Certain investigation on crop disease detection using enhanced classification model

Abstract

Agriculture plays a vital role in the history of human civilization. newlineAgriculture contributes more than 17% of total Indian GDP and provides newlineemployment to over 60% of the population. Horticulture is a division of newlineagriculture which addresses the science of plant cultivation in a scientific newlinemanner. Every year, India loses on an average of 15 25% of crops due to lack newlineof proper support to disease-affected crops and it causes heavy economic loss newlinenot only to the farmers but also to the country. Hence quality assurance is newlinerequired to perform examining the food products and evaluating the farmland newlinefor the disease. The image processing has been proved to be an effective tool newlinefor analysis in various fields and applications including agriculture and newlinehorticulture. The information extraction provides the eminent processing of newlinedata using the digital imaging system which is very helpful to assist the Agriscientist. newlineIn image processing, the digital images are captured and the newlinecorresponding image manipulation operations are performed in the captured newlineimage. After a digital image is obtained, the pre-processing operation is newlineinvoked. The key function of pre-processing is to improve the image in order newlineto get better results for the other processes. It typically deals with techniques newlinefor enhancing contrast, removing noise and isolating regions. During the newlineprocess of noise removal in the image, the enhanced filtering methodology newlineusing the combined form of the linear filter and the probabilistic filter is newlineproposed. newlineThe first step in image analysis is to segment the objects present in newlinethe denoised and enhanced image. Segmentation subdivides an image into its newlinesignificant parts in terms of objects. In general, autonomous segmentation is newlineone of the most difficult tasks in image processing. newline