基于迁移学习的棉花识别

doi:10.3969/j.issn.1004-1524.2020.08.16

浙江农业学报 ›› 2020, Vol. 32 ›› Issue (8): 1457-1465.DOI: 10.3969/j.issn.1004-1524.2020.08.16

基于迁移学习的棉花识别

王见^1,2, 田光宝^1,2, 周勤^1,2

1.重庆大学机械传动国家重点实验室,重庆 400044;
2.重庆大学机械工程学院,重庆 400044

收稿日期:2020-01-16 出版日期:2020-08-25 发布日期:2020-08-28
作者简介:王见(1975—),男,新疆喀什人,博士,副教授,主要从事机电一体化技术与智能控制研究。E-mail:vi@cqu.edu.cn
基金资助:
重庆市创新重大主题专项(cstc2017rgzn-zdyfX0007); 重庆科技计划(cstc2018jszx-cyztzxX0026)

Cotton recognition based on transfer learning

WANG Jian^1,2, TIAN Guangbao^1,2, ZHOU Qin^1,2

1. The State Key Laboratory of Mechanical Transmission, Chongqing University, Chongqing 400044, China;
2. College of Mechanical Engineering, Chongqing University, Chongqing 400044, China

Received:2020-01-16 Online:2020-08-25 Published:2020-08-28

摘要/Abstract

摘要： 提高智能采棉机效率的一个重要途径是实现单个、重叠和遮挡棉花的识别,避免误采摘和漏采摘。针对不同形态棉花的识别,常规的特征提取方法难以达到令人满意的结果,因而采用基于迁移学习的棉花识别方法和基于迁移模型的特征提取与极限学习机(extreme learning machine,ELM)相结合的方法进行棉花识别研究。首先更改AlexNet、GoogleNet、ResNet-50模型分类层和设置相关参数,用训练好的迁移模型对棉花验证集识别,然后利用训练好的迁移模型进行棉花数据集特征提取,再用训练集的特征训练ELM模型,统计不同隐含层神经元个数的ELM模型对棉花的识别准确率。AlexNet、GoogleNet、ResNet-50迁移模型识别率依次为92.03%、93.19%、93.68%;使用特征提取再与ELM结合的方法,准确率比对应迁移模型分别提高了1.97、1.34、1.55百分点。结果表明,迁移模型对小样本棉花识别也有较高准确率,基于特征提取与ELM相结合的方法可进一步提高准确率。

关键词: 棉花, 识别, 迁移学习, 极限学习机, 特征提取

Abstract: An important way to improve the efficiency of the intelligent cotton picker is to realize the identification of single cotton, overlapped cottons, blocked cottons to avoid false picking and missed picking. The conventional feature extraction method is difficult to achieve satisfactory results for the identification of different types of cotton. Methods based on transfer learning and feature extraction based on transfer learning combined with extreme learning machine (ELM) was proposed. Firstly, the classification layer of AlexNet, GoogleNet and ResNet-50 models was changed and the relevant parameters were set, and the trained transfer model was used to identify the cotton validation set; secondly, the trained transfer model was used to extract the characteristics of cotton data set, and then the characteristics of the training set were used to train the ELM model to calculate the accuracy of the ELM model with different hidden layer neurons. The recognition rates of AlexNet, GoogleNet and ResNet-50 transfer models were 92.03%, 93.19% and 93.68%, respectively. The accuracy of the method of feature extraction combined with ELM was 1.97, 1.34 and 1.55 percentage points higher than that of the corresponding transfer model. The results showed that the transfer model could obtain higher accuracy for small sample cotton recognition. The method based on feature extraction and ELM could further improve the accuracy.

Key words: cotton, recognition, transfer learning, extreme learning machine, feature extraction

中图分类号:

S562
S24

王见, 田光宝, 周勤. 基于迁移学习的棉花识别[J]. 浙江农业学报, 2020, 32(8): 1457-1465.

WANG Jian, TIAN Guangbao, ZHOU Qin. Cotton recognition based on transfer learning[J]. , 2020, 32(8): 1457-1465.

