A mobile plant identification application for environmental monitoring

Abstract

In this paper, we present a computer vision-based mobile plant classification application. The objectives to develop such a mobile application is to utilize smartphones and benefit from the photographs taken by their camera in order to perform environmental monitoring. Moreover, enabling individuals to use such a technnology is expected to increase individuals' awareness to their environment. The developed application identifies the plants by processing a photograph of its leaf. The main processing steps of the system are: determining whether there is a leaf in the photo, detecting whether there is a blur, leaf segmentation, feature extraction, and classification. Each building block of the system is tested separately and promising results have been obtained in the experiments.