关键词: 桃树缺磷症, 改进Faster RCNN, RPN, SoftNMS, ResNeXt101

Abstract: The initial symptoms of peach phosphorus deficiency (PPD) are not obvious, and the symptoms differ greatly in different stages. However, the existing peach disease identification model based on computer vision has low identification accuracy and poor generalization of different varieties. Therefore, the Faster RCNN (Faster Region based Convolutional Neural Network) model is improved. Firstly, the RS (Rank & Sort)Loss function is used to replace the crossentropy function in the Region Proposal Network (RPN). Secondly, the SoftNMS (NonMaximum Suppression) algorithm is used to replace the original NMS algorithm. Finally, ResNeXt101 network is used to replace the original feature extraction network to improve the accuracy and generalization of PPD recognition, and the detection test is carried out on the selfbuilt PPD data set. The experimental results show that the improved Faster RCNN network has an average detection accuracy of 92.28%, a recall rate of 92.31% and a recognition accuracy of 92.28% on the selfbuilt PPD data set, which meets the practical application requirements.

Key words: peach phosphorus deficiency, improved Faster RCNN, RPN, SoftNMS, ResNeXt101