参考文献

[1] 张山鹰. 新疆机采棉发展现状及发展方向的思考[J]. 农业工程, 2012, 2(7): 1-6.
ZHANG S Y. Consideration about status quo and development direction of machine pick up cotton in Sinkiang Province[J]. Agricultural Engineering, 2012, 2(7): 1-6.(in Chinese with English abstract)
[2] 王献锋, 丁军, 朱义海. 基于性能改善深度信念网络的棉花病虫害预测方法[J]. 浙江农业学报, 2018, 30(10): 1790-1797.
WANG X F, DING J, ZHU Y H. A forecasting method of cotton diseases and insect pests based on deep belief network with improved performance[J]. Acta Agriculturae Zhejiangensis, 2018, 30(10): 1790-1797.(in Chinese with English abstract)
[3] 王巧华, 翁富炯, 张洪洲, 等. 机采棉中残地膜静电吸附法分级去除[J]. 农业机械学报, 2019, 50(6): 140-147.
WANG Q H, WENG F J, ZHANG H Z, et al. Grading removal of residual plastic film in machine-harvested cotton using electrostatic adsorption method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(6): 140-147.(in Chinese with English abstract)
[4] 田景山, 张煦怡, 张旺锋. 新疆近年机采棉发展过程中的棉纤维品质变化[J]. 中国棉花, 2017, 44(12): 27-31, 34.
TIAN J S, ZHANG X Y, ZHANG W F. Change of fiber quality along with the development of machine-harvested cotton in Xinjiang[J]. China Cotton, 2017, 44(12): 27-31, 34.(in Chinese with English abstract)
[5] LI C, THIBODEAUX D, KNOWLTON A R, et al. Effect of cleaning treatment and cotton cultivar on cotton fiber and textile yarn quality[J]. Applied Engineering in Agriculture, 2012, 28(6): 833-840.
[6] KRIFA M. Fiber length distribution in cotton processing: dominant features and interaction effects[J]. Textile Research Journal, 2006, 76(5): 426-435.
[7] 刘坤, 费树岷, 汪木兰, 等. 基于改进随机Hough变换的棉桃识别技术[J]. 农业机械学报, 2010, 41(8): 160-165.
LIU K, FEI S M, WANG M L, et al. Cotton recognition based on randomized hough transform[J]. Transactions of the Chinese Society for Agricultural Machinery, 2010, 41(8): 160-165.(in Chinese with English abstract)
[8] 王勇, 沈明霞, 姬长英. 基于色差信息的田间成熟棉花识别[J]. 浙江农业学报, 2007, 19(5): 385-388.
WANG Y, SHEN M X, JI C Y. Study on the recognition of mature cotton based on the chromatic aberration in natural outdoor scenes[J]. Acta Agriculturae Zhejiangensis, 2007, 19(5): 385-388.(in Chinese with English abstract)
[9] 刘坤, 费树岷, 汪木兰. 采棉机器人中棉花成熟度的识别[J]. 工业仪表与自动化装置, 2012(6): 3-5.
LIU K, FEI S M, WANG M L. Cotton recognition in the cotton picking robot[J]. Industrial Instrumentation & Automation, 2012(6): 3-5.(in Chinese with English abstract)
[10] 关胤. 基于残差网络迁移学习的花卉识别系统[J]. 计算机工程与应用, 2019, 55(1): 174-179.
GUAN Y. Flower species recognition system based on residual network transfer learning[J]. Computer Engineering and Applications, 2019, 55(1): 174-179.(in Chinese with English abstract)
[11] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks[C]//Advances in Neural Information Processing Systems. 2012: 1097-1105.
[12] SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston, MA, USA. IEEE, 2015: 1-9.
[13] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA. IEEE, 2016: 770-778.
[14] 王彦翔, 张艳, 杨成娅, 等. 基于深度学习的农作物病害图像识别技术进展[J]. 浙江农业学报, 2019, 31(4): 669-676.
WANG Y X, ZHANG Y, YANG C Y, et al. Advances in new nondestructive detection and identification techniques of crop diseases based on deep learning[J]. Acta Agriculturae Zhejiangensis, 2019, 31(4): 669-676.(in Chinese with English abstract)
[15] 曹凤莲. 基于深度学习的植物叶片识别方法研究[D]. 重庆: 重庆大学, 2017.
CAO F L. The study of plant leaf identification based on deep learning method[D]. Chongqing: Chongqing University, 2017.(in Chinese with English abstract)
[16] 海潮. 基于深度学习的红枣缺陷识别技术研究[D]. 郑州: 郑州大学, 2019.
HAI C. The research on deep learning for jujube defects recognition technology[D]. Zhengzhou: Zhengzhou University, 2019.(in Chinese with English abstract)
[17] 廉小亲,成开元,安飒,等.基于深度学习和迁移学习的水果图像分类[J].测控技术,2019,38(6):15-18.
