LEE- Uydu Haberleşme ve Uzaktan Algılama Lisansüstü Programı

Bu topluluk için Kalıcı Uri

http://hdl.handle.net/11527/19486

Gözat

Cooperative vehicular communication systems with physical layer security and noma techniques

(Graduate School, 2021-01-22) Koşu, Semiha ; Ata Durak, Lütfiye ; 705181014 ; Satellite Communication and Remote Sensing

In recent years, with mobile communication systems development, higher bandwidths and higher data rates are required for individual users. Moreover, in the next-generation wireless communications (5G+), with the emergence of smart cities, many autonomous vehicles and infrastructures are expected to connect. In addition to these numerous connections, it must provide ultra-reliable and low-latency communication (URLLC), which is also necessary for next-generation wireless communications. There are studies examining system performance in vehicle-to-vehicle (V2V) and vehicle-to-everything (V2X) communication systems in the literature. The inter-vehicle environment requires lower antenna heights, cost, and hardware complexities due to the high mobility of vehicles compared to other traditional mobile environments. Also, fading environments in inter-vehicle systems are different from those of stationary users in the literature. Besides, this inter-vehicle fading medium is assumed to be the product of channels in the conventional fading medium and is called the cascade channel model. Therefore, the cascade channel model has an adverse effect on overall system performance. However, some techniques have been studied in the literature which improves V2V system performance. Cooperative communications and receive diversity techniques are considered as a potential solution for inter-vehicle communication systems. When vehicles are not close enough to each other, the signal may be transmitted over relay nodes, increasing the source coverage area enabling cooperative communications. On the other hand, multiple antenna systems are used to combine signals in the receiver to increase the reliability of the system. The diversity technique at the receiver, which reaches the optimum result by maximizing the received signal-to-noise ratio (SNR), is considered to be the maximum ratio combining (MRC) technique that corresponds to the sum of all SNR values received at the destination. Thus, system performance is improved compared to the use of a single antenna, and the coverage area of the source node is increased with the help of a relay. Also, the relay can use different transmission protocols while transmitting the information of the source. In the decode-and-forward (DF) relaying protocol, the transmitted signal is decoded in the relay first, and then an estimated version of symbol is transmitted to the target node. In the amplify-and-forward (AF) relaying protocol, the signal received on the relay is amplified and then transmitted to the destination. Unlike DF relaying protocol, the noise component is also amplified and sent to the destination in this transmission technique as a disadvantage. In this thesis, a comparison of DF and AF relaying protocols are studied, assuming that all nodes are mobile in the system. Also, the channels between all vehicle nodes are designated as cascade Rayleigh fading. It is also assumed that the relay is placed co-linearly and with equal distance between source and destination. Moreover, the relay is equipped with a multi-antenna and applies the MRC technique. Results are provided in terms of bit error probability (BEP) versus SNR values. Accordingly, the increasing number of antennas have improved system performance for both AF and DF relaying protocols. As a result, it is shown that the obtained mathematical expressions are consistent with the Monte-Carlo simulation results. With the tremendous increase in mobile devices in recent years, the continuous broadcast feature of mobile nodes has become a fundamental problem for ensuring security in the system. Therefore, information can become available even to illegitimate listeners. In the open system interconnection (OSI) model, as the physical layer is critical, it is crucial to provide security and transmitting secure information to other layers. Jammer and eavesdropper are the two main types of physical layer attacks studied in the literature. In jamming attacks, the jammer deliberately generates a noise, causing the received signal to be distorted at the destination. However, in eavesdropping attacks, the eavesdropper intercepts the confidential information transmitted to the destination. In all types of attacks, the secrecy capacity of the general system decreases. However, physical layer security (PLS) techniques focusing on increasing system security performance are studied in the literature. For instance, a secret key generation is a PLS technique that increases system security by using the randomness of channels. In this method, the secret key is generated based on the channel state information (CSI) between the legitimate users. Therefore, the data is kept confidential since the illegal user fails to predict the key, even empowering them with high power. In this thesis, the eavesdropper is equipped with multiple antennas for a realistic scenario and applies the MRC technique. Moreover, the eavesdropper receives the broadcasted information from both source and relay in the proposed vehicular communication system. The channel models between all mobile nodes are assumed as the cascade Rayleigh fading channel. The secrecy capacity in this system is calculated by subtracting the eavesdropper's capacity from the destination node's capacity. As an evaluation criterion, the secrecy outage probability (SOP) is calculated first. SOP gives the expression when the secrecy capacity falls below a particular threshold value. Moreover, the probability of positive secrecy capacity (PPSC) means that the instantaneous secrecy capacity is always greater than zero is examined. For system performance, it is observed that when the number of receiver antennas of the eavesdropper increases, SOP increases, and PPSC decreases. Finally, the theoretical analyses of SOP and PPSC are verified by Monte-Carlo simulations. In wireless communication networks, several multiple access methods are drawing attention, such as frequency division multiple access (FDMA), time division multiple access (TDMA), and code division multiple access (CDMA). These orthogonal multiple access (OMA) techniques share the same resource and allow multiple users to work simultaneously in a limited spectrum based on frequency, time, or code. In other words, mobile users can access a limited number of the spectrum simultaneously in these techniques. However, spectrum scarcity is encountered in next-generation wireless networks due to users' need for high data rates and limited resources. At this point, the non-orthogonal multiple access (NOMA) technique could be a promising technology for future wireless networks in terms of providing high spectral efficiency and ensuring fairness between users. The basic concept of NOMA is to allocate different power to users and enable them to work on the same resource block (frequency, time, or code). Besides, NOMA can be classified into two categories, power-domain and code-domain. In power-domain NOMA, the signals of current users are superimposed at the base station (BS) and broadcasted towards the users to decode their signals. The transmitted signal is decoded at the users using the successive interference cancellation (SIC) method, starting from the strong user with better channel quality conditions. Unlike traditional OMA techniques, weak users with poorer channel quality are allocated more transmission power in NOMA, while stronger users with better channel quality are allocated less transmission power. This power allocation can considerably compensate for the trade-off between the quality of service of the system and user fairness. In this thesis, the cooperative power-domain downlink NOMA system is studied. The BS communicates with two vehicles via the relay node and operates in half-duplex (HD) mode. Also, relay transmits the signal of the source to the users by applying the AF relaying protocol. Since both relay and users have high mobility, the channel corresponding to link BS and relay is subjected to Rayleigh fading. In contrast, the channels between relay and users are considered as double Rayleigh fading. Since transmitted signals of each user are superimposed, the SIC method helps to decode these signals. It is assumed that the signal of weak user is correctly decoded on the strong user's channel. In other words, the SIC technique is performed perfectly. Additionally, system performance is evaluated in terms of outage probability and ergodic capacity. In both analyzes, the results are provided using different system parameters (power allocation, distance and transmission power of the BS) for the users. Besides, the overall system performance is also taken into account. Finally, the numerical results are consistent with the Monte-Carlo simulation results.
Aircraft detection using deep learning

(Graduate School, 2022) Mutlu, Utku ; Pınar Kent, Sedef ; 732793 ; Satellite Communication and Remote Sensing Programme

For many years, it has been thought whether computers can think like humans and perform these intellectual tasks autonomously. Artificial intelligence studies were started on this idea and today, studies are carried out using this technology. Both machine learning and deep learning are subsets of artificial intelligence and deep learning is actually subset of machine learning. Deep learning is used in the development of technologies such as image recognition, virtual assistant, natural language processing, speech recognition, autonomous and robotic systems. Deep learning applications have been developed for years by using remote sensing images. Deep learning algorithms in remote sensing images are frequently used for the detection of objects such as aircraft, ships, buildings and other similar things for civil and military purposes. Owing to the development of high-performance hardware, the ease of access to big data and the rapid development of deep learning algorithms, progression of new projects have been satisfied with less time and lower cost. Remote sensing is the science of obtaining information about an area without being in contact with it. Remote sensing devices consist of sensor systems in satellites and aircraft. In 1858, the earliest aerial photo was acquired through a hot air balloon attached to the ground with one or more tether. In the early 1900s, aerial photographs were taken with cameras mounted on pigeons, and in 1909, aerial images were obtained with cameras mounted on airplanes for the first time in order to view larger areas. The term "remote sensing" was used for the first time in the 1950s. Remote sensing satellites provide information about the atmosphere, ocean, and land. As a result of the development of satellite sensors that can detect different parameters, the use of remote sensing images has become widespread in more comprehensive projects. The main areas where remote sensing is used are defense, agriculture, aviation, forestry, biodiversity and surface changes. Deep learning is a subset of the machine learning algorithm in artificial intelligence, emulating the working human brain as it processes data and creates patterns for use in decision making. In 1943, the first mathematical model of a neural network that imitates the thought process of the human brain was created by Walter Pitts and Warren McCulloch. An algorithm using a two-layer neural network for pattern recognition was developed by Frank Rosenblatt, and the first perceptron was presented in 1957. Alexey Ivakhnenko and V.G. Lapa published the first working neural network for supervised learning in 1965. Alexey Ivakhnenko described the 8-layer deep learning network in his publication in 1971. Artificial intelligence studies were interrupted between 1974 and 1980 due to the lack of hardware with sufficient processing power and memory to train multilayer networks. Neocognitron, a multilayer artificial neural network, was developed by Kunihiko Fukushima in 1980. The term "deep learning" was first used by Rina Dechter in 1986. Mike Schuster and Kuldip Paliwal introduced bidirectional recurrent neural networks in 1997, which connects two hidden layers, one for the positive time direction and the other for the negative time direction, to the same output. Fei-Fei Li started working on the ImageNet idea in 2006 because of the need for a large amount of labeled images for supervised learning. In 2009, Fei-Fei Li introduced ImageNet that is a database of a large quantity of labeled images.
Performance of 5G codes over a noisy channel

(Graduate School, 2022) Sanfaz, Mohamed ; Helvacı, Mustafa ; 713268 ; Satellite Communication and Remote Sensing Programme

At present, the need for mobile internet keeps increasing every day, especially with the rise of IoT devices, as it's estimated that by the year 2025, there will be more than 5 billion IoT devices connected to the network. For wireless mobile communication, a huge bandwidth is needed to adapt the different rates for different applications. The 5G network will provide lower latency and also achieve higher speeds than previous networks. In 5G wireless communication, both turbo codes and tail-biting convolutional codes failed to meet 5G standards even though they proved their efficiency for the LTE standard. In 5G, a more advanced error correction method is needed for both LDPC codes and polar codes, specifically LDPC codes dealing with data channels and polar codes dealing with control channels. As error correction and detection are the main requirements for 5G wireless communication, the BER performance against the (Eb/NO) performance is really important as you don't want to lose almost any transmitted block. One of the methods used to check BER against EB/NO was to check an un-coded signal under various types of modulation, from BPSK up to 256 QAM; the higher the modulation, the worse the BER against EB/NO performance was getting. With 5G packing more data now, even higher than 256 QAM is possible. A performance test of the codes that are being used in 5G has been simulated here. As is customary, the higher the modulation, the worse the BER against EB/NO. A 5G-NR scenario has been performed using BPSK modulation with an AWGN channel to demonstrate how the codes perform under the best modulation scenario. The 5G standard has been applied to both codes as base graph 1 and base graph 2 have been used for LDPC at different code rates. The same goes for polar as channels are in sequential order from worst to best as specified in the standard. The hardware performance for 5G is very challenging, so a single decoder has been used in both codes, with quantization implemented in both of them. As a result of simulations of BER at both codes, different plots have been shown. For LDPC codes, performance iterations had a noticeable improvement in BER levels starting at 10 iterations to 20 iterations and from 20 to 30 iterations. Not a huge BER improvement was seen, so 20 iterations have been implemented as the main iteration number for most of the graphs. For LDPC codes, both base graphs were used. For rate half, with midsize block BG1, had a better performance; for rates 2/3 and 5/6, rate 2/3 had an overall better performance compared to rate 5/6, with 4096 block size providing the best results in both rates. As for polar codes, successive cancelation was implemented for 256 and 512 block sizes with different rates. The lower the block size, the better the results were obtained for polar codes.
Satellite images super resolution using generative adversarial networks

