DETECTION OF CROP DISEASES ON BASIS OF VARIOUS IMAGE PROCESSING TECHNIQUES

Crops play an important role in our daily meal. But diseases in crops cause major production, so identification and quantification of crops diseases are required for adequate crop production and the determination of crop losses and design of breeding strategies in agriculture. The detection of infected or defected leaves have been done by farmers using naked eyes, which is not an accurate way and its time consuming, which contribute to many errors. So the appropriate ways are needed to be sure the amount of pesticides, insecticides and weedicides to be applied. Automatic detection of crop diseases is an essential research topic as it may prove benefits in monitoring large fields of crops and thus automatically detect symptoms of diseases. The analysis of plant leaves can be effectively done using an image processing by capturing an image of a certain crop leaf followed by extracting a predefined feature from the captured image and finally analyzing these features based on image processing techniques, which would decide the diseases and would also detect the type of crop diseases at early stages and enables early control and protection measures. Main concern of proposed system is to detect the crop diseases automatically. With the help of proposed method, five crops rice(drechslera oryzae), cotton(alternaria gossypina), groundnut(bud necrosis virus), sugarcane(collectotrichum falcatum), wheat(puccinia recondite) diseases will be detected. It is based on feature extraction of an image and various comparison techniques of an image processing.