Recognition of plant disease by photographs of the leaf: A comparative analysis for understanding perspectives

Abstract

The agricultural sector plays a key role of any financial system and depends on the yield of farmers and ranchers. This output has often been affected by micro-level conditions that occur while the fruit-bearing crop is growing. These viruses are numerous and hence improvement in image processing is useful to establish and suggest remedies. For example, Fusarium oxysporum, Mycosphaerella musicola, Gloeosporium musae, Erwinia Carotovora, Pseudomonas Solanaceanim, Pentalonia nigronervosa, Erionota thrax, BSV and BBM Virus are some of the diseases that affect the banana leaf. A large section of research carried out in this area is directed at linear approaches in which the identification of visually visible diseases includes classification, extraction of attributes and definition. But those modalities are not suitable for large and more burden of managing and therefore machine learning and artificial intelligence approaches such as Q-learning and enhanced learning are responsible for identifying conditions that are not visible to the human eye in every component of the processing layers. Given this complexity, the system designers are always unclear about the algorithms to define in order to distinguish combinations of algorithms to classify what kind of diseases. These papers validate and empirically weigh up some of the current techniques in this area and evaluate the best melding of algorithms for the development of a very precise leaf disorder classification system to remove such ambiguity. This document also recommends further guidelines for research, which can be done to improve the performance of the system