先验阈值优化卷积神经网络的作物覆盖度提取算法
|
摘要
农作物的生长观测是农业气象观测的一个重要部分,作物的覆盖度反映了环境对作物综合影响的结果,传统的先验阈值分割法受作物图像中存在的田间杂物、下雨或施肥后的土地以及光照阴影影响较大,会存在误分割的情况,针对这些问题,本文研究基于深度学习的作物与背景的自动分割问题,提出基于RGB和HSI关系阈值法优化的卷积神经网络(RGB-HSI-CNN)的作物图像分割提取覆盖度方法,解决了光照、遮挡、阴影等影响,取得了平均98.3%的模型准确率和97.53%的像素准确率,为后续作物生长状况监测以及农作物病虫草害的识别、监测等提供了有力支持。
The growth of crops is an important part of agricultural meteorological observations. Crop coverage reflects the results of the environmental impact on the crop. The traditional threshold segmentation method may result in misclassification due to sundries,light,shadows,fertilization and rain in the crop image. In this paper,to solve these problems,propose a method to extract crop coverage and obtain image segmentation based on convolution neural network optimized by RGB and HSI relation threshold method( RGB-HIS-CNN). The average model accuracy is 98. 3% and the pixel accuracy is97. 53%. It provides strong support for the monitoring of growth status and the identification of crop diseases and insects.
The growth of crops is an important part of agricultural meteorological observations. Crop coverage reflects the results of the environmental impact on the crop. The traditional threshold segmentation method may result in misclassification due to sundries,light,shadows,fertilization and rain in the crop image. In this paper,to solve these problems,propose a method to extract crop coverage and obtain image segmentation based on convolution neural network optimized by RGB and HSI relation threshold method( RGB-HIS-CNN). The average model accuracy is 98. 3% and the pixel accuracy is97. 53%. It provides strong support for the monitoring of growth status and the identification of crop diseases and insects.
引文
[1]王馥棠.中国气象科学研究院农业气象研究50年进展[J].应用气象学报,2006,17(6):778-785.Wang Futang.Advances of Agro-meteorological Research in CAMS During Recent 50 Years[J].Journal of Applied Meteorological Science,2006,17(6):778-785.(in Chinese)
[2]余正泓.基于图像的玉米发育期自动观测技术研究[D].武汉:华中科技大学,2014.Yu Zhenghong.Research on Image-based Automatic Observation Technology for Critical Growth Stages of Maize[D].Wuhan:Huazhong University of Science Technology,2014.(in Chinese)
[3]Rehkugler G E,Throop J A.Apple Sorting with Machine Vision[J].Transactions of the ASAE-American Society of Agricultural Engineers(USA),1986,29(5):1388-1397.
[4]Gunasekaran S,Cooper T M,Berlage A G,et al.Image Processing for Stress Cracks in Corn Kernels[J].Transactions of the ASAE-American Society of Agricultural Engineers(USA),1987,30(1):266-273.
[5]Woebbecke D M,Meyer G E,Bargen K V,et al.Shape Features for Identifying Young Weeds Using Image Analysis[J].Transactions of the Asae,1995,38(1):271-281.
[6]Woebbecke D M,Meyer G E,Von Bargen K,et al.Color indices for weed identification under various soil,residue,and lighting conditions[C]∥American Society of Agricultural Engineers.Meeting.1994.
[7]Giles.Precision band spraying with machine-vision guidance and adjustable yaw nozzles[J].Transactions of the Asae,1997,40(1):29-36.
[8]Lee W S.Robotic weed control system for tomatoes[C]∥University of California,Davis,1998:95-113.
[9]Lukina E V,Stone M L,Raun W R.Estimating vegetation coverage in wheat using digital images[J].Journal of Plant Nutrition,1999,22(2):341-350.
[10]Tang L,Tian L,Steward B L.Color image segmentation with genetic algorithm for in-field weed sensing[J].Transactions of the Asae,2000,43(4):1019-1027.
