Research on the Algorithm of YOLO Visual Image Classification System for Pests and Diseases

Abstract

During the internship, it was found that artificial intelligence algorithms have problems such as low accuracy and low efficiency in the visual graphic classification of agricultural pests and diseases. These issues not only affect the healthy development of the industry but also make it difficult to meet the development needs of large-scale and precise agriculture. To solve these practical problems, this research topic was proposed in combination with the professional knowledge learned ——the Algorithm of YOLO Visual Image Classification System for Pests and Diseases. This article conducts a comprehensive review of the YOLO series algorithms using the literature review method. Firstly, conduct a comprehensive review of the current research status of the YOLO series of algorithms and the visual image classification of pests and diseases, and the research objectives were clarified(Conduct an in-depth analysis of the technical features and performance, Summarize the existing problems and challenges, Summarize research trends and development directions, Build a systematic review framework).Then, the basic principles, network structures and improved versions (such as YOLOv3, YOLOv4, etc.) of the YOLO series algorithms are discussed. The advantages and disadvantages of different versions of the algorithms and their performance in image classification tasks are compared. The shortcomings and areas for improvement of the existing research are summarized.