Reviewing Important Aspects of Plant Leaf Disease Detection and Classification

Abstract

Agriculture, being the dominant industry from the point of view of economical growth of countries like India, plays a vital role in fulfilling the demand of food. However, extreme weather conditions and several climate changes may invite notable infectious diseases in plants caused by fungi, viruses and bacteria. These plant diseases can be a major threat to food supply and hence it is important to identify and prevent the plants from the diseases at the early stages. The conventional approaches were dependent on the experts in the field and hence time consuming. Since the technology is upgrading day by day and has plenty of its advantages in plant leaf disease detection field as well, various disease identification approaches using different domains have been proposed in the literature to detect and cure the plant diseases that occur on the plant leaves. Although, many of the existing approaches have provided better results, challenges exist in order to achieve optimized results of plant leaf disease detection process. This paper reviews different methodologies under image processing, machine learning, deep learning and swarm intelligence domains for plant leaf disease detection. Understanding of various diseases that occurs on plant leaves is very important in order to deal with it; hence this paper provides a detailed taxonomy about the different plant diseases and dataset that is popularly used in various existing approaches for training and testing purpose of plant leaf disease detection and its classifications.