LIAN X Q, CHENG K Y, AN S, et al. Fruit image classification based on deep learning and transfer learing[J]. Measurement & Control Technology, 2019, 38(6):15-18(in Chinese with English abstract)
[18] PAN S J, YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359.
[19] CUI Y, SONG Y, SUN C, et al. Large scale fine-grained categorization and domain-specific transfer learning[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA: IEEE, 2018: 4109-4118.
[20] 张建华, 孔繁涛, 吴建寨, 等. 基于改进VGG卷积神经网络的棉花病害识别模型[J]. 中国农业大学学报, 2018, 23(11): 161-171.
ZHANG J H, KONG F T, WU J Z, et al. Cotton disease identification model based on improved VGG convolution neural network[J]. Journal of China Agricultural University, 2018, 23(11): 161-171.(in Chinese with English abstract)
[21] 郑一力, 张露. 基于迁移学习的卷积神经网络植物叶片图像识别方法[J]. 农业机械学报, 2018, 49(增刊): 354-359.
ZHENG Y L, ZHANG L. Plant leaf image recognition method based on transfer learning with convolutional neural networks[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(Suppl.): 354-359.(in Chinese with English abstract)
[22] 王艳玲, 张宏立, 刘庆飞, 等. 基于迁移学习的番茄叶片病害图像分类[J]. 中国农业大学学报, 2019, 24(6): 124-130.
WANG Y L, ZHANG H L, LIU Q F, et al. Image classification of tomato leaf diseases based on transfer learning[J]. Journal of China Agricultural University, 2019, 24(6): 124-130.(in Chinese with English abstract)
[23] 张雪芹, 陈嘉豪, 诸葛晶晶, 等. 基于深度学习的快速植物图像识别[J]. 华东理工大学学报(自然科学版), 2018, 44(6): 887-895.
ZHANG X Q, CHEN J H, ZHUGE J J, et al. Deep learning based fast plant image recognition[J]. Journal of East China University of Science and Technology(Natural Science Edition), 2018, 44(6): 887-895.(in Chinese with English abstract)
[24] 王平. 基于迁移学习的烟草叶片病情害识别[J]. 变频器世界, 2018(10): 77-80.
WANG P. Identification of tobacco leaf disease based on transfer learning[J]. The World of Inverters, 2018(10): 77-80.(in Chinese with English abstract)
[25] 袁文翠, 孔雪. 基于迁移学习的图像识别研究[J]. 微型电脑应用, 2018, 34(7): 10-12.
YUAN W C, KONG X. A study of image recognition based on transfer learning[J]. Microcomputer Applications, 2018, 34(7): 10-12.(in Chinese with English abstract)
[26] 杨东旭, 赖惠成, 班俊硕, 等. 基于改进DCNN结合迁移学习的图像分类方法[J]. 新疆大学学报(自然科学版), 2018, 35(2): 195-202.
YANG D X, LAI H C, BAN J S, et al. Image classification method based on improved DCNN combined transfer learning[J]. Journal of Xinjiang University(Natural Science Edition), 2018, 35(2): 195-202.(in Chinese with English abstract)
[27] 高旋, 赵亚凤, 熊强, 等. 基于迁移学习的树种识别[J]. 森林工程, 2019, 35(5): 68-75.
GAO X, ZHAO Y F, XIONG Q, et al. Identification of tree species based on transfer learning[J]. Forest Engineering, 2019, 35(5): 68-75.(in Chinese with English abstract)
[28] 柴帅, 李壮举. 基于迁移学习的番茄病虫害检测[J]. 计算机工程与设计, 2019, 40(6): 1701-1705.
CHAI S, LI Z J. Detection of tomato pests and diseases based on transfer learning[J]. Computer Engineering and Design, 2019, 40(6): 1701-1705.(in Chinese with English abstract)
[29] 冯海林, 胡明越, 杨垠晖, 等. 基于树木整体图像和集成迁移学习的树种识别[J]. 农业机械学报, 2019, 50(8): 235-242.
FENG H L, HU M Y, YANG Y H, et al. Tree species recognition based on overall tree image and ensemble of transfer learning[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(8): 235-242.(in Chinese with English abstract)
[30] 张雪松, 庄严, 闫飞, 等. 基于迁移学习的类别级物体识别与检测研究与进展[J]. 自动化学报, 2019, 45(7): 1224-1243.
ZHANG X S, ZHUANG Y, YAN F, et al. Status and development of transfer learning based category-level object recognition and detection[J]. Acta Automatica Sinica, 2019, 45(7): 1224-1243.(in Chinese with English abstract)
[31] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: a new learning scheme of feedforward neural networks: a new learning scheme of feedforward neural networks[C]//2004 IEEE International Joint Conference on Neural Networks. Budapest, 2004.

