基于二值化的Faster R-CNN柑橘病虫害识别研究

doi:10.13733/j.jcam.issn.20955553.2022.06.020

摘要/Abstract

摘要： 为研究在自然场景下柑橘叶片病害检测和识别技术，提出一种基于二值化的Faster R-CNN（Binary Faster R-CNN）区域检测神经网络模型。改进模型将原始的Faster R-CNN全连接层神经网络转变为二进制全卷积神经网络。试验结果表明，该模型对柑橘的黑斑病、溃疡病、黄龙病、疮痂病、健康叶片的平均准确率分别为87.2%、87.6%、89.8%、86.4%和86.6%，总平均准确率为87.5%；模型识别时间相较于Faster R-CNN网络提高0.53 s，每幅图像的检测时间为0.31 s，模型大小缩小到15.3 MB，FLOPs为2.58×109；同时在保证模型检测有效性的情况下可快速收敛。该方法对复杂自然环境下的柑橘叶片病害检测具有较好的识别速度和鲁棒性，对柑橘类疾病预防有重要的研究意义。

关键词: Faster R-CNN, Binary Faster R-CNN, 二进制网络, 目标检测, 柑橘, 病虫害

Abstract: A binarizationbased Faster R-CNN （Binary Faster R-CNN） region detection neural network model is proposed for the study of citrus leaf disease detection and recognition techniques in natural scenarios. The improved model transforms the original Faster R-CNN fully connected layer neural network into a binary, fully convolutional neural network. The experimental results showed that the average recognition rate of the model was 87.2%, 87.6%, 89.8%, 86.4%, and 86.6% for black spot, ulcer, citrus_greening, scab, and healthy leaves of citrus, respectively, and the overall average recognition rate of the model was 87.5%. The recognition speed of the model was improved by 0.53 s compared with the Faster R-CNN network, and the detection time up to 0.31 s per image. The model size was reduced to 153 MB, and the floatingpoint computational power was 2.58×109, while the model converged quickly, ensuring the effectiveness of detection and robustness of the model. The method has good recognition speed and robustness for citrus leaf disease detection in complex natural environments and is of great research significance for citrus disease prevention.

Key words: Faster R-CNN, Binary Faster R-CNN, binary network, object detection, citrus, diseases and insect pests

中图分类号:

TP391.4

宋中山, 汪进, 郑禄, 帖军, 朱祖桐, . 基于二值化的Faster R-CNN柑橘病虫害识别研究[J]. 中国农机化学报, 2022, 43(6): 150-158.

Song Zhongshan, Wang Jin, Zheng Lu, Tie Jun, Zhu Zutong.. Research on citrus pest identification based on Binary Faster R-CNN[J]. Journal of Chinese Agricultural Mechanization, 2022, 43(6): 150-158.