(Graduate School, 2022) Serdar, Maryam ; Kayran, Ahmet Hamdi ; 717024 ; Satellite Communication And Remote Sensing Programme

The general broad definition of remote sensing is to observe an object and collect data regarding this object without actual contact. From a narrower perspective, it is the science that studies the earth and its atmosphere by gathering data from above the earth. Nowadays earth observation systems with their various sensors in multiple bands produce a huge amount of data that need to be processed and analyzed to get a final product in a certain discipline. Applications like monitoring the water resources, forest fire monitoring, soil type classifications are examples of remote sensing use in different fields of our modern life. Satellite imagery plays a pivotal role in remote sensing .they can be acquired by various types of sensors some of which are passive like optical sensors and some whıch are active like LIDAR and SAR. This study focuses on the satellite images in the visible portion of the spectrum. This type of satellite imagery can vary in resolution whether this resolution is spatial, spectral, temporal, or radiometric. The satellite imagery also can be categorized according to its spatial resolution into low, medium, and high-resolution images and each of them can be deployed in certain applications. Preprocessing these images is a critical stage that would affect the final product or the application that uses these images. High resolution is a desirable characteristic, yet it can be difficult to achieve financially and technically. However, image processing can offer a convenient software solution to this problem by super-resolution techniques. Hence, the importance of superresolution which is one of the preprocessing tasks that obtains high-resolution images is considered fundamental in lots of remote sensing applications. Super-resolution aims to obtain high-resolution images using low-resolution observation. Super-resolution is considered a classical image processing problem that is ill-posed due to the lack of a single unique solution. Thus, lots of algorithms and approaches were proposed over the years. This study gives a general review of the main significant types of super-resolution algorithms which can be divided into interpolation-based, reconstruction-based, and learning-based algorithms. The simplest methods are interpolation-based ones, nevertheless, the results lack high-frequency details. The second type is reconstruction-based methods which require a good prior choice to get better results. designing a good prior can be complex These methods can be complicated. The third category is example-based or learn-based methods which include learning the relationship between the low resolution and high-resolution images by exploiting datasets to learn from. Algorithms like sparse coding super-resolution and deep learning methods are learning-based methods. Super-resolution methods performance is usually evaluated by many metrics such as, peak signal to noise ratio PSNR, which is based on mean squared error, a pixel-wise metric thus, can be misleading, structural similarity index SSIM which is considered more accurate as it considers the structure of the image instead of the individual pixel value. Deep learning, which deploys deep neural networks in its algorithms, is a branch of machine learning which is, in turn, a subfield of artificial intelligence. It is widely used in image processing and computer vision problems, especially after the emergence of convolutional neural networks CNNs. Deep learning models structures in image processing problems usually share common building blocks like CNNs. The default CNN consists of a convolutional layer followed by an activation layer to ensure nonlinearity, hence learning, which is followed by a pooling layer. The backpropagation is used to adjust weights at the end of every epoch of training. The fourth chapter of this thesis elaborates the super-resolution algorithms which were proposed to deal with super-resolution problems that present the state-of-the-art performance compared to the other methods. SRCNN was the first suggested model to deal with super-resolution. It is considered as the benchmark of super-resolution using deep learning. This model was followed by the FSRCNN which tried to overcome the backward of the previous model by using the low-resolution image as an input without upscaling and performed the upscaling later by using deconvolution layer. Very Deep Super Resolution model which mainly consists of deep VGG layers to get better results. Then there was the enhanced deep super-resolution model EDSR that exploited the concept of the residual blocks to be able to increase the depth of the network without getting slower training. SRResNet and SRGAN were proposed in the same paper to give a better performance in image super-resolution. SRResNet deployed the residual blocks in its structure in addition to conv layers and uses the mean squared error dased loss or VGG content loss to optimize. The generative model of generative adversarial neural networks consists of two network models that learn together, the generator aims to learn to generate the required data with the help of a discriminator that tries to differentiate between fake data generated by the generator and ground truth. This approach of training in an adversarial manner presents a state of the art performance in several tasks, It was also used in the super-resolution task by what is called as SRGANs super-resolution networks. In addition to the adversarial structure of this model, another factor that improved its performance is the perceptual loss that was used in optimizing the model. Mentioning all of these deep learning super-resolution algorithms, the next chapter gives a general overview of the use of deep learning in remote sensing. This use is expanding with the increased amount of remote sensing data and its quality and with the development of deep learning algorithms and computational abilitıes. From the preprocessing of the remote sensing data, like image fusion, segmentation, and denoising, to other many applications such as anomaly detection, land use classification, and other classification tasks, deep learning is being deployed in remote sensing. The experiment that is done in this thesis is to examine the performance of super-resolution generative adversarial neural networks on the satellite images and ıts abıltıy of generalization when it is trained with the irrelevant dataset. By training an SRGAN model using the UC-MERCED Land Use dataset which consists of 21 classes each class contains 100 images of size 256x256 these images are used as high-resolution images and downsized versions of them with factor x4 are used as low-resolution images. After training, the model was tested with random images from the NWPU-RESISC45 dataset. In order to examine the ability of generalization of the model, the same architecture was trained using a natural images dataset which is Linnaeus 5 256X256 which consists of 5 classes of 256x256 sized images in the same way as the previous training. testing was done with random images from the NWPU-RESISC45 dataset. In addition, the SRResNet model that uses the mean square error-based optimization was trained to compare it with the performance of the previous generative SRGAN models. Peak signal to noise ratio and structural similarity index was used to evaluate the performance and make a comparison between the previously mentioned methods. The experiment was done using Google Colab Pro environment utilizing its provided GPU.
Bölgesel ölçekte klorofil-a konsantrasyonunun belirlenmesinde sentınel-3 olcı verilerinin kullanım olanaklarının araştırılması

(Lisansüstü Eğitim Enstitüsü, 2022-01-17) Demir, Başak ; Kaya, Şinasi ; 705181022 ; Uydu Haberleşmesi Ve Uzaktan Algılama ; Satellite Communication and Remote Sensing

Türkiye güneyde Akdeniz, kuzeyde Karadeniz, batıda Ege Denizi ile çevrilidir. Ülke sınırları içinde bulunan Marmara Denizi, Karadeniz'i Ege Denizi ve Akdeniz'e bağlar. Birbiriyle bağlantılı olan bu denizlerin izlenmesi doğal çevrenin sağlığı ve canlılar için önemlidir. Uzaktan algılama çalışmalarıyla yapılan araştırmalar sonucu denizlerdeki problemler tespit edilmektedir. Marmara Denizi birçok ekosistemi etkilemesiyle birlikte insan sağlığı içinde önemli bir iç denizdir. Denizin kirliliği doğrudan ya da dolaylı yollarla bütün canlıların hayatına olumsuz yansıyacaktır. Evsel, ticari, endüstriyel, doğal ve tarımsal gibi birçok nedenle deniz kirliliği artmaktadır. Bu durum Marmara Denizi'nde organik ve inorganik kirleticilerin çoğalmasıyla deniz suyuna ve denizde bulunan ekosisteme ciddi zararlar vermektedir. Bununla birlikte Avrupa'nın en büyük ikinci nehri olan Tuna Nehri'nin sanayinin yoğun olduğu ülkelerden, yerleşim yerlerinden ve tarım alanlarından geçerek Karadeniz'e ulaşması ve Karadeniz'de oluşturduğu kirliliğin Marmara Denizi'ne yansımasıda deniz kirliliğinde büyük bir etkendir. Sentinel-3 OLCI deniz ve yeryüzü hakkında bilgi kaydeden bir uydudur. Çevre ve iklimsel gözlem çalışmalarında da kullanılmaktadır. Sentinel-3A ve Sentinel-3B olmak üzere iki özdeş uyduya sahiptir. Tez kapsamında Sentinel-3 OLCI uydu görüntü verileriyle Karadeniz'in batısında ve Marmara Denizi'ndeki klorofil-a konsantrasyonunun neden olduğu kirlilik incelenmiştir. Veriler Yerüstü Su Kalitesi Yönetmeliğinde belirlenen su kalite sınıflarına göre 9 sınıfa ayrılmış ve makine öğrenme algoritması olan destek vektör makinesi (SVM) kullanılarak kontrollü sınıflandırma yapılmıştır. Destek vektör makinesi iki sınıflı doğrusal verilerin veya çok sınıflı doğrusal olmayan verilerin sınıflandırılması için tasarlanmış bir algoritmadır. Farklı mevsimlerde alınan 2020 yılına ait sınıflandırılmış uydu görüntülerine göre klorofil-a konsantrasyonunun ilkbaharda en yüksek sonbaharda en düşük olduğu belirlenmiştir. Bununla ilişkili olarak alg konsantrasyonunun da ilkbahar aylarında artış göstermesi bu durumu desteklemektedir. Ek olarak Tuna Nehri'nin Karadeniz'de yarattığı kirliliğin Marmara Denizi'ne yansımasıyla deniz kirliliğinin arttığı değerlendirilmiştir. Suda yaşayan algler fotosentez yapan canlılardır. Deniz yüzeyinin kirli olması güneş ışınlarını engellediği için bu canlıların fotosentez yapmasına izin vermemektedir. Deniz yüzeyinin kirliliği sonucu ısı dengesi de bozulmaktadır. Bunların sonucunda algler çoğalarak salgı üretir ve müsilaj (deniz salyası) problemine neden olmaktadır. 2021 yılının bahar mevsimine ait Sentinel-3 OLCI (Ocean and Land Colour Instrument) uydu görüntü verileri incelendiğinde 2020 yılının bahar mevsimine kıyasla Marmara Denizi'ndeki klorofil-a oranında artış görülmüştür. Buna bağlı olarak oluşan müsilaj Marmara Denizi'nde büyük bir soruna neden olmuştur. Müsilaj probleminin kontrol altına alınabilmesi için Marmara Denizi ve Batı Karadeniz'in düzenli olarak izlenmesi gerekmektedir. Sentinel-3 OLCI verileriyle yapılan çalışmalar, yıl içinde mevsimlere bağlı klorofil-a konsantrasyonundaki değişimin takip edilmesi için iyi bir seçenek olabilir. Bunun dışında OLCI için geçmiş yıllara ait verileri bulmak her zaman mümkün olmamaktadır. Bu durumda yapılan çalışmaların farklı uydu görüntüleriyle desteklenmesi çalışmaların sürdürülebilirliğinin sağlanması açısından bir gereklilik olacaktır.
Design of a reverberation chamber from a shielded room