基于迁移学习的棉花识别

Cotton recognition based on transfer learning

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	石婧, 刘东洋, 张凤华. 棉花幼苗对盐胁迫的生理响应与耐盐机理[J]. 浙江农业学报, 2020, 32(7): 1141-1148.
[2]	钟伟镇, 刘鑫磊, 杨坤龙, 李丰果. 基于Mask-RCNN的复杂背景下多目标叶片的分割和识别[J]. 浙江农业学报, 2020, 32(11): 2059-2066.
[3]	刘珊珊, 韦鑫, 盛福瑞, 乔清华. 棉花不同发育时期根际微生物的动态变化[J]. 浙江农业学报, 2019, 31(8): 1361-1371.
[4]	孟庆龙, 张艳, 尚静. 基于高光谱成像的猕猴桃表面疤痕无损识别[J]. 浙江农业学报, 2019, 31(8): 1372-1378.
[5]	刘志, 贺正, 苗芳芳, 贾彪. 基于无人机的水肥一体化玉米出苗率估算方法与试验[J]. 浙江农业学报, 2019, 31(6): 977-985.
[6]	梁长江, 吴雪梅, 王芳, 宋朱军, 张富贵. 基于无人机的田间地膜识别算法研究[J]. 浙江农业学报, 2019, 31(6): 1005-1011.
[7]	王彦翔, 张艳, 杨成娅, 孟庆龙, 尚静. 基于深度学习的农作物病害图像识别技术进展[J]. 浙江农业学报, 2019, 31(4): 669-676.
[8]	李颀, 赵洁, 杨柳, 王俊, 高一星. 基于GA-BP神经网络和特征向量优化组合的黄瓜叶片病斑识别[J]. 浙江农业学报, 2019, 31(3): 487-495.
[9]	闫建伟, 苏小东, 赵源, 刘进平. 基于图像特征的小麦胚芽鞘识别[J]. 浙江农业学报, 2019, 31(2): 326-332.
[10]	杨国亮, 许楠, 康乐乐, 龚曼, 洪志阳. 基于参数指数非线性残差神经网络的脐橙病变叶片识别[J]. 浙江农业学报, 2018, 30(6): 1073-1081.
[11]	刘全凤, 刘永震, 曹金锋, 杨振立, 翟玉柱, 赖德强, 孙一, 刘贞贞. 控释氮肥用量对环渤海潮土区棉花产量及氮肥利用率的影响[J]. 浙江农业学报, 2018, 30(2): 275-279.
[12]	邓晓雷, 潘晶晶, 祝水金, 陈进红. 铃期极端高温对转iaaM基因棉花棉铃产量性状、纤维品质和棉仁营养品质的影响[J]. 浙江农业学报, 2018, 30(11): 1805-1810.
[13]	王献锋, 丁军, 朱义海. 基于性能改善深度信念网络的棉花病虫害预测方法[J]. 浙江农业学报, 2018, 30(10): 1790-1797.
[14]	赵学观, 王秀, 李翠玲, 高原源, 王松林, 冯青春. 基于主成分分析及LVQ神经网络的番茄种子品种识别[J]. 浙江农业学报, 2017, 29(8): 1375-1383.
[15]	刘玉姣, 裘波音, 王汐妍, 徐晓建, 祝水金, 陈进红. 棉花microRNA功能研究进展[J]. 浙江农业学报, 2017, 29(6): 1050-1056.