Rice pest identification in natural scene based on lightweight residual network

Bao Wenxia; Wu Dezhao; Hu Gensheng; Liang Dong; Wang Nian; Yang Xianjun

doi:10.11975/j.issn.1002-6819.2021.16.018

Transactions of the Chinese Society of Agricultural Engineering > 2021 > 37(16): 145-152. > DOI: 10.11975/j.issn.1002-6819.2021.16.018

Bao Wenxia, Wu Dezhao, Hu Gensheng, Liang Dong, Wang Nian, Yang Xianjun. Rice pest identification in natural scene based on lightweight residual network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(16): 145-152. DOI: 10.11975/j.issn.1002-6819.2021.16.018

Citation:

PDF (2118 KB)

Rice pest identification in natural scene based on lightweight residual network

1.
National Engineering Research Center for Agro-ecological Big Data Analysis and Application, Anhui University, Hefei 230601, China
2.
Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China

More Information

Received Date: June 18, 2021
Revised Date: August 12, 2021
Published Date: August 14, 2021

Graphical Abstract

Abstract

Abstract

Abstract: Accurate identification of rice pests is of great significance for timely protection and management of rice. However, the rice pests are similar with the background in the color and texture, and the morphology of the pests varies greatly during different growth stages. Therefore, it is difficult to accurately identify the rice pests in natural scenes. In this paper, the Light Weight Residual Network (LW-ResNet) composed of feature extraction, global optimization and local optimization modules was designed to improve the ability to identify rice pests in natural scene images. Firstly, in order to reduce the influence of complex background and enhance the feature extraction and expression capabilities of the residual network, the residual block is improved to constitute the feature extraction module. The improved residual block increases the number of convolutional layers and branches of the original residual block, which can effectively extract the deep global features of rice pest images. Secondly, the deep global features are further optimized through the convolutional layers in the global optimization module. Finally, in order to obtain the local discriminative characteristics of rice pest images to distinguish the morphological differences between similar pests, the lightweight attention sub-module constitutes the local optimization module. The light weight attention sub-module uses depth separable convolution to reduce the redundancy of channel features and realize the aggregation of different channel characteristics, so it can highlight the local key features of rice pests. Because the improvement of the residual block in the feature extraction module reduces the number of residual blocks, and the use of deep separable convolution in the attention sub-module and the channel-based global average pooling and global maximum pooling encoding operations reduce floating point operations, the LW-ResNet network has achieved lighter weight. In the HSV space, Gamma transform is used to preprocess the v component of rice pest images and then proceed to the data expansion. After the expansion, there are 4 380 images in the training set and 492 images in the test set. In order to verify the rationality and effectiveness of the method in this paper, in the training phase, the cosine learning rate decay strategy was used to train the network model. By analyzing the number of the improved residual blocks in the feature extraction module, the lightweight attention sub-module in the local optimization module, and the global optimization module, the rationality of the method in this paper was verified. In the testing phase, the LW-ResNet network model achieves a identification accuracy of 92.5% on the test data set of 13 types of rice pest images. The identification accuracy of the LW-ResNet network model is higher than that of classic convolutional neural network models such as VGG16, ResNet, and AlexNet. The parameter amount of the LW-ResNet model is 1.62×106, and the amount of floating-point operations is 0.34×109. The number of parameters and floating-point operations of the LW-ResNet model are both lower than those of MobileNetV3, which verified the effectiveness of the method in this paper. The LW-ResNet network model has achieved light weight and a good identification effect, so it can be used for rice pest identification on the mobile terminal.
- crops,
- models,
- image identification,
- rice pest,
- attention mechanism,
- deep separable convolution

FullText(HTML)

References (31)

