FOLIAR DISEASE DETECTION IN APPLE TREES

Apple is the key crop of Kashmir valley.Apple occupies a prominent place in its agricultural economy because of right climatic conditions with comparative advantage for apple farming.The main reason for limited apple cultivation is insects and diseases.Apple cultivation is vulnerable to various plant diseases, significantly affecting fruit yield and quality.Among them, Apple Scab, Rust, and co-occurring multiple diseases are few very common and destructive.In way to protect the plants from diseases it is necessary to detect them at early stage.Detecting at early stage and providing accurate verdict of these diseases are crucial for the disease management and prevention.The current work is carried out using CNN algorithm on pycharm and jupyter notebook.For time constraint the 2 nd and 3 rd layer of CNN algorithm is combined to increase the efficiency of the algorithm.The dataset used is freely available on kaggle publishes by the cornell university.The classification is done on the basis of parameters as entirely green with no spots for healthy leaf, Brown spots for leaves affected by scab, Yellow spots for leaf affected by Cedar rust and yellow and brownish spots for leaves possessing multiple disease.As a result the system classifies the leaf image based on the parameters into scab, rust, healthy and multiple disease.