Crop variety identification method, device and equipment and storage medium

• Main Authors: WANG WEI, ZHU ZHILING, JIANG XIAOYU, GUO XINGCHENG, FAN QIQIANG
• Format: Patent
• Language: Chinese; English
• Published: 09.06.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a crop variety identification method, device and equipment and a storage medium, and the method comprises the steps: inputting a feature map obtained after a leaf image passes through a multilayer convolutional neural network into an FPN network, and generating a multi-scale feature map; using a sliding window to detect whether a local area of each feature map under the current window contains discrimination information, and generating a plurality of candidate areas; carrying out discriminative sorting on the candidate regions by using KL divergence, and selecting k candidate regions with Top-k discriminations; and carrying out feature splicing on the global feature of the leaf image and the k discriminative candidate features by using a full connection layer neural network, training and outputting to obtain a variety classification result. According to the method, the FPN is used for detecting the multi-scale local features of the leaf image, the RPN is used for generating the plurali