基于光度立体视觉的蔬菜秧苗叶片形态测量方法
|
摘要
为了实现蔬菜秧苗长势智能化在线评估,设计了基于光度立体视觉的蔬菜秧苗叶片曲面形态测量方法和装置,用于对叶片倾角、长度和面积进行精确测量。以15~30 d苗龄辣椒秧苗为测量对象,构建了针对其冠层叶片形态测量的试验装置。根据4组光源在标定球表面成像的反射关系组合方程,标定其对秧苗叶片的照射向量;以D65白色标准板为参考,采用图像RGB分量线性矫正的方法,对不同方位光源辐射强度差异进行补偿,克服光源强度波动对叶片图像明暗信息的影响。根据叶片在不同方位光源照射下图像的明暗特征,基于光度立体视觉获取主叶脉区域离散梯度信息,在此基础上采用最小二乘方法拟合叶片空间平面,以恢复叶片空间倾斜信息,进一步结合叶片图像尺寸测算其实际长度和面积。试验结果表明,视觉系统对秧苗倾角、叶片长度和面积的测量结果与人工测量结果相比平均偏差分别为6.29°、3.82 mm、56.53 mm2,叶片长度和面积与人工测量结果的决定系数R2分别为0.936 3、0.866 4,且对于叶片伸展充分、无遮挡的幼龄秧苗测量精度较高,可为进一步开发温室苗床秧苗长势在线监测设备提供重要技术支撑。
In order to achieve the on-line nondestructive assessment for vegetable seedling growth,the method and device for measuring the seedling 's leaf spatial morphology based on photometric stereo vision was introduced,and the morphology information,including seedling leaf 's inclination angle,length and size. The 15 ~ 30 d pepper seedlings were adopted as research objects,and the test device for measuring their canopy leaves was built,which involved the four different LED lights ' orientation calibration and the image color's linear correction referring to the D65 palette's color variation under the various illumination. According to the light reflexive relation on the calibration ball from the LED lights to the camera,the spatial orientations of illuminants under the camera coordinate were determined. And the image color correction was supposed to overcome the color distortion caused from the lights' luminance fluctuation. According to the seedling leaves' light-shadow feature under various illuminants,the discrete gradient information along the leaf's main vein was obtained based on photometric stereo vision. And referring to the gradients vectors,the leaf's space plane was fit by adopting the least-square principle.The leaf's inclination angle was measured as the angle between the fitting plane and camera imaging plane,thus the actual spatial leaf size also could be accurately measured on the basis of the image size.As the test result shown,compared with the manual measurement result,the auto-measurement method had an average errors of 6. 29° for leaf inclination angle,3. 82 mm for leaf length,and 56. 53 mm2 for leaf size,and the determination coefficients on the length and leaf size measurement were 0. 936 3 and0. 866 4,respectively. Besides,the measuring result would be more accurate for the younger seedling with fully-opened and unoverlapped leaves.
In order to achieve the on-line nondestructive assessment for vegetable seedling growth,the method and device for measuring the seedling 's leaf spatial morphology based on photometric stereo vision was introduced,and the morphology information,including seedling leaf 's inclination angle,length and size. The 15 ~ 30 d pepper seedlings were adopted as research objects,and the test device for measuring their canopy leaves was built,which involved the four different LED lights ' orientation calibration and the image color's linear correction referring to the D65 palette's color variation under the various illumination. According to the light reflexive relation on the calibration ball from the LED lights to the camera,the spatial orientations of illuminants under the camera coordinate were determined. And the image color correction was supposed to overcome the color distortion caused from the lights' luminance fluctuation. According to the seedling leaves' light-shadow feature under various illuminants,the discrete gradient information along the leaf's main vein was obtained based on photometric stereo vision. And referring to the gradients vectors,the leaf's space plane was fit by adopting the least-square principle.The leaf's inclination angle was measured as the angle between the fitting plane and camera imaging plane,thus the actual spatial leaf size also could be accurately measured on the basis of the image size.As the test result shown,compared with the manual measurement result,the auto-measurement method had an average errors of 6. 29° for leaf inclination angle,3. 82 mm for leaf length,and 56. 53 mm2 for leaf size,and the determination coefficients on the length and leaf size measurement were 0. 936 3 and0. 866 4,respectively. Besides,the measuring result would be more accurate for the younger seedling with fully-opened and unoverlapped leaves.