(Graduate School, 2022-01-27) Aba, Rıdvan ; Yapar, Ali ; 705181038 ; Satellite Communication and Remote Sensing

Electromagnetic Compatibility (EMC) has become more important in the last half-century. The reason for this situation is sensors and similar sensitive electronic circuits, which are increasingly used in electronic systems. As a result of increasing electromagnetic pollution, electromagnetic interference (EMI) to systems has started to have very serious consequences. For this reason, some standards and restrictions have been introduced for almost all electronic products to be produced. These standards and restrictions describe the testing of products for electromagnetic compatibility. One of the test environments used in these tests is the reverberation chambers (RC). RCs are systems based on obtaining a uniform field level in a certain volume by stirring the electromagnetic waves emitted from a source in a closed room with the help of a stirrer. In this thesis, studies on the conversion of a screened room to an RC are included. Within the scope of this study, it was investigated whether the screened room is suitable for conversion to an RC and some preliminary measurements and simulations were made. Afterwards, the stirrer design, which is the most important part of an RC, was started. When the studies in the literature were examined, it was thought that the use of a Z-folded stirrer was appropriate. RC simulations were started with RC simulation with a vertically positioned stirrer, designed using ALTAIR FEKO electromagnetic analysis program. Subsequently, RC simulations with horizontally positioned 4-panel and 5-panel stirrers were performed. Then, an RC simulation with two stirrers was performed. In all these simulations, the position and angle of the antenna were kept constant and changes were made on the stirrer. Since the desired uniform field could not be obtained in this way, it was decided to follow a different plan. In the new simulations, the stirrer position was kept the same and changes were made on the location and angles of the antenna. As a result of the simulations, it was decided to produce the configuration that provides the desired uniform field. After the production and assembly of the stirrer were completed, the verification measurements of the RC were made. The matching of the measurements with the simulations showed the successful completion of the work. Although a difference was observed between the lowest usable frequencies (LUF), this was thought to be due to the simplified modelling of the RC. As a result, an RC that can be used in EMC tests was made for the first in Turkey. In the future, studies will continue to develop the established system.
Direct and inverse scattering problem related to real breast models

(Graduate School, 2022-06-10) Çarıkçı, Ozan ; Yapar, Ali ; 705181013 ; Satellite Communication and Remote Sensing

Meme kanseri dünyada en çok bilinen kanser türlerinden biridir. Tüm dünyada meme kanseri oranları her geçen gün artmakta ve bu oranların ihmal edilebilir düzeyde olmadığı gözlemlenmektedir. Sonuç olarak bu konu tedavi yöntemleri açısından araştırmacılar için oldukça önemli hale gelmiştir. Meme içindeki tümörün görüntülenmesi ve zamanında tıbbi müdahalenin yapılması meme kanseri tedavi yöntemleri açısından oldukça önemli bir konudur. Meme içindeki tümörün görüntülenebilmesi için öncelikle tümörün memedeki elektromanyetik dalga yayılımı üzerindeki etkilerinin ortaya çıkarılması gerekir. Bu etkiler, memelerin farklı konfigürasyonlarda analiz edilmesiyle ortaya çıkarılabilir. Bu tezde literatür taraması yapıldıktan sonra farklı konfigürasyonlar için tümörlü ve tümörsüz meme açısından analizler yapılmıştır. Meme kategorisi, memenin kesiti, frekansı, arka plandaki εr değeri, kaynak sayısı, tümör boyutu gibi farklı parametreler değiştirilerek ve tümörün elektromanyetik dalga yayılımı üzerindeki etkileri, elektrik alan, enerji ve gösterge fonksiyon dağılımı analiz edilmiştir. Bu analizler Momentler Metodu (MoM) ve Ters Zamanlı Geçiş Metodu (RTM) yardımıyla düz ve ters saçılma problemleri çözülerek oluşturulmuştur. Sonuç olarak, tümörlü ve tümörsüz meme için bu dağılımları analiz ederek ve yorumlayarak gerçekleştirilen bu çalışma, mikrodalga görüntüleme problemlerine bir arka plan oluşturabilmek için yapılmıştır. Gerçek göğüs modelleri ile ilgili düz saçılma problemleri bölümünde, yapılan analizler sonucunda, elektrik alan ve enerji dağılımı açısından en iyi sonuç ikinci göğüs kategorisi olan dağınık fibroglandüler doku göğüs kategorisinde bulunmuştur. Bu göğüs kategorisinde, tümörün göğüs içinde olduğu yer neresi olursa olsun, 1 GHz frekans bandı, R=0,5 cm tümör çapı, göğüsün z eksenindeki dilim 24. dilim, arka plan değeri εr=1 olan ve kaynak sayısı 8 e eşit olan, elektrik alanı ve enerji dağılımı açısından tümörü en iyi ¸sekilde görselleştirebilmek için diğer tüm parametreler arasında en iyi seçimlerdi. Gerçek göğüs modelleri ile ilgili ters saçılma problemleri bölümünde, yapılan analizler sonucunda, gösterge fonksiyon dağılımı açısından en iyi sonuç, neredeyse tüm yağlı göğüs kategorisi olarak adlandırılan birinci göğüs kategorisinde bulunmuştur. Bu göğüs kategorisinde, tümör göğüsün neresinde olursa olsun, z ekseninin hangi dilimi seçilirse seçilsin, kaynak sayısı 10'dan az olmadığı sürece, 1 GHz frekans bandı, R=0,5 cm tümör çapı, 10'a eşit arka plan εr değeri gösterge fonksiyon dağılımı açısından tümorü en iyi ¸sekilde görselleştirebilmek için tüm diğer parametreler arasında en iyi seçimlerdi. Sonuç olarak gerçek göğüs modelleri düz ve ters saçılma problemi açısından analiz edilmiştir. Gelecekte, gerçek göğüs modellerinde tümörlerin elektromanyetik dalga yayılımı üzerindeki etkilerini anlamaya yönelik çalışmalar devam edecektir.
Vessel detection from very high-resolution satellite images with deep learning methods

(Graduate School, 2022-06-22) Büyükkanber, Furkan ; Yanalak, Mustafa ; 705181008 ; Satellite Communication and Remote Sensing

Vessel detection from remote sensing images is becoming increasingly important component in marine surveillance applications such as maritime traffic control, anti-illegal fishing applications, oil discharge control, marine pollution and marine safety. Increasingly, very high and medium resolution (VHR and MR) earth observation satellites both significantly increase the detectability of many terrestrial objects and shorten recurring visit times in orbit like never before, making the use of this technology attractive for a variety of maritime monitoring missions. However, the difficulty and complexity of object detection in huge satellite images that cover hundreds of square kilometers and derive results under near real-time constraints cause traditional methods to face many difficulties when processing satellite images of this size. Processing these images and applying them to deep learning methods makes it possible to minimize unforeseen errors that can be made by analysts, and to save labor, time and cost. In order to create the artificial neural network and make it successful by determining the deep learning method, it is necessary to train using as much as possible examples of the objects targeted to be detected. By using the designed convolutional neural networks, it is possible to detect more than one object in a given test input image and perform change analysis as well. The weights are updated in each layer for the input image processed in the multilayer convolutional neural network, and the error rate is found by looking at the difference between the detected value and the actual ground truth value. Many vessels for commercial, military and civil purposes are observed in international maritime areas, usually in areas close to ports and coasts. High resolution satellite images, which provide wide field of view and altitude monitoring, are very useful for vessel detection. Vessel detection from satellite images plays a significant role for inspecting maritime areas, controlling maritime transport traffic and applications for defense purposes. Open source datasets are widely used in object detection applications, since it takes a substantial amount of time and cost to build a dataset for object recognition and detection from satellite images. Within the scope of this thesis, models developed using convolutional neural networks including single-stage and two-stage deep learning methods were used by applying our own dataset images that we build with the open source DOTA dataset selected for vessel detection. For the purposes of the experiments in this research, three separate datasets were built. All the images were labelled with YOLO annotation format, then in accordance of use for various models, they have been converted to COCO and Pascal VOC annotation format. Both inshore and offshore vessel images have been collected with having wide variety of scales, shapes, orientations and weather conditions (fuzzy, cloudy, sunny, etc.). Experiments were performed by using Faster R-CNN, YOLOv3, YOLOv5 and YOLOX deep learning models on all three different datasets. Any dataset containing various examples of the target object considerably improves the accuracy of outcomes in deep learning applications by implementing various data augmentation techniques, such as mosaic, mixup, and rotating images, are utilized for remote sensing. In some experiments, more than one augmentation approach is being used simultaneously to improve the accuracy of the results. Not all data augmentation approaches had the same effect on the experiment outcomes. As a result, there is no logical answer to the question of which data augmentation strategy is the most effective. The outcomes of the studies were compared using the mean average precision metric (mAP), and the YOLOv5 model achieved on top results. All of the experiments have yielded the same result: raising the depth of the network by increasing the size of the input images. mAP value results improved as the input sizes were increased, however this caused the selected models longer to train. Experiments in deep learning studies are made easier by machines that have powerful graphics cards. Faster R-CNN, YOLOv3, YOLOv5 and YOLOX model trainings were conducted on a local machine workstation equipped with NVIDIA GeForce RTX 2080Ti graphics card and Intel® Core™ i9-9900K 3.60 GHz CPU processor. Deep learning applications were carried out using Python programming language and PyTorch framework deep learning library.
Single-frame and multi-frame super-resolution on remote sensing images via deep learning approaches

(Graduate School, 2022-07-29) Wang, Peijuan ; Sertel, Elif ; 705172003 ; Satellite Communication and Remote Sensing

