SubCoPLeD: Superpixel Based Color Distribution Driven Plant Leaf Disease Detection

Plant leaf diseases are haveily threatening the agricultural products. It is responsible for losses in agriculture economic. Observing plant leaf for early detection of diseases is essential to control disease spread and reduce its impact on the agriculture economic. Therefore, plant leaf disease automation is required for effectively controlling the plant leaf diseases in agricultural activities. Recent advances in image processing techniques emerged new possibility for plant leaf disease detection in effective manner. This paper present a simple and robust methodology for early and fast plant leaf disease detection using visual of plant leaf. The proposed method utilizes superpixel of the image in several color models jointly with empirically leant thresholds for identification of plant leaf either diseased or non-diseased. This method The experimental results and analysis on publicly available Plant village dataset supports the advantages of the proposed method.