[11]纪寿文,王荣本,陈佳娟,等.应用计算机图像处理技术识别玉米苗期田间杂草的研究[J].农业工程学报,2001,17(2):154-156.Ji Shouwen,Wang Rongben,Chen Jiajuan,et al.Research on Recognizing Weed From Corn Seedling by Using Computer Image Processing Technology[J].Transactions of the CSAE,2001,17(2):154-156.(in Chinese)
[12]毛文华.基于机器视觉的田间杂草识别技术研究[D].北京:中国农业大学,2004.Mao Wenhua.Study on Infield Weed Detection Using Machine Vision[D].Beijing:China Agricultural University,2004.(in Chinese)
[13]毛文华,王一鸣,张小超,等.基于机器视觉的苗期杂草实时分割算法[J].农业机械学报,2005,36(1):83-86.Mao Wenhua,Wang Yiming,Zhang Xiaochao,et al.Realtime Algorithm Based on Machine Vision to Segment Weeds at Seedling[J].Transactions of the Chinese Society of Agricultural Machinery,2005,36(1):83-86.(in Chinese)
[14]毛罕平,胡波,张艳诚,等.杂草识别中颜色特征和阈值分割算法的优化[J].农业工程学报,2007,23(9):154-158.Mao Hanping,Hu Bo,Zhang Yancheng,et al.Optimization of color index and threshold segmentation in weed recognition[J].Transactions of the Chinese Society of Agricultural Engineering,2007,23(9):154-158.(in Chinese)
[15]Tellaeche A,Burgosartizzu X P,Pajares G,et al.A new vision-based approach to differential spraying in precision agriculture[J].Computers&Electronics in Agriculture,2008,60(2):144-155.
[16]何姣,张雪芬,汤志亚,等.棉花覆盖度特征及与其他生长参数关系分析[J].气象与环境科学,2015,38(2):18-23.He Jiao,Zhang Xuefen,Tang Zhiya,et al.Analysis of Cotton Coverage Characteristics and the Relationships with Other Growth Parameters[J].Meteorological and Environmental Sciences,2015,38(2):18-23.(in Chinese)
[17]张丽芬.基于计算机视觉技术的樱桃自动分选系统设计[J].农机化研究,2017,39(9):212-214.Zhang Lifen.Design of Cherry Sorting System Based on Computer Vision Technology[J].Journal of Agricultural Mechanization Research,2017,39(9):212-214.(in Chinese)
[18]严锦程.作物冠层覆盖率、株高的远程自动监测技术[D].浙江:浙江理工大学,2016.Yan Jincheng.Technology on Remote Automatic Monitoring of Crop Canopy Coverage and Height[D].Zhejiang:Zhejiang Sci-tec University,2016.(in Chinese)
[19]Lecun Y,Boser B,Denker J S,et al.Backpropagation applied to handwritten zip code recognition[J].Neural Computation,1989,1(4):541-551.
[20]Krizhevsky A.Convolutional Deep Belief Networks on CIFAR-10[J].2010.
[21]余凯,贾磊,陈雨强,等.深度学习的昨天、今天和明天[J].计算机研究与发展,2013,50(9):1799-1804.Yu Kai,Jia Lei,Chen Yuqiang,et al.Deep Learning:Yesterday,Today,and Tomorrow[J].Journal of Computer Research and Development,2013,50(9):1799-1804.(in Chinese)
[22]Russakovsky O,Deng J,Su H,et al.Image Net Large Scale Visual Recognition Challenge[J].International Journal of Computer Vision,2015,115(3):211-252.
[23]Krizhevsky A,Sutskever I,Hinton G E.Image Net classification with deep convolutional neural networks[C]∥International Conference on Neural Information Processing Systems.Curran Associates Inc.2012:1097-1105.
[24]Wu Z,Huang Y,Yu Y,et al.Early Hierarchical Contexts Learned by Convolutional Networks for Image Segmentation[C]∥International Conference on Pattern Recognition.IEEE Computer Society,2014:1538-1543.
[25]Long J,Shelhamer E,Darrell T.Fully convolutional networks for semantic segmentation[C]∥IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2015:3431-3440.
[26]Zheng S,Jayasumana S,Romeraparedes B,et al.Conditional Random Fields as Recurrent Neural Networks[C]∥IEEE International Conference on Computer Vision.IEEE,2015:1529-1537.