As a quite significant computer vision task, image super-resolution (SR) has been widely applied in remote sensing (RS), medical imaging, video surveillance, and biometrics. Image SR aims to restore high-resolution (HR) images by enhancing the spatial, spectral, or temporal resolution of the low-resolution (LR) inputs. In recent years, great efforts have been made for improving the SR approaches. One of the approaches is to design deeper networks. Nevertheless, this greatly increases computation and memory consumption. As a result, some mechanisms (such as cascading networks, attention mechanisms, and back projection) are proposed to improve the performance and the training process of the complex networks. Satellite imagery can be seen in various fields, namely Land cover/Land use classification, road and building extraction, observation of climate, and earthquake prediction. However, in some cases, the resolution of satellite images can not meet the application requirements due to the technology and cost limitations during the satellite design; therefore, the improvement of image resolution might be necessary. Since it is not possible to upgrade the equipment onboard the launched satellite, the software-based SR algorithms are deserved to be explored in RS fields. This thesis aims to strengthen the spatial resolution of optical satellite imageries by using deep learning (DL) methods. Generally, image SR algorithms can be categorized as single-frame image SR (SFSR) and multi-frame image SR (MFSR). The inputs of the SFSR can be a single LR image. While multi-frame image SR aims to restore HR image by using multiple LR images, which can be obtained under different conditions and at different angles. Recently, great contributions have been made to improve the SR methods including two aspects: (1) increasing the value of PSNR (Peak-Signal-Noise-Ratio); (2) improving the image quality perceptually. Nevertheless, some algorithms obtain a high PSNR but with a low perceptual quality which is more important to human perception. Therefore, this thesis has the following objectives: (1) Explore a perceptual-driven approach to enhance the SR image quality visually on single-frame and multi-frame RS imageries; (2) Explore Generative Adversarial Network (GAN)-based models for single-frame and multi-frame RS imagery super-resolution task to fulfill the multi-scale problem and blind to the degradation model; (3) Explore an image fusion method that can generate an arbitrary size of the super-resolved image rather than a small patch. This thesis firstly gives an overview of single-frame and multi-frame RS image SR methods. The single-frame RS image SR methods are briefly classified into supervised and unsupervised methods. The former mainly includes Convolutional Neural Networks (CNN)-based, GAN-based, attention-based, and Back-projection based methods. In addition, the commonly used attention mechanisms including self-attention, channel attention, spatial attention, mixed high-order attention (MHOA), non-local attention (NLA), and non-local sparse attention (NLSA) are also introduced. Moreover, loss functions including pixel-wise loss, perceptual loss, adversarial loss, and cycle consistency loss are presented. For the single-frame RSISR, firstly, an attention CNN-based SR method is proposed. Although CNN-based algorithms have made outstanding achievements in computer vision tasks, the traditional CNNs methods treat the abundant low-frequency information included in the LR inputs equally across channels. Attention-guided algorithms play a vital role in the informative features extraction in various tasks including image SR. With the application of the attention mechanism, the proposed CNN-based method can further learn the deeper relationships among the different channels. Instead of simply integrating the attention module with the residual blocks, a Layer Attention Module (LAM) and Spatial Attention Module (SAM) are proposed to further learn the relationships among the Residual Groups (RG). Moreover, the perceptual loss function is adopted in the training process to enhance the generated image quality perceptually, and Random down-sampling is applied to strengthen the model's generalization ability. Secondly, an attention GAN-based super-resolution method is explored for the single-frame RS images. CNN-based methods have made great contributions to increasing the value of PSNR/SSIM. Nevertheless, the generated outputs tend to be overly smooth and blurry. GAN can generate more realistic images than normal CNN-based methods and has been introduced to single image super-resolution (SRGAN, ESRGAN, EEGAN). Standard GANs only function on spatially local points in LR feature maps. The attention mechanism can directly learn the long-range dependencies in the feature maps both in the generator and discriminator in a powerful way. By applying the attention mechanism, the network allocates attention based on the similarity of color and texture. Therefore, based on ESRGAN, an attention GAN-based method is for the single-frame RS image SR. The ESRGAN was mainly improved from two aspects: (1) we further improved the architecture of the residual blocks by adding more skip connections; (2) we add attention modules to the residual blocks for further feature extraction. Moreover, instead of working on aerial photographs or low-resolution and medium-resolution satellite images, we are focusing on the Very High-Resolution (VHR) satellite imageries, such as the Pleiades, and Worldview-3. The spatial resolutions of the multispectral images for the Pleiades, and Worldview-3 are 2m, and 1.24m, respectively. Furthermore, for the attention CNN-based method, we evaluated the method on the Pleiades and Worldview-3 datasets with scaling factors of 2, 4, and 8. For the attention GAN-based method, we evaluated the method on the Pleiades and Worldview-3 datasets with a scaling factor of 4. The experimental results show the attention-based method can provide better perceptual results both quantitatively and qualitatively. At last, we proposed an attention GAN-based method for the multi-frame RS image SR. Firstly, we introduced an attention mechanism to the Generator and proposed a space-based network that worked on every single frame for better temporal information extraction. Secondly, we proposed a novel attention module for better spatial and spectral information extraction. Thirdly, we applied an attention-based discriminator to enhance the discriminator's discriminative ability. Finally, the experimental results on the SpaceNet7 dataset and Jilin-1 dataset exhibit the superior of the proposed model both quantitatively and qualitatively.
Makine öğrenme yöntemleriyle uydu görüntülerinin sınıflandırılması ve zamansal değişiminin izlenmesi

(Lisansüstü Eğitim Enstitüsü, 2022-09-22) Solmaz, Babakan ; Algancı, Uğur ; 705181015 ; Uydu Haberleşmesi ve Uzaktan Algılama

Gümüzde, hızlı nüfus artışı ve kentleşmenin ivme kazandığı dünya kapsamında görünmektedir. Bu sürecin bir doğal sonucu olarak ise sürecin gereksinimlerini karşılamak için Arazi Örtüsü ve Arazi Kullanımı (AÖ/AK) sınflarında değişiklikler meydana gelmektedir. Öte yandan, karşılaştığımız küresel ısınma ve iklim değişiklikleri nedeni ile AÖ/AK sınıflarının değişimi daha sıklıkla rastalanabilmektedir. Dolayısıyla, bu değişikliklerin kontrol altında tutulmasında ve özellikle yeşil alanların korunmasını sağlamak, ileri yıllarda yaşanabilecek doğal afetleri öngörebilmek ve tedbir amaçlı uygulamaların ele alınması bakımından önemli olacaktır. Bu çalışmada, Türkiye'de Bursa ili bölgesi ele alınmış ve AÖ/AK sınıflarındaki bölgesel değişiklikler 2001 yılı itibari ve yaklaşık 10'er yıllık aralıklar ile değerlendirilmiştir. Bu değerlendirmelerde, Makine Öğrenme yöntemleri yardımıyla bölgeye ait Lansat uydu görüntülerinin AÖ/AK sınıflandırılması ve zamansal değişiminin incelenmesi yapılmıştır. Doğal ve tarihi güzelliklerinin yanı sıra termal turizm açısından da ülkemizden önde gelen bölgelerinden biri olan Bursa ilinde kentleşme süreci hızla yaşanmaktadır. Dolayısıyla bölgede, yıllar itibariyle uygulanan politikaların AÖ/AK sınıflarının değişiminde olan etkisini incelemek ve yapılan değerlendirmelere dayalı olarak, bölgede yeşil alanların korunmasına odaklı bir politika yürütmek ve sürdürebilir bir kentsel gelişim sağlamak oldukça büyük önem taşımaktadır. Bu çalışmada, uydu verilerinin analiz edilmesinde ve bölgesel kullanım arazi değişimlerinin tespit edilmesi için ücretsiz olarak uydu görüntülerine ulaşmayı ve çevrimiçi incelenmesine imkan sağlayan Google Eath Engine platformuu kullanılmıştır. Çalışmada Makine Öğrenme yöntemlerinden Destek Vektör Makineleri (DVM) ve Rastgele Orman (RO) algoritmaları kullaılmıştır. Bu doğrultuda iki uygulama gerçekleşmiştir. İlk uygulamada, yıllara ait Landsat görüntülerinin görünür ve yakın kızıl ötesi bantları üzerinde Makine Öğrenme sınıflandırıcıları uygulanmıtşr. İkinci uygulamada ise, sınıflandırmalarda daha güçlü performans elde edebilmek ve farklı bileşenlerin etkisini değerlendirmek hedefiyle, indikatör faktör haritaları sınıflandırma için kullanılmıştır. Bu amaçla görüntü iyileştirme yöntemlerinden Bant Oranlaması ve Temel Bileşenler Analizi (TBA) farklı AK/AÖ sınıflarının ayrışmasını kolaylaştırmak için kullanılmıştır. Çalışmada, Bant Oranlaması yöntemleri bölgede litolojik, bitki örtüsü ve kentsel alan bileşenlerin haritalanması amacıyla uygulanmıştır. Böylelikle, en büyük pay sahibi olan ilk temel bileşen görüntüsü, Normalleştirilmiş Fark Bitki İndeksi, Kentsel Alan İndeksi kullanılmıştır. Aynı zamanda, çorak alanların ve kayaçların ayrışmasını güçlendirmek amacıyla, farklı bant oranlaması yöntemleri kullanılmıştır. Bu doğrultuda, 5/7 Landsat görüntü bant oranlaması kil minerallerini görüntülemek, 5/4 Landsat görüntü bant oranlaması demirli mineralleri (Fe2+) haritalamak ve 3/1 Landsat görüntü bant oranlaması demir oksitlerin haritalanması için hesalanmıştır. Çalışmada, CORINE sınıflandırma sisteminden ilham alınarak ve bölgesel değerlendirmeler dikkate alınarak, altı AÖ/AK sınıfı incelenmeye alınmıştır. Bu sınıflar bölgedeki Su kütlesi, Orman alanı, Tarım alanı, Çorak alan, Kentsel alan ve Maden ocaklarından oluşturulmuştur. Çalışma bölgesi olan Bursa ilinin zamasal süreçte AÖ/AK sınıflarında meydaa gelen değişiklikleri incelenmek için, yaklaşık on yıllık periyotlarda alınan Landsat 5 TM ve Landsat 8 OLI/TIRS uydu görüntüleri kullanılmıştır. Arazi sınıflandırmasının zamansal değişimiyle ilgili kullanılan uygulama ve yöntemler ele alındığında, sonuçların bir biri ile örtüştüğü görünmektedir. Aynı zamanda, doğruluk oranları değerlendirildiğinde, ikinci uygulama olan indikatör faktör haritalarına dayalı sınıflandırma yönteminin daha iyi performans sergilediği ortaya konulmuştur. Sınıflandırma sonuçlarında ortak sonuç olarak ise, Bursa ili bölgesinde 2001 – 2022 yılları arasında çorak alanlarda azlma tespit edilirken, maden ocaklarında, tarımsal ve kentsel alanlarda genişleme olduğu dikkat çekmektedir. Doğru ve gerçek zamanlı AÖ/AK haritaları, Dünya'nın dinamiklerinin izlenmesi, planlanması ve yönetimi için kesin bilgiler sağlayabilecek niteliktedir. Bulut bilişim platformları, zaman serisi öznitelik çıkarma teknikleri ve makine öğrenme sınıflandırıcılarının ortaya çıkmasıyla, daha doğru ve büyük ölçekli AÖ/AK haritaları üretilebilmek doğrultusunda önemli gelişmelere yol açmıştır.
Design and simulation of fractal-based ring antennas for 5G wireless communications

(Graduate School, 2022-12-02) Altaleb, Abdulazeez Ethar ; Eker, Sebahattin ; 705181001 ; Satellite Communications and Remote Sensing

