Computing Amount of Disease in Crop Using Artificial Intelligence Techniques

Abstract

Artificial intelligence techniques and the image analysis technology have a vital role to play in newlinebiology and agricultural sectors. As it is well known fact that, a healthy crop can yield quality newlineoutput if a preventive measure is adopted. Automatic detection of plant diseases and cultivation newlineof healthy crops is of great importance and agricultural automation. The term plant disease is newlinedefined as any impairment happening to the normal physiological function, producing newlinecharacteristic symptoms. The studies of crop diseases refer to studying the visually observable newlinepatterns of a particular crop leaf. The identification of crop, leaves, stems and finding out the newlinepests or diseases, or its percentage is found very effective in the successful cultivation of crops. newlineThe manual observation is usually practiced by the farmers for the detection and identification of newlinecrop diseases. It requires continuous monitoring and which is difficult work for farmers on large newlinefarms. Expert knowledge. With the aid of imaging technology the crop disease detection systems newlineautomatically detect the symptoms that appear on the leaves and stem of a crop and helps in newlinecultivating healthy crop in a farm. With the help of imaging technology system on parts like newlinesuch as leaves and stem and any variation observed from its characteristic features, variation will newlinebe automatically identified and also will be informed to the user. This thesis provides an newlineevaluative study on the existing disease detection well in advance for preventive measures. newline In this work, Employed a crops database collected from OSF-Home, Plant village newlineAgriculture University and also a self prepared dataset are employed for the study. Features like newlinetexture color, shape are employed and to detect the disease, suitable classification techniques newlinelikes CNN, ANN, SVM, AOSMO-CNN, and HCO-DNN. This research work presents the newlineestimations of crop disease to an accuracy level of 97.44% Artificial intelligence techniques and the image analysis technology have a vital role to pla