Plant Disease Detection Techniques based on Deep Learning Models: A Review

Abstract

Plants must be checked at an early stage of their life cycle in order to avoid illnesses. Visual observation, which takes longer, and costly expertise are the conventional approach utilised for this monitoring. Therefore, illness detection systems need to be automated in order to speed up this procedure. This study analyses the possibility of technologies for the identification of pest leaf diseases in plants to support agricultural growth. It covers many processes, such as image retrieval, image segmentation, extraction of features and classification. Two key phases comprise plant disease detection technology: segmentation of an open input to detect the ill portion and an extraction approach to extract the image feature and classify the functionality that is removed using different classifiers. The technology consists of two important steps. In this study, segmentation, characteristic removal, and classification approaches are examined and clarified from the perspective of different parameters.