基于轻量化VGG的植物病虫害识别

doi:10.13733/j.jcam.issn.20955553.2022.04.005

摘要/Abstract

摘要： 深度神经网络模型被广泛应用在植物病虫害识别任务中，并取得巨大成功。同时，这些网络的计算复杂度和参数量也在不断增加，这将对神经网络的部署提出重大挑战，尤其是在硬件资源有限的设备或实时应用上。针对该问题，提出一种轻量化的病虫害识别模型，结合Ghost模块对VGG16进行改进，同时减少模型卷积层的卷积核个数，并引入Ranger优化器。试验结果表明，该模型在PlantVillage数据集上准确率为99.37%，FLOPs为88.45 M，比VGG16下降71.86%，有较快的收敛速度，在复杂环境下，模型的准确率为92.40%，识别时间为VGG16的50%。

关键词: 轻量化, 病虫害识别, Ghost模块, VGG网络, Ranger优化器

Abstract: The deep neural network model is widely used in the identification of plant diseases and insect pests and has achieved great success. At the same time, the computational complexity and parameter quantities of these networks are also increasing, which will pose a major challenge to the deployment of neural networks, especially on devices with limited hardware resources or realtime applications. Aiming to solve this problem, a lightweight pest identification model with Ranger optimizer is proposed, in which the VGG16 is improved by using the Ghost module and reducing the number of convolution kernels in the convolution layer. Experimental results show that the accuracy of the model on the PlantVillage dataset is 99.37%, and the FLOPs is 88.45 M, which is 71.86% lower than VGG16, and it has a faster convergence rate. In a complex environment, the accuracy of the model is 92.40%, and the time is 50% of VGG16.

Key words: lightweight, pests and diseases identification, Ghost module, VGG network, Ranger optimizer

中图分类号:

TP391.41

王江晴, 冀星, 莫海芳, 帖军, 刘畅, . 基于轻量化VGG的植物病虫害识别[J]. 中国农机化学报, 2022, 43(4): 25-31.

Wang Jiangqing, Ji Xing, Mo Haifang, Tie Jun, Liu Chang.. Plant disease detection based on lightweight VGG[J]. Journal of Chinese Agricultural Mechanization, 2022, 43(4): 25-31.

参考文献

［1］ Liang Q, Xiang S, Hu Y, et al. PD 2 SENet: Computerassisted plant disease diagnosis and severity estimation network ［J］. Computers and Electronics in Agriculture, 2019, 157:518-529.

［2］ Lecun Y, Bengio Y. Convolutional networks for images,speech, and time series ［M］. Cambridge: MIT Press, 1995.

［3］ Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks ［C］. NIPS. Curran Associates Inc. 2012.

［4］ Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision ［C］. Proceedings of the IEEE conference on computer vision and pattern recognition, Taipei: IEEE Press, 2016: 2818-2826.

［5］ 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. Taipei: IEEE Press, 2016: 770-778.

［6］ Durmus H, Gunes E O, Kirci M. Disease detection on the leaves of the tomato plants by using deep learning［C］. 2017 6th International Conference on AgroGeoinformatics. Fairfax: 2017: 1-5.

［7］ Iandola F N, Han S, Moskewicz M W, et al. SqueezeNet: AlexNetlevel accuracy with 50x fewer parameters and< 0.5 mb model size ［EB/OL］. https://arxiv. org/pdf/1602. 07360. pdf , 2018-11-13.

［8］ Lü M, Zhou G, He M, et al. Maize leaf disease identification based on feature enhancement and DMSrobust alexnet ［J］. IEEE Access, 2020, 8:57952-57966.

［9］王国伟, 刘嘉欣. 基于卷积神经网络的玉米病害识别方法研究［J］. 中国农机化学报, 2021. 42（2）: 139- 145.

Wang Guowei, Liu Jiaxin. Research on corn disease recognition method based on convolutional neural network ［J］. Journal of Chinese Agricultural Mechanization, 2021, 42（2）: 139-145.

［10］李鑫然, 李书琴, 刘斌. 基于改进Faster R＿CNN的苹果叶片病害检测模型［J］. 计算机工程, 2021, 47（11）: 298-304.

