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

[1]  Ole Winther,et al.  Recurrent Spatial Transformer Networks , 2015, ArXiv.

[2]  T. Poggio,et al.  What and where: A Bayesian inference theory of attention , 2010, Vision Research.

[3]  Daoliang Li,et al.  EfficientNet-B4-Ranger: A novel method for greenhouse cucumber disease recognition under natural complex environment , 2020, Comput. Electron. Agric..

[4]  G. Humphreys,et al.  Attentional control and the self: The Self-Attention Network (SAN) , 2016, Cognitive neuroscience.

[5]  Quoc V. Le,et al.  EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks , 2019, ICML.

[6]  Davide Di Ruscio,et al.  Automated fruit recognition using EfficientNet and MixNet , 2020, Comput. Electron. Agric..

[7]  F. Carvalho,et al.  Amanita phalloides poisoning: Mechanisms of toxicity and treatment. , 2015, Food and chemical toxicology : an international journal published for the British Industrial Biological Research Association.

[8]  Ryosuke Shibasaki,et al.  Distinction of Edible and Inedible Harvests Using a Fine-Tuning-Based Deep Learning System , 2019 .

[9]  Dong Soo Lee,et al.  Moth Detection from Pheromone Trap Images Using Deep Learning Object Detectors , 2020, Agriculture.

[10]  Ashnil Kumar,et al.  Multimodal Spatial Attention Module for Targeting Multimodal PET-CT Lung Tumor Segmentation , 2020, IEEE Journal of Biomedical and Health Informatics.

[11]  A. Nayır,et al.  Haemoperfusion in Amanita phalloides poisoning. , 1995, Journal of tropical pediatrics.

[12]  T. Bau,et al.  [Determination of main peptide toxins from Amanita pallidorosea with HPLC and their antifungal action on Blastomyces albicans]. , 2011, Wei sheng wu xue bao = Acta microbiologica Sinica.

[13]  W. Chilton,et al.  Toxic metabolites of Amanita pantherina, A. cothurnata, A. muscaria and other Amanita species. , 1976, Lloydia.

[14]  Andrew Zisserman,et al.  Spatial Transformer Networks , 2015, NIPS.

[15]  D. Michelot,et al.  Amanita muscaria: chemistry, biology, toxicology, and ethnomycology. , 2003, Mycological research.

[16]  Ning Ye,et al.  TA-CNN: Two-way attention models in deep convolutional neural network for plant recognition , 2019, Neurocomputing.

[17]  C. Bas.  Morphology and subdivision of Amanita and a monograph of its section Lepidella , 1969 .

[18]  X. Yang.  An Overview of the Attention Mechanisms in Computer Vision , 2020, Journal of Physics: Conference Series.

[19]  Jun Sun,et al.  Detection of Key Organs in Tomato Based on Deep Migration Learning in a Complex Background , 2018, Agriculture.

[20]  H. Faulstich.  New aspects of Amanita poisoning , 1979, Klinische Wochenschrift.

[21]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[22]  R. Busuttil,et al.  Amanita poisoning: treatment and the role of liver transplantation. , 1989, The American journal of medicine.

[23]  Peng Wang,et al.  Computer Vision Based Automatic Recognition of Pointer Instruments: Data Set Optimization and Reading , 2021, Entropy.

[24]  Sashank J. Reddi,et al.  Why ADAM Beats SGD for Attention Models , 2019, ArXiv.

[25]  D. Barałkiewicz,et al.  Leaching of arsenic and sixteen metallic elements from Amanita fulva mushrooms after food processing , 2017 .

[26]  Jae Youn Hwang,et al.  Channel Attention Module With Multiscale Grid Average Pooling for Breast Cancer Segmentation in an Ultrasound Image , 2020, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control.