After the rolling-out of 5G communication systems the development of smaller and more effective components is still ongoing since it is always important to keep up with the development of the technology, therefore smaller compact and easy-to-fabricate components are the main aim of the scientific community these days. Since the 5G systems are somehow smaller than the old systems' components it arises the fact that the newly- designed components have to have space limitations during the design stages. In this work, by focusing on two of the main 5G bands which are the bands centered on 3.5 GHz and 7GHz three types of antennas were designed and implemented by using CST Microwave studio simulator. The antennas were designed using the fractal concept, characterized by space-filling and self-similarity, so there is no need for extra space when we already have a limited one. The design of the first antenna started by designing a cut-angles rectangular patch antenna that propagates at 3.5 GHz, then by copying and then scaling down the same patch and later subtracting it from the main patch we got a single ring cut-angles rectangular patch antenna that propagates at 3.5 GHz with a reflection coefficient of -19 dB and a gain of 2dBi. The second antenna was created by scaling down the full ring of the first antenna and creating a similar inner ring that propagates at 7 GHz center frequency and has a bandwidth between 6.25-8.1 GHz, this antenna can propagate at two different 5G frequency bands centered at 3.5 GHz and 7 GHz respectively. This antenna has a reflection coefficient S11 of around -20 dB for both bands' resonant frequencies and a gain of 2.29 dBi and 2.51 dBi for the two bands at their center frequency. All these antennas have a microstrip feeding line with a length of 16 mm which is equal to something around λ/4 of the first band's center frequency, all the antennas have an FR-4 substrate thickness of 1 mm and a width of the feeding line of 1.6 mm so that together they provide a 50-ohm impedance at the input port which assure that most of the input port's waves are being propagated. Finally, to increase the gain a 4x1 antenna array was designed to propagate at the same bands, this array has two feeding ports that are designed in an inverted way to improve the matching between the array elements, each port is connected to only two propagating elements by a tree-shaped λ/4 length microstrip has a reflection coefficient of around -45 dBi and -35 dB for both bands at their center frequencies, respectively. This array antenna also has a gain for the 3.5Ghz centered band of 5.64 dBi for port 1 and 5.648 dBi for port 2, and for the 7 GHz band, the gain was equal to 8.39 dBi and 8.4 dBi for port 1 and port 2, respectively.
District-based urban sprawl monitoring and modelling using CA-Markov model: application in two mega cities

(Graduate School, 2022-12-12) Azabdaftari, Anali ; Sunar, Filiz ; 705152008 ; Satellite Communication and Remote Sensing

There is a well-known fact that the population all around the world is growing so fast and current estimation and trends forecast a continuous rise in world population in the future. Urbanization occurs due to the high rate of immigration from rural areas to urban areas resulting in urban growth and forcing urban development toward the border of the city. The unstructured growth of urban areas can lead to urban sprawl. There are various methods and data to observe and quantify the changes induced through urban expansion. Remote sensing and Geographical Information Systems (GIS) can be named as one of the most helpful tools in the detection of the spatio-temporal dynamics of patterns of urban growth and Land Use/Land Cover (LULC) changes at local, regional, and global scales. The main aims and objectives of this study are to evaluate the spatio-temporal pattern of urban growth over time, to predict future urban expansion, and draw attention to district- based urban expansion. Thus, in this research, the most urbanized districts of Istanbul and Sydney, were selected to study. Identification of potential areas for future urban development is important in the selection of study areas. Arnavutköy district of Istanbul has a high potential of urban development due to the construction of the Istanbul airport. On the other hand, Hills Shire Local Government Area (LGA) was selected due to the improvement in the transportation system which can affect future urban development. This study consists of two main parts: urban expansion analysis and urban expansion modelling. As a first step in urban expansion analysis multitemporal Landsat satellite images of Arnavutköy district (18-July-1997, 31-August-2007, and 25-July-2017) and Hills Shire LGA (08-December-1996, 05-January-2007, and 19-January-2018) were acquired. After pre-processing of obtained satellite images of both study areas, images were classified into five LULC classes using Maximum Likelihood supervised classification algorithm. Accuracy assessment was also done to evaluate the accuracy of LULC maps of each study area. The overall accuracy values of Arnavutköy district in 1997, 2007, and 2017 were found as 91%, 86%, and 88%, respectively. Likewise, in Hills Shire LGA, the overall accuracy values were acquired as 92%, 95%, and 91% in 1996, 2007, and 2018, respectively. The post-classification change detection technique was employed to identify conversions from one LULC class to another LULC class. The results of change detection analysis showed that the built-up area in Arnavutköy district grew up to 197% while Hills Shire LGA expanded by 78% through the study period. In both study areas, in opposition to the built-up area, forest, and agricultural area experienced a downward trend and lost a lot of hectares during the period of study. To analyze the impact of urban expansion on the Land Surface Temperature (LST) of the surrounding environment, the correlation analysis between satellite retrieved LST and the built-up area in the selected buffer zones were carried out. The results revealed that, with the increase in built-up areas, the LST has also increased over the study period in Arnavutköy district and Hills Shire LGA. This indicates that urban expansion can cause to increase in the temperature of the urban areas which may also have negative effects on the climate changes. The degree of urban sprawl and the spatial pattern of urban expansion were measured using Shannon's entropy approach for each study area. The results indicated that, in both cases, urban areas had undergone a considerable urban sprawl. Furthermore, it was observed that built-up areas mostly expanded toward the North direction in both study areas, which shows the role of human-induced activities in urban expansion. As a next step, selected landscape metrics at the class level were used to better understand the urban structure. Accordingly, six metrics namely, Class Area (CA), Number of Patches (NP), Edge Density (ED), Largest Patch Index (LPI), Patch Density (PD), and Area Weighted Mean Patch Fractal Dimension (AWMPFD/FRAC-AM) were selected to describe the pattern of urban growth. The CA metric, which describes what portion of the landscape is composed of a specific patch type, was used to calculate the absolute size of the built-up area. The NP metric is a simple measure of the discontinuous built-up patch in the landscape and used as a indication on the fragmentation degree of the urban area. The PD metric is a fragmentation metric that reveals the number of patches per unit area. The absolute size of urban patches was determined using the LPI metric, which measures the percentage of entire landscape area covered by the largest patch. The ED and FRAC_AM metrics which measure the patch shape complexity were used to analyze the the shape irregularity of urban areas and spatial heterogeneity of the landscape. Metrics results which provide useful information about the aggregation and dispersion of urban also confirmed an expansion in Arnavutköy district and Hills Shire LGA. The results also showed that Arnavutköy district is more fragmented and Hills Shire LGA has a more compact urban expansion process. These results also confirm the findings of the change detecion and Shannon's entropy analysis. In the final step of this research, the second part of this study which was urban expansion modelling was carried out. Several modelling methods such as; Cellular Automata (CA), Markov Chain (MC), Logistic Regression and Artificial Neural Network have been used by many resarchers to evaluate the current and possible future state of the urban areas. The Cellular Automata model is a computational method that can simulate the process of growth by using a set of simple rules in analyzig complex systems. This model is comprised of a set of rules that define the interaction of cells and collection of variables that lead to the investigation of various urban forms. The acpability of CA to simulate urban expansion, land use change, and population growth has become appropriate for simulating complex spatial systems. The Markov Chain model is a stochastic process that experiences transitions from one state to another considering certain probabilistic rules and predicts future changes based on the past. Logistic Regression creates a model that can best describe the relationship between dependent and independent variables using the least number of variables. The Artificial Neural Networks model is a powerful tool that models patterns using a machine learning approach. ANN output values vary from 0 to 1. In urban modelling, a value of 1 represents the greatest potential for future urban growth, while a value of 0 represents the least potential for future urban growth. Each of these methods has its own advantages and disadvantages, however, the integrated models have proved to be more accurate. The integrated CA–Markov model does multiple principle evaluation functions which combines Cellular Automata and Markov Chain models together. The MC is a widely known model, which predicts future change by considering the past, however it does not take into account the spatial distribution of the categories, while CA is a dynamic model which detects the spatial location of changes. To this end, the integrated CA-Markov model was used for predicting future urban expansion. In the modelling, first, the suitability map of each LULC class were created using Multi Criteria Evaluation (MCE) method. Second, transition area and transition probability matrices were determined by using the Markov Chain model. After that, the LULC map of Arnavutköy district in 2017 and LULC map of Hills Shire LGA in 2018 were simulated using CA_Markov model. Subsequently, the model was validated by comparing the actual and simulated LULC maps of Arnavutköy district and Hills Shire LGA. After obtaining the reasonable results of validation which were above 80% for Arnavutköy district and Hills Shire LGA, the suitability maps, transition area and transition probability matrices were created to predict the future LULC maps of 2050 for both study areas. To better evaluate the urban expansion of both study areas in the future, the area statistics of the LULC map of 2050 were calculated and the results revealed that urban growth will continue in the future by increasing 45% and 51% in Arnavutköy district and Hills Shire LGA, respectively. On the other hand, the forest area will decrease in Arnavutköy district by 26%, while in Hills Shire LGA 30% of the forest will be lost throughout the 32 years. This study presents a significant contribution to land use modelling, which logically integrates remote sensing data and ancillary data to be used as an input in CA- Markov model to successfully simulate and predict the temporal and spatial changes of LULC classes. Moreover, it has been determined that using different spatial pattern analysis such as Shannon's entropy and landscape metrics can help to better evaluate the urban expansion and model the future urban growth. The findings of this study can help to better evaluate the dynamic nature of built-up areas in both Arnavutköy district and Hills Shire LGA. The simulated future LULC maps can be used as a system of early warning to recognize the consequences of land use changes. The simulation results can also be investigated as a strategic guide for urban planners to better evaluate a complex system and create optimized land use managemnet that can better control urban expansion and environmental protection. In conclusion, this study highlights the importance of district-based urban expansion analysis. However, there are many research studies on urban expansion in big cities all over the world, but there are relatively few studies that consider district-based urban expansion in megacities. Most researchers mainly focus on the physical and socio-economic issues of districts neglecting the role of urbanization in the district. In this regard, considering this issue is very important, as district-based analysis of the causes and effects of urban sprawl can provide urban planners with more detailed insights to analyze and understand the nature of this phenomenon on a city basis.
Uzaktan algılama verileri temel alınarak verimlilik tahmininin oluşturulmasI

(Lisansüstü Eğitim Enstitüsü, 2023) Karimli, Nilufar ; Selbesoğlu, Mahmut Oğuz ; 798396 ; Uydu Haberleşmesi ve Uzaktan Algılama Bilim Dalı

Artan insan nüfusunun yeterli gıda ile temin edilmesine ilişkin kaygılar, dikkatleri Gıda Güvenliği alanına çekmektedir. Tarımsal verilerin geleceğe odaklı analiz edilmesi ve işlenmesi, bu alandaki planlama potansiyelini geliştirmekle birlikte gerekli önlemlerin önceden alınmasını da sağlamaktadır. Ancak, bu bölgelerin genişliği ve sayısı göz önüne alındığında, saha araştırması pahalı ve zaman alıcı bir prosedür olmaktadır. Uzaktan Algılama ve optik sensörlerin ortaya çıkmasıyla, çeşitli verileri uzaktan, hızlı ve düşük maliyetli bir şekilde elde etmek mümkün hale gelmiştir. Bu tez çalışması, Gıda Güvenliği alanında Uzaktan Algılama veri uygulamasının sınırlamalarını ve kapasitesini araştırmıştır. Sonuç olarak, Mamatkulov yaklaşımı ve MEDALUS modeli kullanılarak, Sentinel 1 ve Sentinel 2 verilerinden kışlık buğdayda oldukça doğru Verim Tahmini sonuçları (%98,03) hiçbir maliyet olmadan ve yüksek kullanılabilirlikle elde edilmiştir. Bu yöntem, regresyon modellerinin oluşturulmasını veya herhangi bir saha çalışmasını beklemeye gerek kalmadan, yeni oluşturulmuş veya önceki yılların verimliliği hakkında bilgi sahibi olmadığımız ekin alanlarının verimliliği hakkında tahminlerde bulunmayı mümkün kılabilir. Çıkan sonuca bakıldığında bu konuda daha kapsamlı analizler yapılabilir.
LTE için geni̇ş bantlı ve yüksek veri̇mli̇li̇kli̇ Doherty güç yükselteç tasarımı

