Plant Leaf Disease Classification using Deep Learning: A Survey

With the increase in global population, food supply must be increased correspondingly while simultaneously protecting crops from numerous fatal diseases. Traditionally, plant disease identification was done by naked eyes by using experience-based studies of farmers and plant pathologists. Performing the traditional process is difficult, time-consuming, and offered inaccurate diagnosis at times, resulting in significant economic loss in agribusiness. Later, several studies have employed machine learning in the field of plant disease identification, but the findings were not promising and were too slow for practical use. Recently, Convolution Neural Networks have made an essential breakthrough in the field of computer vision due to their characteristics like automatic feature extraction and leverage effective results with small dataset in a short span of time when compared to machine learning. This paper discusses about the challenges faced in identifying the plant leaf diseases and it tries to solve the problem of inaccurate and time consuming analysis of disease detection and classification by reviewing different methods and state-of-the-art algorithms, which are trying to overcome this issue.