参考文献

［1］苏鸿, 温国泉, 谢玮, 等. 基于区域卷积神经网络模型的广西柑橘病虫害识别方法研究［J］. 西南农业学报, 2020, 33（4）: 805-810.
Su Hong, Wen Guoquan, Xie Wei, et al. Research on citrus pest and disease recognition method in Guangxi based on regional convolutional neural network model ［J］. Southwest China Journal of Agricultural Sciences, 2020, 33（4）: 805-810.
［2］王建涛, 吴叶兰, 廖禺, 等. 基于卷积神经网络的柑橘病叶高光谱分类［J］. 信息技术与信息化, 2020（3）: 84-87.
［3］张敏, 刘杰. 基于卷积神经网络的柑橘溃疡病识别方法［J］. 计算机应用, 2018, 38（Z1）: 48-52, 76.
Zhang Min, Liu Jie. Recognition method of citrus canker disease based on convolution neural network ［J］. Journal of Computer Applications, 2018, 38（Z1）: 48-52, 76.
［4］ Sharif M, Khan M A, Iqbal Z, et al. Detection and classification of citrus diseases in agriculture based on optimized weighted segmentation and feature selection ［J］. Computers and Electronics in Agriculture, 2018, 150: 220-234.
［5］任守纲, 贾馥玮, 顾兴健, 等. 反卷积引导的番茄叶部病害识别及病斑分割模型［J］. 农业工程学报, 2020, 36（12）: 186-195.
Ren Shougang, Jia Fuwei, Gu Xingjian, et al. Recognition and segmentation model of tomato leaf diseases based on deconvolutionguiding ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36（12）: 186-195.
［6］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 diseases and pests based on multistream Gaussian probability fusion network ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36（13）: 148-157.
［7］张善文, 王振, 王祖良. 多尺度融合卷积神经网络的黄瓜病害叶片图像分割方法［J］. 农业工程学报, 2020, 36（16）: 149-157.
Zhang Shanwen, Wang Zhen, Wang Zuliang. Method for image segmentation of cucumber disease leaves based on multiscale fusion convolutional neural networks ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36（16）: 149-157.
［8］李就好, 林乐坚, 田凯, 等. 改进Faster R-CNN的田间苦瓜叶部病害检测［J］. 农业工程学报, 2020, 36（12）: 179-185.
Li Jiuhao, Lin Lejian, Tian Kai, et al. Detection of leaf diseases of balsam pear in the field based on improved Faster R-CNN ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36（12）: 179-185.
［9］王昱潭, 朱超伟, 赵琛, 等. 基于Faster R-CNN的灵武长枣图像检测方法［J］. 计算机工程与应用, 2021, 57（4）: 216-224.
Wang Yutan, Zhu Chaowei, Zhao Chen, et al. Image detection method of Lingwu long jujube based on Faster R-CNN ［J］. Computer Engineering and Applications, 2021, 57（4）: 216-224.
［10］ He K, Zhang X, Ren S, et al. Identity mappings in deep residual networks［C］. European conference on computer vision. Springer, Chan, 2016: 630-645.
［11］ Pan S J. Qiang Y. A survey on transfer learning ［J］. IEEE Transactions on Knowledge and Data Engineering, 2010, 22（10）: 1345-1359.
［12］ Behroozikhazaei N, Malekl M R. A robust algorithm based on color features for grape cluster segmentation ［J］. Computers and Electronics in Agriculture, 2017, 42: 41-49.
［13］ Wang D, Song H, He D. Research advance on vision system of apple picking robot ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33（10）: 59-69.
［14］ Swietojanskl P, Ghoshal A, Rnals S. Convolutional neural networks for distant speech recognition ［J］. IEEE Signal Processing Letters, 2014（9）: 1120-1124.
［15］ Rumpf T, Mahlein, A K, Steiner V, et al. Early detection and classification of plant diseases with Support Vector Machines based on hyperspectral reflectance［J］. Computers and Electronics in Agriculture, 2010, 74（1）: 91-99.
［16］ Xu Y, Imou K, Kaizu Y, et al. Twostage approach for detecting slightly overlapping strawberries using HOG descriptor ［J］. Biosystems Engineering, 2013, 115（2）: 144-153.
［17］ Zhao C, Lee W S, He D. Immature green citrus detection based on colour feature and sum of absolute transformed difference （SATD） using colour images in the citrus grove ［J］. Computers & Electronics in Agriculture, 2016, 124: 243-253.
［18］ Wu X, DSahoo, Hoi S. Recent advances in deep learning for object detection ［J］. Neurocomputing, 2020, 396: 39-64.
［19］ Zhang Q, Zhang R M, Chen B. Research review of image recognition technology based on deep learning ［J］. Journal of the Hebei Academy of Sciences, 2019, 36（3）: 28-36.
［20］ Goodridge W, Bernard M, Jordan R, et al. Intelligent diagnosis of diseases in plants using a hybrid MultiCriteria decision making technique ［J］. Computers and Electronics in Agriculture, 2017, 133（6）: 80-87.
［21］ Xiang R. Image segmentation for whole tomato plant recognition at night ［J］. Computers and Electronics in Agriculture, 2018, 154: 434-442.
［22］张星, 高巧明, 潘栋, 等. 基于改进YOLOv3的田间复杂环境下菠萝拾捡识别研究［J］. 中国农机化学报, 2021, 42（1）: 201-206.
Zhang Xing, Gao Qiaoming, Pan Dong, et al. Picking recognition research of pineapple in complex field environment based on improved YOLOv3 ［J］. Journal of Chinese Agricultural Mechanization, 2021, 42（1）: 201-206.
［23］秦立峰, 张熹, 张晓茜. 基于高光谱病害特征提取的温室黄瓜霜霉病早期检测［J］. 农业机械学报, 2020, 51（11）: 212-220.
Qin Lifeng, Zhang Xi, Zhang Xiaoqian. Early detection of cucumber downy mildew in greenhouse by hyperspectral disease differential feature extraction ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51（11）: 212-220.
［24］景亮, 王瑞. 基于双目相机与改进YOLOv3算法的果园行人检测与定位［J］. 农业机械学报, 2020, 51（9）: 34-39, 25.
Jing Liang, Wang Rui. Orchard pedestrian detection and location based on binocular camera and improved YOLOv3 algorithm ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51（9）: 34-39, 25.
［25］袁洪波, 赵努东, 程曼. 基于图像处理的田间杂草识别研究进展与展望［J］. 农业机械学报, 2020, 51（S2）: 323-334.
Yuan Hongbo, Zhao Nudong, Cheng Man. Review of weeds recognition base on image processing ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51（S2）: 323-334.
［26］姚青, 谷嘉乐, 吕军, 等. 改进RetinaNet的水稻冠层害虫为害状自动检测模型［J］. 农业工程学报, 2020, 36（15）: 182-188.
Yao Qing, Gu Jiale, Lü Jun, et al. Automatic detection model for pest damage symptoms on rice canopy based on improved RetinaNet ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36（15）: 182-188.
［27］刘翱宇, 吴云志, 朱小宁, 等. 基于深度残差网络的玉米病害识别［J］. 江苏农业学报, 2021, 37（1）: 67-74.
Liu Aoyu, Wu Yunzhi, Zhu Xiaoning, et al. Corn disease recognition base on deep residual network ［J］. Jiangsu Journal of Agricultural Sciences, 2021, 37（1）: 67-74.