(Lisansüstü Eğitim Enstitüsü, 2023) Koca, Kaan ; Savcı, Hüseyin Şerif ; Pınar Kent, Sedef ; Uydu Haberleşmesi ve Uzaktan Algılama Bilim Dalı

Bu tezde, LTE (Uzun Dönem Evrim) Band-7 ve Wi-Fi uygulamaları için uygun olan AB sınıfı ve C sınıfı GY ile tasarlanmıs ̧ genis ̧ bantlı bir Doherty güç yükselteç tasarlanmıs ̧tır. Çıkıs ̧uyumlandırmaag ̆ı,yardımcıyükselteçidealolmayansonsuz çıkıs ̧ empedansının parazitik cihazlar üzerindeki etkisine vurgu yapılarak teorik olarak analiz edilmis ̧tir. Yeni bir Doherty Güç Yükselteç (DGY) 25 W GaN HEMT (Yüksek Elektron Mobiliteli Transistör) ile tasarlanmıs ̧tır. DGY, temel tasarımlar ve açıklamalar için tipik olan 6 dB'lik bir OPBO (Güç Geri Çekme) deg ̆erini varsayar, çünkü ilgili voltaj seviyesi 1:4'tür ve tepe güç yükselteç giris ̧ voltajının dinamik aralıg ̆ının yarısında etkinles ̧tirilir. Ancak OPBO (Güç Geri Çekme) deg ̆erini arttırmak için öncelikle sinyalin PAPR (Tepe Etkin Güç Oranı) deg ̆eri ile uyumlu olması gerekmektedir. Asimetrik bir Doherty güç yükselteç tasarımı, tam çıkıs ̧ gücünde dog ̆rusallıg ̆ı korurken yüksek kazanç dag ̆ıtımını koruyarak verimlilig ̆i en üst düzeye çıkarmaya yardımcı olabilir. DGY, 3G (3. Nesil Mobil ̇Iletis ̧im) /LTE (Uzun Dönem Evrim) modülasyon hızı ayarlarında yüksek RF GY verimlilig ̆i sag ̆lamayı amaçlamaktadır. Bu, Doherty GY'nın yüksek bir ortalama çıkıs ̧ gücünde yüksek PAPR (Tepe Etkin Güç Oranı) için DE (drenaj verimlilik) artırmasına ve PAPR (Tepe Etkin Güç Oranı) zayıf oldug ̆u yerlerde GY ısınmasını önemli ölçüde azaltmasına olanak tanır. OPBO (Güç Geri Çekme) aralığı asimetrik DGY teknig ̆i kullanılarak genis ̧letilmis ̧tir. Daha önce DGY topolojisinde kullanılan çeyrek dalga dönüs ̧türücüsü, ilgili Klopfenstein tapper ag ̆ı ile deg ̆is ̧tirildi. Gerçek dünyadaki prototip uygulamalar, bu deg ̆is ̧iklig ̆in verimlilik deg ̆erlerini korurken geleneksel topolojilere kıyasla elde edilen DGY bant genis ̧lig ̆ini (BW) artırdıg ̆ını göstermis ̧tir (Kesirli bant genis ̧lig ̆i %24'e es ̧ittir). Çıkıs ̧ birles ̧tirici, optimum karakteristik empedans ve faz ofset deg ̆eri kombinasyonlarına sahip konik empedans transformatörleri ve yük empedanslarından olus ̧ur. Bu, yüksek çıkıs ̧ gücü seviyesinde yük modülasyonu ve yüksek geri tepme sag ̆lamak için yapılır. Simülasyonların bir sonucu olarak, DGY'nın CG2H40025 transistörlerle uygulanması, %69'dan daha yüksek bir doymus ̧ verimlilikle 79 W'tan daha yüksek bir çıkıs ̧ gücü sag ̆lar. Tüm frekans bandı boyunca, maksimum çıkıs ̧ gücü 47 dBm'den fazladır ve bu da bu transistörün maksimum güç is ̧leme faktörüne karşılık gelir. Verimlilik açısından, doygunlukta %69 ile %79 arasında ve 6-dB geri çekmede %50 ile %72 arasında deg ̆is ̧mektedir.
A statistical framework for degraded underwater video generation

(Graduate School, 2023) Şatak, Serkan ; Töreyin, Behçet Uğur ; 834354 ; Satellite Communication and Remote Sensing Programme

Computer vision in the underwater medium presents unique challenges due to the distinct properties and conditions encountered beneath the water’s surface. Underwater environments are characterized by limited visibility, color distortion, scattering of light, and various water conditions such as turbidity and currents. These factors severely impact the performance of traditional computer vision algorithms designed for terrestrial images, leading to significant difficulties in underwater image and video analysis. One of the primary hardships in underwater computer vision is the degradation of image quality caused by the attenuation of light. As light travels through water, it is absorbed and scattered, resulting in reduced contrast, loss of details, and color distortion. These effects make object detection, recognition, and tracking challenging tasks. Additionally, the scattering of light causes blurring and reduces the sharpness of underwater images, further impeding accurate analysis. Another significant hurdle is the lack of reliable, in-depth information. Estimating depth in underwater scenes is complex due to the varying water conditions and the absence of well-defined visual cues. This limitation poses challenges for tasks such as 3D reconstruction, scene understanding, and object localization.
A deep learning based framework for identification of ship types using optical satellite images

(Graduate School, 2023-01-18) Kızılkaya, Serdar ; Sertel, Elif ; 705152003 ; Satellite Communication and Remote Sensing

Today, monitoring of ship traffic in open and coastal seas is one of the primary and even indispensable activities of coastal countries for many reasons. Preventing activities such as illegal fishing, migration, smuggling, pollution of the seas, protection of underground resources and sea creatures, defensive reconnaissance, and surveillance activities can be listed as the main reasons for monitoring ship activity in the seas. Periodic monitoring of open and coastal seas using Earth observation satellites is a significant approach, providing fast and cost-effective results as well as large coverage extents. With this perspective, the thesis presents research about an effective and end to end ship monitoring approach via optical satellites. The use of deep learning (DL) techniques for the detection and classification of ships using optical satellite images, the creation of a ship database consisting of optical satellite images for the verification of the subject, and optical constellation modeling are the topics of this research. In the thesis, a comprehensive literature review is presented with the scope of gabs in the ship monitoring with using optical satellites. To mimic the real satellite image input, an optical satellite based image dataset – VHRShips - was formed. VHRShips consists of a total of 6312 images, 1000 without ships and 5312 with ships. There are a total of 11337 ships in the images. The database created includes 24 different types, and navy ships, one of these types, contain 11 different types. This dataset stands out with its rich ship database content and large sample size. Thesis reinterprets DL-based target detection and classification algorithms and proposes a flexible target detection and classification approach in a hierarchical design (HieD) that allows easy addition and removal of different algorithms. In addition, the detection and localization, recognition, and identification (DLRI) steps are staged and the outputs of the algorithm are detailed. In the phase of detecting, the presence of ships is verified in the images provided by the satellites. The determination of the positions of the ships in the images is carried out during the localization stage. The classification of the ships among the determined basic ship types is defined in the recognition stage. Lastly, the classification of navy ships, which is one of the main ship types, is handled during the identification stage. The feasibility of the developed approach has been demonstrated by the preliminary feasibility analysis for the covering of the Turkish surrounding seas with optical constellations. In this analysis, an optimization was carried out by using a software on how the satellite images required for the proposed method can be realized with a set of satellite design. It has been determined that 40 optical satellites are required to fully cover the selected sea area in 24 hours, and 100 optical satellites with the specified characteristics to be covered twice. The results of proposed method, HieD are presented in three formats which are; the individual stage performances, a comprehensive end-to-end evaluation and lastly a comparison with a well-known method, YOLOV4. The results of the thesis are very promising. F1 scores for detection, recognition, localization, and identification were respectively 99.17%, 94.20%, 84.08%, and 82.13% as a consequence of stage-by-stage optimization. The F1 scores for the same order after complete implementation of our suggested method were 99.17%, 93.43%, 74.00%, and 57.05%. End-to-end YOLOv4 produced F1-scores for DLRI of 99.17%, 86.59%, 68.87%, and 56.28%, in opposition. For the steps of localization, recognition, and identification, we outperformed YOLOv4 using HieD. In the thesis; it has been shown that the ship detection, localization, recognition and identification method developed in a hierarchical structure using a challenging data set containing images with different backgrounds in open and coastal seas in many different geographies, works successfully. The method is very open to development with the antecedent and intermediate methods to be added. The data set created in the thesis has been used with various detection and classification techniques, proving that there is no dataset that provides a limited opportunity for the study. VHRShips can serve as a standard for the developed approach as well as for further research into the use of deep learning and advancing geospatial AI applications in the maritime environment. Finally, the thesis presents a vision of the future along with a fair self-criticism.
Akıllı yansıtıcı yüzey destekli telsiz haberleşme sistemi ve insansız hava aracı konumlandırma

(Lisansüstü Eğitim Enstitüsü, 2023-05-05) Aslandoğan, Emir ; Yazıcı, Mehmet Akif ; 705191025 ; Uydu Haberleşmesi ve Uzaktan Algılama