[27]毋立芳,贺娇瑜,张加楠,等.一种基于多分辨率CNN的毫米波雷达云图分割方法[P].2016.中国专利:CN106127725A.Wu Lifang,He Jiaoyu,Zhang Jianan,et al.A Millimeterwave Cloud Image Segmentation Based on Multi-resolution Convolutional Neural Network[P].2016.CN Patent:CN106127725A.(in Chinese)
[28]张加楠,毋立芳,贺娇瑜,等.基于多分辨率卷积神经网络的毫米波雷达云图像分割[C]∥第十二届和谐人际环境会议(HHME2016),2016.Zhang Jianan,Wu Lifang,He Jiaoyu,et al.Millimeterwave Cloud Image Segmentation Based on Multiresolution Convolutional Neural Network[C]∥HHME2016,2016.(in Chinese)
[29]韩思奇,王蕾.图像分割的阈值法综述[J].系统工程与电子技术,2002,24(6):91-94.Han Siqi,Wang Lei.A Survey of Thresholding Methods for Image Segmentation[J].Systems Engineering and Electronics,2002,24(6):91-94.(in Chinese)
[30]丁亮,张永平,张雪英.图像分割方法及性能评价综述[J].软件,2010,31(12):78-83.Ding Liang,Zhang Yongping,Zhang Xueying.A Survey of Image segmentation Techniques and Performance Evaluation[J].Software,2010,31(12):78-83.(in Chinese)
[31]石建萍,梁继,隋凌志.基于全卷积网络人脸五官定位与判别的方法及系统[P].2016.中国专利:CN105354565A,2016-02-24.Shi Jianping,Liang Ji,Sui Lingzhi.Method and system for facial features localization and discrimination based on full convolution network[P].2016.CN Patent:CN105354565A,2016-02-24.(in Chinese)
[32]Kamnitsas K,Ledig C,Newcombe V F,et al.Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation[J].Medical Image Analysis,2017,36:61-78.
[2]余正泓.基于图像的玉米发育期自动观测技术研究[D].武汉:华中科技大学,2014.Yu Zhenghong.Research on Image-based Automatic Observation Technology for Critical Growth Stages of Maize[D].Wuhan:Huazhong University of Science Technology,2014.(in Chinese)
[3]Rehkugler G E,Throop J A.Apple Sorting with Machine Vision[J].Transactions of the ASAE-American Society of Agricultural Engineers(USA),1986,29(5):1388-1397.
[4]Gunasekaran S,Cooper T M,Berlage A G,et al.Image Processing for Stress Cracks in Corn Kernels[J].Transactions of the ASAE-American Society of Agricultural Engineers(USA),1987,30(1):266-273.
[5]Woebbecke D M,Meyer G E,Bargen K V,et al.Shape Features for Identifying Young Weeds Using Image Analysis[J].Transactions of the Asae,1995,38(1):271-281.
[6]Woebbecke D M,Meyer G E,Von Bargen K,et al.Color indices for weed identification under various soil,residue,and lighting conditions[C]∥American Society of Agricultural Engineers.Meeting.1994.
[7]Giles.Precision band spraying with machine-vision guidance and adjustable yaw nozzles[J].Transactions of the Asae,1997,40(1):29-36.
[8]Lee W S.Robotic weed control system for tomatoes[C]∥University of California,Davis,1998:95-113.
[9]Lukina E V,Stone M L,Raun W R.Estimating vegetation coverage in wheat using digital images[J].Journal of Plant Nutrition,1999,22(2):341-350.
[10]Tang L,Tian L,Steward B L.Color image segmentation with genetic algorithm for in-field weed sensing[J].Transactions of the Asae,2000,43(4):1019-1027.
[11]纪寿文,王荣本,陈佳娟,等.应用计算机图像处理技术识别玉米苗期田间杂草的研究[J].农业工程学报,2001,17(2):154-156.Ji Shouwen,Wang Rongben,Chen Jiajuan,et al.Research on Recognizing Weed From Corn Seedling by Using Computer Image Processing Technology[J].Transactions of the CSAE,2001,17(2):154-156.(in Chinese)
[12]毛文华.基于机器视觉的田间杂草识别技术研究[D].北京:中国农业大学,2004.Mao Wenhua.Study on Infield Weed Detection Using Machine Vision[D].Beijing:China Agricultural University,2004.(in Chinese)
[13]毛文华,王一鸣,张小超,等.基于机器视觉的苗期杂草实时分割算法[J].农业机械学报,2005,36(1):83-86.Mao Wenhua,Wang Yiming,Zhang Xiaochao,et al.Realtime Algorithm Based on Machine Vision to Segment Weeds at Seedling[J].Transactions of the Chinese Society of Agricultural Machinery,2005,36(1):83-86.(in Chinese)
[14]毛罕平,胡波,张艳诚,等.杂草识别中颜色特征和阈值分割算法的优化[J].农业工程学报,2007,23(9):154-158.Mao Hanping,Hu Bo,Zhang Yancheng,et al.Optimization of color index and threshold segmentation in weed recognition[J].Transactions of the Chinese Society of Agricultural Engineering,2007,23(9):154-158.(in Chinese)
[15]Tellaeche A,Burgosartizzu X P,Pajares G,et al.A new vision-based approach to differential spraying in precision agriculture[J].Computers&Electronics in Agriculture,2008,60(2):144-155.
