Convolutional Neural Network Based Potato Leaf Disease Detection

Abstract

Agriculture is an area that has a significant influence on human lives and economic condition. Inadequate management of agricultural goods in sectors leads to loss. Farmers generally lack the understanding of diseases and hence produce less crops and lead to losses. Kisan telephone centers are accessible but do not offer 24-hour service and communications break too often. Farmers cannot effectively describe their sickness on phone and this is not the way they actually want to. While photographs and videos from crops provide a greater insight, agro-scientists can offer a better way of resolving healthy crop problems and farmers do not have so much knowledge and information. It should be mentioned that, if the crop yield is not healthy, then healthy nutrition will be diminished. With technological development and intelligent recognition of disease, earlier treatment can be used to reduce adverse harvest effects. The focus of this work is primarily on the identification of plant diseases by image treatment. This study uses an accessible dataset of 1000 pictures of potato leaves separated into two healthy and diseased groups. This research provides a model CNN for the detection and classification of potato leaf disease. For automated extraction and categorization of functions we have designed a CNN.