References

[1]	Bao W X, Qiu X, Liang D, et al. Recognition insect images at the order level based on elliptic metric learning[J]. American Society of Agricultural and Biological Engineers, 2021, 37(1): 163-170.
[2]	鲍文霞，邱翔，胡根生，等. 基于椭圆型度量学习空间变换的水稻虫害识别[J]. 华南理工大学学报，2020，48(10)：136-144.Bao Wenxia, Qiu Xiang, Hu Gensheng, et al. Rice pest identification based on elliptic metric learning space transformation[J]. Journal of South China University of Technology, 2020, 48(10): 136-144. (in Chinese with English abstract)
[3]	Jensen H G, Jacobsen L B, Pedersen S M, et al. Socioecomomic impact of widespread adoption of precision farming and controlled traffic in Denmark[J]. Precision Agriculture, 2012, 13(6): 661-677.
[4]	Lim S, Kim S, Park S, et al. Development of Application for Forest Insect Classification using CNN[C]. //2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV). Singapore: IEEE, 2018.
[5]	边柯橙，杨海军，路永华，等. 深度学习在农业病虫害检测识别中的应用综述[J]. 软件导刊. 2021，20(3)：26-33.Bian Kecheng, Yang Haijun, Lu Yonghua, et al. Overview of the application of deep learning in the detection and identification of pests and diseases in agriculture[J]. Software Guide. 2021, 20(3): 26-33. (in Chinese with English abstract)
[6]	田杰，韩冬，胡秋霞，等. 基于PCA和高斯混合模型的小麦病害彩色图像分割[J]. 农业机械学报，2014，45(7)：267-271.Tian Jie, Han Dong, Hu Qiuxia, et al. Color image segmentation of wheat diseases based on PCA and Gaussian mixture model[J]. Transactions of the Chinese Society for Agricultural Machinery, 2014, 45(7): 267-271. (in Chinese with English abstract)
[7]	Liu T, Chen W, Wu W, et al. Detection of aphids in wheat fields using a computer vision technique[J]. Biosystems Engineering, 2016, 141(2): 82-93.
[8]	Hafiz G A U, Qaisar A, Fatima G. Insect classfication using image processing and bayesian network[J]. Journal of Entomology and Zoology Studies, 2017, 5(6): 1079-1082.
[9]	Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks[J]. Advances in neural information processing systems, 2012, 25(2): 84-90.
[10]	Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. Computer Science, 2014, 18(2): 1409-1556.
[11]	He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]. // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE Computer Society, 2016.
[12]	Lecun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521(1): 436-444.
[13]	鲍文霞，孙庆，胡根生，等. 基于多路卷积神经网络的大田小麦赤霉病图像识别[J]. 农业工程学报，2020，36(11)：174-181.Bao Wenxia, Sun Qing, Hu Gensheng, et al. Field wheat head blight image recognition based on multi-channel convolutional neural network[J]. Transactions of the Chinese Society of Agricultural Engineering (Transaction of the CSAE), 2020, 36(11): 174-181. (in Chinese with English abstract)
[14]	孔建磊，金学波，陶治，等. 基于多流高斯概率融合网络的病虫害细粒度识别[J]. 农业工程学报，2020，36(13)：148-157.Kong Jianlei, Jin Xuebo, Tao Zhi, et al. Fine-grained recognition of pests and diseases based on multi-stream Gaussian probability fusion network[J]. Transactions of the Chinese Society of Agricultural Engineering (Transaction of the CSAE), 2020, 36(13): 148-157. (in Chinese with English abstract)
[15]	Li Y, Wang H, Dang L M, et al. Crop pest recognition in natural scenes using convolutional neural networks[J]. Computers and Electronics in Agriculture, 2020, 169(1): 302-310.
[16]	Liu Z, Gao J, Yang G, et al. Localization and classification of paddy field pests using a saliency map and deep convolutional neural network[J]. Scientific Reports(Sci Rep), 2016, 6(1): 1-9.
[17]	Nanni L, Maguolo G, Pancino F. Insect pest image detection and recognition based on bio-inspired methods[J]. Ecological Informatics, 2020, 57(1): 2-14.
[18]	Jie H, Li S, Gang S, et al. Squeeze-and-Excitation Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 42(8): 2011-2023.
[19]	Woo S, Park J, Lee J Y, et al. CBAM: Convolutional block attention module[C]. European Conference on Computer Vision, Munich, Germany: Springer, 2018.
[20]	Wang X, Girshick R, Gupta A, et al. Non-local neural networks[J]. Computer Science, 2017, 1711(2): 7794-7803.
[21]	Zeng W, Li M. Crop leaf disease recognition based on Self-Attention convolutional neural network[J]. Computers and Electronics in Agriculture, 2020, 172(1): 4-14.
[22]	王美华，吴振鑫，周祖光. 基于注意力改进CBAM的农作物病虫害细粒度识别研究[J]. 农业机械学报，2021，52(4)：239-247. Wang Meihua, Wu Zhenxin, Zhou Zuguang. Research on fine-grained recognition of agricultural diseases and insect pests based on improved CBAM of attention[J]. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(4): 239-247. (in Chinese with English abstract)
[23]	Gonzalez, R C, Woods. Digital image processing[J]. Beijing: Electronic Industry Press, 2005: 175-207.
[24]	陆涛. 基于统计特征分类耦合自适应Gamma校正的图像增强算法[J]. 电子测量与仪器学报，2020，34(6)：154-162. Lu Tao. Image enhancement algorithm based on statistical feature classification coupled with adaptive Gamma correction[J]. Journal of Electronic Measurement and Instrument, 2020, 34(6): 154-162. (in Chinese with English abstract)
[25]	Ren F, Liu W, Wu G. Feature reuse residual networks for insect pest recognition[J]. IEEE Access, 2019, 7(1): 122758-122768.
[26]	Loshchilov I, Hutter F. SGDR: Stochastic gradient descent with warm restarts[C]. Toulon, France: OpenReview.net, 2016.
[27]	周志华. 机器学习[M]，北京：清华大学出版社，2016：230-245.
[28]	Saini R, Jha N K, Das B, et al. ULSAM: Ultra-lightweight subspace attention module for compact convolutional neural networks[C]. IEEE Winter Conference on Applications of Computer Vision (WACV). Snowmass. CO. USA, IEEE, 2020, 12(4): 1616-1625.
[29]	Selvaraju R R, Cogswell M, Das A, et al. Grad-CAM: Visual Explanations from deep networks via gradient-based localization[J]. International Journal of Computer Vision, 2020, 128(2): 336-359.
[30]	Park J, Woo S, Lee J Y, et al. BAM: Bottleneck attention module[C]. Newcastle, UK:British Machine Vision Conference (BMVA). 2018.
[31]	Howard A, Sandler M, Chen B, et al. Searching for MobileNetV3[C]. Seoul, Korea (South): IEEE. 2019.