Review of Classification and Detection for Insects/Pests Using Machine Learning and Deep Learning Approach

Farmers usually have to deal with insects/pests and the diseases they cause. The diseases caused by these pests create several health issues and crops get severely damaged. As a result, the country’s economy is at risk. Thus, developing an intelligent system that will help farmers identify insects/pests has become necessary. The aim is to present efforts taken for pest classification and pest detection. This paper aims to find missing links in the existing work done for the classification and detection of pests and conduct an analysis of the same. The survey paper will assist to understand various machine learning models built for pest classification and deep-learning models used for pest detection. The paper will also highlight various advanced techniques used for pest localization in an image with multiple pests of multiple classes. This paper analyzes the performance of classification methods and detection methods based on parameters such as accuracy, F1 score, precision, and recall. A comprehensive survey, observations drawn from the existing research, and its analysis are presented in this paper. The survey will lead directions for future research.