[16]何姣,张雪芬,汤志亚,等.棉花覆盖度特征及与其他生长参数关系分析[J].气象与环境科学,2015,38(2):18-23.He Jiao,Zhang Xuefen,Tang Zhiya,et al.Analysis of Cotton Coverage Characteristics and the Relationships with Other Growth Parameters[J].Meteorological and Environmental Sciences,2015,38(2):18-23.(in Chinese)
[17]张丽芬.基于计算机视觉技术的樱桃自动分选系统设计[J].农机化研究,2017,39(9):212-214.Zhang Lifen.Design of Cherry Sorting System Based on Computer Vision Technology[J].Journal of Agricultural Mechanization Research,2017,39(9):212-214.(in Chinese)
[18]严锦程.作物冠层覆盖率、株高的远程自动监测技术[D].浙江:浙江理工大学,2016.Yan Jincheng.Technology on Remote Automatic Monitoring of Crop Canopy Coverage and Height[D].Zhejiang:Zhejiang Sci-tec University,2016.(in Chinese)
[19]Lecun Y,Boser B,Denker J S,et al.Backpropagation applied to handwritten zip code recognition[J].Neural Computation,1989,1(4):541-551.
[20]Krizhevsky A.Convolutional Deep Belief Networks on CIFAR-10[J].2010.
[21]余凯,贾磊,陈雨强,等.深度学习的昨天、今天和明天[J].计算机研究与发展,2013,50(9):1799-1804.Yu Kai,Jia Lei,Chen Yuqiang,et al.Deep Learning:Yesterday,Today,and Tomorrow[J].Journal of Computer Research and Development,2013,50(9):1799-1804.(in Chinese)
[22]Russakovsky O,Deng J,Su H,et al.Image Net Large Scale Visual Recognition Challenge[J].International Journal of Computer Vision,2015,115(3):211-252.
[23]Krizhevsky A,Sutskever I,Hinton G E.Image Net classification with deep convolutional neural networks[C]∥International Conference on Neural Information Processing Systems.Curran Associates Inc.2012:1097-1105.
[24]Wu Z,Huang Y,Yu Y,et al.Early Hierarchical Contexts Learned by Convolutional Networks for Image Segmentation[C]∥International Conference on Pattern Recognition.IEEE Computer Society,2014:1538-1543.
[25]Long J,Shelhamer E,Darrell T.Fully convolutional networks for semantic segmentation[C]∥IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2015:3431-3440.
[26]Zheng S,Jayasumana S,Romeraparedes B,et al.Conditional Random Fields as Recurrent Neural Networks[C]∥IEEE International Conference on Computer Vision.IEEE,2015:1529-1537.
[27]毋立芳,贺娇瑜,张加楠,等.一种基于多分辨率CNN的毫米波雷达云图分割方法[P].2016.中国专利:CN106127725A.Wu Lifang,He Jiaoyu,Zhang Jianan,et al.A Millimeterwave Cloud Image Segmentation Based on Multi-resolution Convolutional Neural Network[P].2016.CN Patent:CN106127725A.(in Chinese)
[28]张加楠,毋立芳,贺娇瑜,等.基于多分辨率卷积神经网络的毫米波雷达云图像分割[C]∥第十二届和谐人际环境会议(HHME2016),2016.Zhang Jianan,Wu Lifang,He Jiaoyu,et al.Millimeterwave Cloud Image Segmentation Based on Multiresolution Convolutional Neural Network[C]∥HHME2016,2016.(in Chinese)
[29]韩思奇,王蕾.图像分割的阈值法综述[J].系统工程与电子技术,2002,24(6):91-94.Han Siqi,Wang Lei.A Survey of Thresholding Methods for Image Segmentation[J].Systems Engineering and Electronics,2002,24(6):91-94.(in Chinese)
[30]丁亮,张永平,张雪英.图像分割方法及性能评价综述[J].软件,2010,31(12):78-83.Ding Liang,Zhang Yongping,Zhang Xueying.A Survey of Image segmentation Techniques and Performance Evaluation[J].Software,2010,31(12):78-83.(in Chinese)
[31]石建萍,梁继,隋凌志.基于全卷积网络人脸五官定位与判别的方法及系统[P].2016.中国专利:CN105354565A,2016-02-24.Shi Jianping,Liang Ji,Sui Lingzhi.Method and system for facial features localization and discrimination based on full convolution network[P].2016.CN Patent:CN105354565A,2016-02-24.(in Chinese)
[32]Kamnitsas K,Ledig C,Newcombe V F,et al.Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation[J].Medical Image Analysis,2017,36:61-78.