标题：Progress in image recognition technology for crop diseases based on deep learning
作者：Zhang L.; Wang D.; Zhao Y.; Li Y.; Zhao Y.
作者机构：[Zhang, L] School of Mechanical & Automotive Engineering, Liaocheng University, Liaocheng, Shandong 252000, China;[ Wang, D] College of Agriculture, 更多
通讯作者地址：[Wang, D] College of Agriculture, Yulin Normal UniversityChina;
来源：Revista de la Facultad de Agronomia
关键词：Deep Learning; Neural Network Model; Pest and Disease Identification
摘要：The traditional digital image processing technology has its limitations. It requires manual design features, which consumes manpower and material resources, and identifies crops with a single type and the results are bad. Therefore, to find an efficient and fast real-time disease image recognition method is very meaningful. Deep learning is a machine learning algorithm that can automatically learns representative features to achieve better results in areas of image recognition. Therefore, the purpose of this paper is to use deep learning methods to identify crop pests and diseases, and to find efficient and fast real-time image recognition methods of disease. This paper analyzes the classical and the latest neural network structure based on the theory of deep learning. For this problem that the network based on natural images classification is not suitable for crop pest and disease identification tasks, this paper has improved the network structure that can take care of both recognition speed and recognition accuracy. Discussed the the influence of crop pest and disease feature extraction layer on recognition performance. Finally, used the inner layer as the main structure to be pest and disease feature extraction layer by comparing the advantages and disadvantages of inner and global average pooling layers. Analyze various loss functions such as Softmax Loss, Center Loss and Angular Softmax Loss for pest identification. In view of the shortcomings of difficulty in loss function training,convergence and operation, By making the distance between pests and diseases smaller and the distance between classes more greater, improved the loss function, and introduced techniques such as feature normalization and weight normalization. The experimental results show that the method can effectively enhance the characteristic expression ability of pests and diseases and thus improve the recognition rate of pests and diseases. Moreover, the method makes the pest identification network training simpler and can improve the pest and disease recognition rate better. © 2020, Universidad del Zulia. All rights reserved.