MSFResNet: A ResNeXt50 model based on multi-scale feature fusion for wild mushroom identification

Abstract

To solve the problems of redundant feature information, the insignificant difference in feature representation, and low recognition accuracy of the fine-grained image, based on the ResNeXt50 model, an MSFResNet network model is proposed by fusing multi-scale feature information. Firstly, a multi-scale feature extraction module is designed to obtain multi-scale information on feature images by using different scales of convolution kernels. Meanwhile, the channel attention mechanism is used to increase the global information acquisition of the network. Secondly, the feature images processed by the multi-scale feature extraction module are fused with the deep feature images through short links to guide the full learning of the network, thus reducing the loss of texture details of the deep network feature images, and improving network generalization ability and recognition accuracy. Finally, the validity of the MSFResNet model is verified using public datasets and applied to wild mushroom identification. Experimental results show that compared with ResNeXt50 network model, the accuracy of the MSFResNet model is improved by 6.01% on the FGVC-Aircraft common dataset. It achieves 99.13% classification accuracy on the wild mushroom dataset, which is 0.47% higher than ResNeXt50. Furthermore, the experimental results of the thermal map show that the MSFResNet model significantly reduces the interference of background information, making the network focus on the location of the main body of wild mushroom, which can effectively improve the accuracy of wild mushroom identification.

Keywords

multi-scale feature fusion / attention mechanism / ResNeXt50 / wild mushroom identification / deep learning

References

[1] WAN R, LIU Z T, WAN Q Q, et al. Analysis on wild mushroom poisoning in Yunnan province from 2011 to 2017. Soft Science of Health, 2019, 33(10): 84-86.

[2] ZHANG L C, MA R, ZHANG Y X. Application of improved LeNet-5 model in apple image recognition. Computer Engineering and Design, 2018, 39(11): 3570-3575.

[3] ZHOU Z Y, HE D J, ZHANG H H, et al. Non-destructive detection of moldy core in apple fruit based on deep belief network. Food Science, 2017, 38(14): 297-303.

[4] WANG J, LIU X H. Pathological recognition of apple leaves based on deeply separable convolution. Computer Systems & Applications, 2020, 29(11): 190-195.

[5] LU Z D, ZHANG C D, ZHANG J Q, et al. Identification of apple leaf disease based on dual branch network. Computer Science and Exploration, 2022, 16(4): 917-926.

[6] LIU B, DING Z F, TIAN L L, et al. Grape leaf disease identification using improved deep convolutional neural networks. Frontiers in Plant Science, 2020, 11: 1082.

[7] QIU J Y, LUO J, LI X, et al. Multi-scale grape image recognition method based on convolutional neural network. Journal of Computer Applications, 2019, 39(10): 2930-2936.

[8] FENG X, LI D D, WANG W J, et al. Image recognition of wheat leaf diseases based on lightweight convolutional neural network and transfer learning. Journal of Henan Agricultural Sciences, 2021, 50(4): 174-180.

[9] XU G Z, WANG Z M, LI C Q. Recognition model of tomato leaf diseases based on deep learning. Microprocessors, 2020, 41(3): 30-36.

[10] SHEN R L, HUANG Y L, WEN X, et al. Mushroom classification based on Xception and ResNet50 models. Journal of Heihe University, 2020, 11(7): 181-184.

[11] XIAO J W, ZHAO C B, LI X J, et al. Research on mushroom image classification based on deep learning. Software Engineering, 2020, 23(7): 21-26.

[12] CHEN D G, AZRAGUL, YIN P B, et al. Research on identification of wild fungus species based on improved Xception migration learning. Progress in Laser and Optoelectronics, 2021,58(8): 245-254.

[13] YUAN P S, SHEN C J, XU H L. Fine-grained mushroom phenotype recognition based on transfer learning and bilinear CNN. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(7): 151-158.

[14] XIE S N, GIRSHICK R, DOLLÁR P, et al. Aggregated residual transformations for deep neural networks//2017 IEEE Conference on Computer Vision and Pattern Recognition, July 21-26, 2017, Honolulu, HI, USA. New York: IEEE, 2017: 5987-5995.

[15] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition//2016 IEEE Conference on Computer Vision and Pattern Recognition, June 27-30, 2016, Las Vegas, NV, USA. New York: IEEE, 2016: 770-778.

[16] HU J, SHEN L, SUN G. Squeeze-and-excitation networks//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 18-23, 2018, Salt Lake City, UT, USA. New York: IEEE, 2018: 7132-7141.

[17] MAJI S, RAHTU E, KANNALA J, et al. Fine-grained visual classification of aircraft. ArXiv e-prints, 2013: arXiv:1306.5151.

[18] POWERS D M W. Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation. 2020: 2010.16061.

[19] YU Y, YANG T T, YANG B X. Confusion matrix classification performance evaluation and Python implementation. Modern Computer, 2021, 27(20): 70-73.

[20] THARWAT A. Classification assessment methods. Applied Computing and Informatics, 2021, 17(1): 168-192.

[21] SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM: visual explanations from deep networks via gradient-based localization. International Journal of Computer Vision, 2020, 128(2): 336-359.