Akıllı yansıtıcı yüzeyler (Reconfigurable Intelligent Surface, RIS) telsiz haberleşme sistemlerindeki kullanım olasılığı ve bunu destekleyen bir çok çalışma bu teknolojinin hem endüstriyel hem de akademik anlamda dikkat çekmesini sağlamıştır. Olağan haberleşme sistemlerinin propagasyon ortamı üzerindeki kabiliyeti oldukça sınırlıydı. Meta malzemelerin geliştirilmesi ve buna bağlı olarak RIS'ler üzerine yapılan çalışmalar propagasyon ortamı üzerindeki kabiliyetimizi artırdı. Bu sebeple telsiz haberleşme sistemlerindeki kanallara ait beklenmeyen bozucu etkilerin ve uygulamalarda karşılaşılan sınırlamaların RIS teknolojisi ile birlikte oldukça azalacağı öngörülmektedir. RIS'ler düşük maliyetli küçük devre elemanlarından oluşmaktadır. Bu sebeple üretim maliyeti açısından günümüz teknolojilerine yükü çok fazla olmayacaktır. Ayrıca bina ve araç yüzeylerine kolaylıkla entegre edilebilir yapıdadır. Bu, RIS teknolojisinin kolaylıkla ve düşük maliyetle kurulumunun yapılacağını göstermektedir. Tezin amacı RIS'in İHA konumlandırma ve güzergah oluşturma sistemlerinde, sabit ve hareketli kullanıcıların bulunduğu telsiz haberleşme sistemi için optimizasyon algoritması kullanarak enerji performansı iyileşmesi sağladığını göstermektir. Bu kapsamda RIS'in İHA'ya entegre edildiği iki kullanıcılı telsiz haberleşme sistemi üzerinden RIS'in performans analizi gerçekleştirilmiştir. Tezin ikinci bölümünde ise sabit kullanıcıların bulunduğu İHA güzergah belirleme çalışmasından yararlanarak, hareketli kullanıcılar için İHA konumlandırma ve güzergah oluşturma sistemi oluşturulmuştur. Bu kapsamda iki sisteme ait performans analizleri optimizasyon algoritması üzerinden gerçekleştirilmiştir. İlk olarak bu tezde RIS'lerin olası kullanım senaryoları üzerine bilgi verildi. Milimetre dalga haberleşme, eş zamanlı bilgi ve güç transferi, fiziksel katman güvenliği, mobil uç hesaplama, cihaz-cihaz haberleşmesi ve insansız hava aracı haberleşmesi alanında yapılan çalışmalardan bahsedildi. Sonraki kısımda RIS yapısı ve çalışma prensibi incelendi. Üç farklı RIS türü hakkında bilgi verildi. Ayrıca RIS kanal modellerinden bahsedildi. Oluşturulan RIS destekli telsiz haberleşme sistemi sönümleme ve gölgeleme etkisinde olduğundan belli başlı sönümleme ve gölgeleme modellerine de değinildi. Bu tezin diğer kısmında RIS üzerine performans analizi gerçekleştirilmiştir. Görüş hattı iletiminin olmadığı ve ortamdaki bozucu etkilerin var olduğu ortam göz önünde bulundurulmuştur. Bu sebeple görüş hattı iletiminin olmadığı senaryo için sönümleme ve gölgeleme etkisi varlığında sistem modeli oluşturulup kesinti olasılığı üzerinden sistem performansı incelenmiştir. Oluşturulan sistemde RIS'in İHA'ya entegre edildiği düşünüldü. Öncelikle iki kullanıcılı sistem için RIS üzerinde bulunan yansıtıcı eleman sayısına bağlı olarak kesinti olasılığının değişimi gözlemlenmiştir. Bu işlem gerçekleştirilirken Nakagami-m sönümleme ve ters Gamma gölgeleme etkisi altında sonuçlar elde edilmiştir. Yansıtıcı eleman sayısının kesinti olasılığına etkisini doğru gözlemlemek için sönümleme ve gölgeleme parametreleri bu inceleme esnasında sabit tutulmuştur. Elde edilen sonuçlarda yansıtıcı eleman sayısı artışının transfer için gereken verici gücünü ciddi miktarda düşürdüğü gözlemlenmiştir. Örneğin -2.5 dB verici SNR değeri için N=8 yansıtıcı eleman sayısında kesinti olasılığı değeri 3.7x10^-1 olarak hesaplanırken N=16 yansıtıcı eleman sayısında kesinti olasılığı değeri 1.6x10^-3 olarak hesaplanmıştır. Oluşturulan sistem modelinde bir diğer gerçekleştirilen inceleme gölgeleme ve sönümleme etkisinin şiddetidir. Bu inceleme yapılırken yansıtıcı eleman sayısı ve RIS ve kullanıcılara ait mesafeler sabit tutulmuştur. Nakagami-m sönümleme ve ters Gamma gölgeleme etkisinin incelenmesi için bu modellere ait biçim parametreleri değiştirilmiştir. m=1,1.5,2 ve α=2,2.5,3 değerleri için kesinti olasılığı hesaplanmıştır. Sönümleme ve gölgeleme etkisinin kanal performansını düşürdüğü ve kesinti olasılığı üzerinde artırıcı etki yaptığı görülmüştür. Kötü kanal koşullarında belirli kesinti olasılığı değeri altında kalmak için daha fazla verici gücü harcanacağı gözlemlenmiştir. RIS üzerine yapılan bu çalışma ile telsiz haberleşme sistemlerinde RIS'in sistem performansını artıracağı gözlemlenmiştir. Bu bulgular doğrultusunda, RIS teknolojisinin İHA konumlandırma ve güzergah planlama uygulamalarında kullanılabilirliği vurgulanmıştır. Sabit ve hareketli kullanıcılar için, RIS'li ve RIS'siz telsiz haberleşme sistemleri üzerinde ayrı ayrı İHA güzergah optimizasyonu üzerine çalışılmıştır. Ayrıca, RIS varlığında faz optimizasyonu sağlandığında performans analizi yapılmıştır. İlk aşamada, sabit ve hareketli kullanıcılar için RIS+P, RIS-P ve NO-RIS senaryoları için 3-boyutlu güzergahlar oluşturulmuştur. Bu güzergahlar, sistem enerji performansını doğrudan etkilemektedir. Hareketli kullanıcıların bulunduğu senaryoda, sistem performansında düşüş gözlemlenmiştir. Ancak RIS+P ve RIS-P senaryolarının, NO-RIS senaryosuna göre enerji performansını önemli ölçüde artırdığı ve enerji verimliliğini sağladığı görülmüştür. Sonuç olarak, bu bulgular, RIS teknolojisinin kullanılmasının, İHA uygulamalarındaki enerji performansını geliştirme potansiyeline sahip olduğunu göstermektedir. Benzer şekilde, sabit ve hareketli kullanıcılar için ayrı ayrı sistemin ortalama veri hızı ve throughput performansları incelenmiştir. RIS'in kullanıldığı senaryolarda performans iyileşmesi gözlemlenmiştir. Bu tezde, telsiz haberleşme sistemlerinde kullanılmak üzere geliştirilen İHA ve RIS teknolojilerinin etkinliği incelenmiştir. Elde edilen sonuçlar, RIS destekli İHA konumlandırma uygulamalarının hem sabir hem de hareketli kullanıcılar için enerji performansı açısından avantajlı olduğunu göstermektedir. Bu nedenle, gelecekte İHA konumlandırma ve güzergah oluşturma uygulamalarında RIS'in kullanılacağı öngörülmektedir.
The spatio-temporal dynamics of aerosols in the Marmara region and impact of land cover/use on atmospheric environment

(Graduate School, 2023-06-16) Osgoui Ettehadi, Paria ; Kaya, Şinasi ; 705152002 ; Satellite Communication and Remote Sensing

The impacts of urbanization and industrialization on air quality are well known, and air pollution control strategies have been implemented, but the effectiveness of these strategies is limited to developed countries. Developing countries continue to rely on fossil fuels and experience air pollution from agriculture emissions, crop residue burning, biomass fuel, and low-quality coal combustion. Air pollution contributes to health, visibility, ecosystems, and climate change. Solid air pollutants, also known as particulate matter (PM) consist of tiny particles suspended in the air has the most severe health impacts. Governments need to measure, monitor, and control the concentration of air pollutants and maintain the pollution levels under the values defined by the World Health Organization (WHO). There have been some efforts to reduce air pollution in developing countries. However, there is still a long way to go in terms of clean air in many developing countries. The data collected from air quality monitoring (AQM) stations are commonly used to study the impacts of air pollution on human health. However, the initial investment and maintenance of ground-based AQM stations are expensive. The limited quantity of air quality monitoring stations on the ground poses a hindrance to conducting thorough research into the effects of elevated levels of air pollution on the environment. Measurements from earth-observing satellites can be a useful substitute for ground-based data in environmental studies. Although ground-based data are often more detailed and accurate, satellite data provide a broader view of larger spatial extents. The advantages of satellite data collection over a lengthy period allow for long-term monitoring of environmental trends and changes. Aerosols, tiny particles that are suspended in the Earth's atmosphere can be of natural or human-made origin. The impact of atmospheric aerosol loading on various aspects of the Earth's climate and environment cannot be overstated. Besides its adverse effect on human health, it affects global temperature, atmospheric radiation, the Earth's albedo, and terrestrial heat budget, as well as clouds and precipitation processes, and ecological systems. Therefore, understanding atmospheric aerosols is essential for comprehending Earth's climate and ecosystem. The columnar aerosol pollution or the number of microscopic aerosol particles in a vertical slice of the atmosphere can be measured by satellite sensors. The main remotely sensed geophysical quantity and column-effective particle property is total column aerosol optical depth (AOD). The AOD is a measure of how much sunlight is absorbed or scattered by aerosol particles in the atmosphere, and it provides information on the concentration and distribution of atmospheric aerosols. The concentrations of PM and AOD are important measures of air pollution. They represent the amount of particulate matter present in the air and are strongly correlated with each other. While ground-based air pollutant concentration records show the concentration of pollutants near the Earth`s surface, satellite-based and ground-measured AOD data provide a measure of the total column amount of aerosols from the Earth's surface up to the top of the atmosphere. Satellite-based AOD measurements can identify hotspots of particle pollution and short-term spikes, which can then be targeted for more detailed ground-based measurements of air pollution. Space-based remote sensing plays a vital role in characterizing the spatial and temporal distributions of atmospheric aerosols from the local to global scale. The popularity of satellite aerosol products has led to the development of various aerosol retrieval algorithms. To ensure the accuracy of satellite aerosol data in interpreting regional and global aerosol patterns, it's crucial to evaluate the performance of the retrieval algorithms. Validation using accurate ground-based AOD measurements is employed for this purpose. AErosol RObotic NETwork (AERONET) program is a federation of ground-based remote sensing aerosol networks, a vital worldwide network for monitoring aerosols that employ numerous sun-photometers stationed across the globe to measure various aerosol optical and microphysical characteristics, such as AOD. It's important to note that the efficacy of satellite aerosol products may vary depending on factors such as geography, climate, and weather conditions. Thus, it's essential to identify the most reliable algorithm by validating satellite AOD retrievals against the nearest ground-based data to the region of interest. The primary aim of this research is to better understand the behavior of aerosols in the atmosphere and gain insight into the spatial distribution and temporal variability of aerosols in the region, as well as to identify the factors that influence the magnitude of AOD and its correlation with land cover/use (LCU). This understanding will aid in devising an efficient strategy to manage environmental air pollution. Furthermore, by gathering, verifying, and analyzing appropriate data through particular techniques, the alterations in aerosol concentrations over a prolonged period will be assessed. For this major objective, the study is divided into three key sections. First, various aerosol products are evaluated for their accuracy and reliability by comparing satellite-derived AODs with AERONET AOD measurements at three different sites. The most effective AOD product were determined by comparing the performances of aerosol data sets. In this context, multiple Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS) AOD products were compared with AERONET AOD measurements at various surface classes/types (i.e. land, ocean, coastal) in the eastern Mediterranean over five years (2014-2018) and the results were validated. MODIS aerosol products based on Dark Target (DT), Deep Blue (DB), and Multi-Angle Implementation of Atmospheric Correction (MAIAC) are available at different spatial resolutions (1 km, 3 km, and 10 km), while the VIIRS aerosol product has a resolution of 6 km. To obtain VIIRS AOD values, the improved DB approach was used for land and the Satellite Ocean Aerosol Retrieval (SOAR) method for the ocean. The best-performing products over the urban/land surface are the MODIS MAIAC and VIIRS DB aerosol products with Root-Mean-Squared Errors (RMSE) of 0.048-0.061 and a higher percentage (81%-89%) of retrievals falling within the expected error for land. This is while the MODIS MAIAC AOD data also has the advantage that it has a better spatial resolution of 1 km, which enables it to capture sub-grid aerosol features and a higher number of AOD retrievals. In the second section, I focus on the spatiotemporal variation of MODIS MAIAC AOD over the Marmara region. The objective of this section is to pinpoint areas within a region that have high AOD levels and examine how various environmental factors such as seasonal weather patterns, local emission sources, and LCU are linked to the presence of aerosol pollution in those areas. Our study area was the Marmara region because it is subject to the production of aerosols originating from diverse sources, both natural and anthropogenic. The Marmara Region is the country's most populous region, despite being the second smallest geographically. This is due to Istanbul being located there. The region also contains other important developing cities like Bursa, Kocaeli, and Tekirdag. It is a significant economic center with a lot of agricultural, commercial, and industrial activity. Aerosol formation in the region is influenced by a range of sources, including sea salt, agricultural practices, maritime transport, industrial activities, and the seasonal transportation of dust from the Sahara Desert. To accomplish the objective of the study's second part, I analyzed the MODIS MAIAC AOD data at annual, monthly, and seasonal scales between 2000 and 2021 to investigate the spatiotemporal variability of AOD in the Marmara region. The monthly mean AOD increases gradually from January to May and fluctuates between May and August and reaches its highest value in August. The monthly mean AOD decreases after August and reaches its lowest monthly mean at the end of the year during December. Seasonal variation in AOD is significant: summer (0.148) > spring (0.136) > autumn (0.116) > winter (0.09). According to an analysis of the AOD's multi-year variation, there were two maxima for the AOD between 2000 and 2010 with values of 0.146 and 0.137 in 2002 and 2008, respectively. AOD exhibits a decreasing tendency from 2000 to 2021, with a 0.005/yearly decline. Using the MODIS MAIAC AOD data at a 1-km scale, I performed a comprehensive assessment of aerosol loading and statistical-visual analyses highlighting the influence of land use/cover on aerosol properties. The findings revealed that there are significant regions with high aerosol concentrations across the region and that these regions show significant temporal variations. The aerosol loading was higher over the western side of the Marmara region (Edirne, Tekirdag, Kırklareli, Canakkale, Istanbul, Kocaeli) until 2011 while the eastern part of the region (Bursa, Sakarya, and Balikesir) was exposed to higher aerosol concentrations after 2016. From 2000 to 2021 the largest number of days with lower aerosol pollution is seen in Bilecik (≈ 35%) while Edirne experienced the highest percentage of days with the highest aerosol pollution level (≈ 9%). In the region, aerosols are mostly generated by urban activities and industries, as well as mineral aerosols originating from soils. Finally, a novel approach to LCU classification is proposed. The last part of the paper proposes a strategy to identify and distinguish different LCU patterns in Mediterranean cities. The focus was on separating built-up areas from bare land, which can be challenging due to urban landscape complexity and heterogeneity. Separation is also necessary since urbanized/industrialized zones and bare soil areas contribute significantly to atmospheric aerosol pollution. For this purpose, the separation of these two classes was well-addressed by using the proposed multi-index methods on satellite image data. The multi-index combination of the normalized difference tillage index (NDTI), the red-edge-based normalized vegetation index (NDVIre), and the modified normalized difference water index (MNDWI) showed outstanding overall performance with 93% accuracy and a 0.91 kappa value for all LCU classes. The improvement achieved in separating built-up regions from bare land is of substantial significance, as it significantly reduces the misclassification of bare land as built-up regions. This is particularly important for aerosol studies in the study region, where the two factors with the highest impact on aerosol loading are found. The enhanced accuracy of land cover classification provided by this improvement can greatly enhance the reliability and precision of aerosol studies in the area.
Entegre mast yapısının gemi radar kesit alanı üzerindeki etkilerinin incelenmesi

