Leaf Disease Detection And Classification System For Soybean Culture

Our country's economy highly depends on agricultural productivity and thus disease detection plays a major role in agricultural field. The aim of this project is to support the farmers for detecting the type of disease in soybean culture. The idea is to identify whether the leaf is healthy or diseased and if it is affected, finding out the disease and to identify the percentage of infection. The segmentation phase is completed with the help of clustering algorithm and followed by classification using unsupervised learning algorithm. The system is trained using combinations of color and texture features. Using our idea it is possible to identify the soybean disease with 91% accuracy in average.