Li Xinran, Li Shuqin, Liu Bin. Apple leaf disease detection method based on improved Faster R_CNN ［J］. Computer Engineering, 2021, 47（11）: 298-304.

［11］ Lin T Y, Dollár P, Girshick R, et al. Feature pyramid networks for object detection ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE Press, 2017: 2117-2125.

［12］ Zhang P, Yang L, DaoLiang L, et al. EfficientNet-B4-Ranger: A novel method for greenhouse cucumber disease recognition under natural complex environment ［J］. Computers and Electronics in Agriculture, 2020, 176: 105652.

［13］ Tan M, Le Q V. Efficientnet: Rethinking model scaling for convolutional neural networks ［EB/OL］. https://arxiv. org/pdf/1905.11946.pdf, 2020-09-11.

［14］肖经纬, 田军委, 王沁, 等. 基于改进残差网络的果实病害分类方法［J］.计算机工程, 2020, 46（9）: 221-225.

Xiao Jingwei, Tian Junwei, Wang Qin, et al. Fruit disease classification method based on improved residual network ［J］. Computer Engineering, 2020, 46（9）: 221-225.

［15］ Simonyan K, Zisserman A. Very deep convolutional networks for largescale image recognition ［J］. Computer Science, 2014.

［16］ Han S, Mao H, Dally W J. Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding ［J］. Fiber, 2015, 56（4）: 3-7.

［17］蒲秀夫，宁羋，雷印杰，等. 基于二值化卷积神经网络的农业病虫害识别［J］. 中国农机化学报， 2020， 41（2）: 177-182.

Pu Xiufu, Ning Qian, Lei Yinjie, et al. Identification of agricultural plant diseases based on binarized convolutional neural network ［J］. Journal of Chinese Agricultural Mechanization, 2020, 41（2）: 177-182.

［18］ Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications ［J］. arXiv preprint arXiv:1704.04861, 2017.

［19］ Zhang X, Zhou X, Lin M, et al. Shufflenet: An extremely efficient convolutional neural network for mobile devices ［C］. Proceedings of the IEEE conference on computer vision and pattern recognition. Salt Lake City, USA: IEEE Press, 2018: 6848-6856.

［20］ Han K, Wang Y, Tian Q, et al. GhostNet: More features from cheap operations ［C］. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE Press, 2020: 1580-1589.

［21］ Wright L. New Deep Learning Optimizer, Ranger: Synergistic combination of RAdam+LookAhead for the best of both ［EB/OL］. https://github. com/lessw2020/RangerDeepLearningOptimizer, 2019-08-20.

［22］ Liu L, Jiang H, He P, et al. On the variance of the adaptive learning rate and beyond ［J］. 2019.

[1]	苏斐, , 张泽旭, 赵妍平, 李天华, , 祖林禄, . 基于轻量化YOLO⁃v3的绿熟期番茄检测方法[J]. 中国农机化学报, 2022, 43(3): 132-137.
[2]	刘泽松, 王浩屹, 李骅, 王永健, 丁元庚, 曾晓萍. 基于有限元的大蒜播种机机架轻量化设计[J]. 中国农机化学报, 2022, 43(1): 27-32.
[3]	任帅阳, 高爱民, 张勇, 韩伟, . 六旋翼植保无人机旋翼折叠机构有限元分析及拓扑优化[J]. 中国农机化学报, 2021, 42(9): 53-58+194.
[4]	杨桃月;顾苑婷;舒成松;曹阳;张大斌;. 烟秆拔秆破碎机机架轻量化设计与随机振动疲劳分析[J]. 中国农机化学报, 2021, 42(5): 100-106.
[5]	冯叶陶;廖敏;张小军;张涛;潘群林;. 丘陵山地拖拉机前驱动桥壳组件轻量化设计与试验研究[J]. 中国农机化学报, 2021, 42(5): 107-113.
[6]	陆雅诺;陈炳才;. 基于注意力机制的小样本啤酒花病虫害识别[J]. 中国农机化学报, 2021, 42(3): 189-196.
[7]	刘萍;李玉龙;臧勇;李秀荣;. 基于最大形状系数的齿轮泵轻量化研究[J]. 中国农机化学报, 2020, 41(8): 127-131+147.