(Lisansüstü Eğitim Enstitüsü, 2023-06-23) Çakal, Seyhan ; Helvacı, Mustafa ; 705201015 ; Uydu Haberleşmesi ve Uzaktan Algılama

Düşük görünürlük teknolojilerinde, uçar, yüzer ve kara platformlarının radar, kızılötesi ve sonar gibi algılayıcı sistemler tarafından tespitinin önlenmesi amaçlanmaktadır. Askeri açıdan bakıldığında düşük görünürlük kavramı, platform veya platorma ait alt sistemlerin düşman radar sistemleri tarafından tespitine, teşhis ve takibine yakalanmadan ilerlemesi platformlara avantaj sağlamaktadır. Yüzer, uçar ve kara platformları düşük görünürlük teknolojisi konsepti dahilinde incelendiğinde bahse konu platformlar için "Radar Kesit Alanı" kavramının önemine değinmek gerekmektedir. Kabaca bir cismin radar kesit alanı, hedefin elektromanyetik sinyali yansıtıcılığının bir ölçütü olarak tanımlanır. Platformun radar kesit alanını azaltmak, hem platformun savaş gemilerine konuşlandırılan radar sistemleri tarafından geç algılanmasını sağlamakta hem de teşhis ve takibini zorlaştırmaktadır. Bu nedenle, radar kesit alanı savaş gemileri için önemli bir tasarım kavramı haline gelmektedir. Savaş gemileri gelişen radar teknolojisinden faydalanarak, tehdit platformları kolaylıkla tespit edebilmektedir. Ancak roller değiştiğinde platformların yeni nesil radar sistemleri tarafından fark edilmeden hedefine ilerlemesi oldukça zordur. Gemiler üzerine yerleştirilen antenler, silah sistemleri ve birçok sensör platformların radar kesit alanını artırmaktadır. Görülmeden gören platformlar geliştirmek için literatürde çeşitli yöntemler mevcuttur. Bu yöntemlerden nispeten maliyet etkin bir yöntem olan şekillendirme, gemilerin radar kesit alanı üzerinde önemli bir etkiye sahiptir. Günümüzde, gemi üzerinde çeşitli mevkilerde konuşlanmış radar sistemlerinin radar kesit alanı üzerindeki etkisinin azaltmak maksadıyla geliştirilen entegre mast sistemlerinden faydalanılmaktadır. Bu çalışmada, platformlarda yerini almaya başlayan yeni nesil entegre mast yapılarının gemilerin radar kesit alanları üzerinde etkilerinin incelenmesi amaçlanmıştır. Çalışma kapsamında ilk olarak, iki farklı mast yapısı tasarlanmış, gemi üzerindeki etkisini incelemek amacıyla 3 boyutlu gemi modeli oluşturulmuş ve son olarak her bir mastın etkisini incelemek maksadıyla gemi üzerine konuşlandırılarak analizler koşulmuştur. Bahse konu analizlerde radar kesit alanı tahmin yazılımları tarafından sıklıkla kullanılan "fiziksel optik" yöntemi kullanılmış, elektromanyetik propagasyonda kaybın en düşük olduğu iki farklı frekans bölgesi seçilmiş ve yanca/yükseliş açıları için çeşitli açı değerleri seçilerek çalışmalar yapılmıştır. Analizler neticesinde; nispeten daha düz yüzeylerin geminin radar kesit alanına ne yönde katkı sağladığı ve tasarlanan iki farklı mast yapısının platformun radar kesit alanı üzerindeki etkisinin bakış açısına (yanca/yükseliş), polarizasyona ve frekansa bağlı olarak nasıl değiştiği incelenmiş ve grafiklerle sergilenmiştir.
Earthquake damage detection with satellite imagery and deep learning approaches: A case study of the february 2023, Kahramanmaraş, Turkey earthquake sequence

(Graduate School, 2023-08-14) Elik, Fatma ; Sertel, Elif ; 705201004 ; Satellite Communication and Remote Sensing

In recent years, the fusion of deep learning techniques, remote sensing technology, and artificial intelligence (AI) has profoundly transformed the field of disaster management and damage assessment. The increased availability of high-resolution satellite imagery and advanced computer vision techniques now makes it possible to analyze Earth observation data at a large scale and with unparalleled precision. This thesis investigates the application of remote sensing and deep learning techniques to perform post-earthquake damage classification using computer vision and focuses specifically on the earthquakes that occurred on February 6th, with an emphasis on Kahramanmaraş province. The objective of this thesis is to investigate the potential of a variety of deep learning techniques, evaluate their accuracy in recognizing structurally compromised buildings, and utilize satellite imagery in conjunction with diverse open-source spatial data to enhance research on earthquakes. This master's thesis specifically delves into the integration of remote sensing, computer vision, and earth observation methods within the field of geophysics and earthquake studies. Thus, in this study it is aimed to showcase the application of computer vision in the analysis of post-earthquake damage and underscore the importance of rapid intervention in such critical situations. The thesis places significant emphasis on the use of satellite imagery and pixel-based classification for the classification of images in earthquake damage assessment. The UNet, DeepLabV3, and PSPNet architectures are implemented using the ArcGIS Pro API for Python, an innovative and supportive tool for scientific research. The primary data source for the investigation is RGB images from Maxar Technologies. The research examines three cities that were affected by the February 6, 2023, Kahramanmaraş earthquake sequences: Kahramanmaraş, Hatay, and Gaziantep. Damage-assessed data points are received thanks to Yer Çizenler Non-Governmental Organization (NGO), and recently modified building footprints are taken from Humanitarian OpenStreetMap (HOTOSM), and they are all used to analyze the damage. Labeled polygons are generated within a 5-meter distance of the damage points. However, assigning values for further and closer distances has a negative impact on the model accuracy. The training data, exported based on the satellite imagery and damage level assigned data points, provides a balanced dataset for Kahramanmaraş, where the building footprints match the images most effectively. In Hatay, the damage level assigned data distribution is the most balanced, but the building footprints do not align well with the images. Gaziantep presents a good match between the building footprints and images, but the distribution of the damaged data classes is highly imbalanced. Consequently, the decision is made to focus on training the model for Kahramanmaraş province due to the similarity in roof and building types, which has the potential to adapt the approach to other cities in the region as well as the earthquake-affected region under investigation. Image sizes of 256x256 pixels with 128 strides and 4 batches gave us the optimum model results among other options in the DeepLabV3 ResNet50 encoder. In conclusion, this master's thesis demonstrates the potential of combining remote sensing, computer vision, and earth observation techniques for geophysics and earthquake studies. Also, it is aimed to use different data types from open sources and use these different data types to make damage detection after earthquakes. The utilization of the ArcGIS Pro Python API, satellite imagery, pixel based classsification, and labeled training data provides insights into damage assessment after earthquakes, with Kahramanmaraş Province serving as the focal point for model training. The findings contribute to the development of efficient and accurate disaster management strategies and lay the foundation for further research in this field.

Gözat

Çıkarma tarihi ile LEE- Uydu Haberleşme ve Uzaktan Algılama Lisansüstü Programı'a göz atma

Sayfa başına sonuç

Sıralama Seçenekleri