[1]	陈品岚, 张小花, 朱立学, 李浩林. 基于树莓派及深度学习的柑橘识别系统设计[J]. 中国农机化学报, 2022, 43(9): 131-136.
[2]	齐亚聪, 杨会民, 陈毅飞, 周欣, 马艳, 王学农, . 变量喷雾靶标探测技术研究现状与展望[J]. 中国农机化学报, 2022, 43(6): 83-89.
[3]	史方青, 王虎林, 黄华. 基于卷积神经网络的马铃薯芽眼检测识别研究[J]. 中国农机化学报, 2022, 43(6): 159-165.
[4]	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.
[5]	何颖, 陈丁号, 彭琳. 基于改进YOLOv5模型的经济林木虫害目标检测算法研究[J]. 中国农机化学报, 2022, 43(4): 106-115.
[6]	王菁, 范晓飞, 赵智慧, 张君, 孙磊, 索雪松. 基于YOLO算法的不同品种枣自然环境下成熟度识别[J]. 中国农机化学报, 2022, 43(11): 165-171.
[7]	尚文卿, 齐红波. 基于改进Faster R-CNN与迁移学习的农田杂草识别算法[J]. 中国农机化学报, 2022, 43(10): 176-182.
[8]	赵越, 赵辉, 姜永成, 任东悦, 李阳, 卫勇. 基于深度学习的马铃薯叶片病害检测方法#br# #br# [J]. 中国农机化学报, 2022, 43(10): 183-189.
[9]	刘冬梅, 杨杭旭, 周宏平, 郑加强, 茹煜, . 茶树植保机械及减量施药技术研究进展[J]. 中国农机化学报, 2021, 42(9): 59-67.
[10]	李小敏, 张日红, 陈天赐, 侯炳法, 张权, 李雄, . 基于深度学习的林下落果识别方法与试验[J]. 中国农机化学报, 2021, 42(9): 202-208.
[11]	李子茂, 刘恋冬, 夏梦, 帖军, 张玥, . 基于深度学习的月季多叶片病虫害检测研究[J]. 中国农机化学报, 2021, 42(8): 169-176.
[12]	项新建;周跃琪;姚佳娜;宋晓敏;郑永平;. 野外农用视频监控运动目标检测算法研究及系统开发[J]. 中国农机化学报, 2021, 42(5): 166-174.
[13]	成伟;张文爱;冯青春;张万豪;. 基于改进YOLOv3的温室番茄果实识别估产方法[J]. 中国农机化学报, 2021, 42(4): 176-182.
[14]	闫建伟;张乐伟;赵源;张富贵;. 改进RetinaNet的刺梨果实图像识别[J]. 中国农机化学报, 2021, 42(3): 78-83.
[15]	孙瑞琳;孙全;孙成明;刘涛;李冬双;吴峰峰;. 基于不同平台的小麦病虫害遥感监测研究进展[J]. 中国农机化学报, 2021, 42(3): 142-150.