Classification of Amanita Species Based on Bilinear Networks with Attention Mechanism

The accurate classification of Amanita is helpful to its research on biological control and medical value, it can also prevent mushroom poisoning incidents. In this paper, we constructed the Bilinear convolutional neural networks (B-CNN) with attention mechanism model based transfer learning realize Amanita. When trained, weight ImageNet used for pre-training, Adam optimizer update network parameters. test process, images at different growth stages were further generalization ability model. After comparing our other models, results show that greatly reduces number parameters while achieving high accuracy (95.2%) has good ability. It an efficient model, which provides a new option in areas limited computing resources.