引文
1张凯,魏敏芝,陈青云,等.黄瓜穴盘苗壮苗指标的初步研究[J].华中农业大学学报,2004,35(增刊2):240-244.ZHANG Kai,WEI Minzhi,CHEN Qingyun,et al.Study on the healthy index of the plug seedling of the cucumber[J].Journal of Huazhong Agricultural University,2004,35(Supp.2):240-244.(in Chinese)
2 韦泽秀,梁银丽,周茂娟,等.水肥组合对日光温室黄瓜叶片生长和产量的影响[J].农业工程学报,2010,26(3):69-74.WEI Zexiu,LIANG Yinli,ZHOU Maojuan,et al.Physiological characteristics of leaf growth and yield of cucumber under different watering and fertilizer coupling treatments in greenhouse[J].Transactions of the CSAE,2010,26(3):69-74.(in Chinese)
3 郝金魁,张西群,齐新,等.工厂化育苗技术现状与发展对策[J].江苏农业科学,2012,40(1):349-351.HAO Jinkui,ZHANG Xiqun,QI Xin,et al.Present situation and development countermeasures of factory nursery technology[J].Jiangsu Agricultural Sciences,2012,40(1):349-351.(in Chinese)
4 HUANG Kuoyi.Application of artificial neural network for detecting Phalaenopsis seedling diseases using color and texture features[J].Computers and Electronics in Agriculture,2007,57:3-11.
5 张彦娥,李民赞,张喜杰,等.基于计算机视觉技术的温室黄瓜叶片营养信息检测[J].农业工程学报,2005,21(8):102-105.ZHANG Yan'e,LI Minzan,ZHANG Xijie,et al.Nutrition information extraction of the cucumber leaves in the greenhouse based on computer vision technology[J].Transactions of the CSAE,2005,21(8):102-105.(in Chinese)
6 张雷,贺虎,武传宇.蔬菜嫁接机器人嫁接苗特征参数的视觉测量方法[J].农业工程学报,2015,31(9):32-38.ZHANG Lei,HE Hu,WU Chuanyu.Vision method for measuring grafted seedling properties of vegetable grafted robot[J].Transactions of the CSAE,2015,31(9):32-38.(in Chinese)
7 杨意,初麒,杨艳丽,等.基于机器视觉的白掌组培苗在线分级方法[J].农业工程学报,2016,32(8):33-40.YANG Yi,CHU Qi,YANG Yanli,et al.Online grading method for tissue culture seedlings of Spathiphyllum floribundum based on machine vision[J].Transactions of the CSAE,2016,32(8):33-40.(in Chinese)
8 FLORA Yuhui,LAI Tsungcheng,LIU Tingyu.An automated growth measurement system for leafy vegetables[J].Biosystems Engineering,2014,117:43-50.
9 GISELSSON T M,MIDTIBY H S.Seedling discrimination with shape features derived from a distance transform[J].Sensors,2013,13:5585-5602.
10 YU Zhenghong,CAO Zhiguo,WU Xi.Automatic image-based detection technology for two critical growth stages of maize:emergence and three-leaf stage[J].Agricultural and Forest Meteorology,2013,174:68-84.
11 TONG J H,LI J B,JIANG H Y.Machine vision techniques for the evaluation of seedling quality based on leaf area[J].Biosystems Enginering,2013,115:369-379.
12 孙国祥,汪小旵,何国敏.基于边缘链码信息的番茄苗重叠叶面分割算法[J].农业工程学报,2010,26(12):206-211.SUN Guoxiang,WANG Xiaochan,HE Guomin.Segmentation algorithm of overlapping tomato seedling leaves based on edge chaincode information[J].Transactions of the CSAE,2010,26(12):206-211.(in Chinese)
13 JULIO C Pastrana,THOMAS Rath.Novel image processing approach for solving the overlapping problem in agriculture[J].Biosystems Enginering,2013,115:106-115.
14 张新,赵燕东,郑力嘉,等.基于三维机器视觉的植物叶片萎蔫预测模型[J/OL].农业机械学报,2014,45(9):260-267.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20140942&flag=1&journal_id=jcsam.DOI:10.6041/j.issn.100-1298.2014.09.042.ZHANG Xin,ZHAO Yandong,ZHENG Lijia,et al.Prediction model of plant leaf wilting using 3-D machine vision[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2014,45(9):260-267.(in Chinese)
15 冯青春,刘新南,姜凯,等.基于线结构光视觉的穴盘苗外形参数在线测量系统研制及试验[J].农业工程学报,2013,29(21):143-149.FENG Qingchun,LIU Xinnan,JIANG Kai,et al.Development and experiment on system for tray-seedling on-line measurement based on line structured-light vision[J].Transactions of the CSAE,2013,29(21):143-149.(in Chinese)
16 王传宇,赵明,阎建河,等.基于双目立体视觉技术的玉米叶片三维重建[J].农业工程学报,2010,26(4):198-202.WANG Chuanyu,ZHAO Ming,YAN Jianhe,et al.Three-dimensional reconstruction of maize leaves based on binocular stereovision system[J].Transactions of the CSAE,2010,26(4):198-202.(in Chinese)
17 于合龙,黄浦,苏红宏,等.基于双目视觉的植物叶片三维形态与光照度同步测量[J].农业工程学报,2016,32(10):149-156.YU Helong,HUANG Pu,SU Honghong,et al.Synchronous measurement of 3D morphology and illuminance of plant leaves based on binocular vision[J].Transactions of the CSAE,2016,32(10):149-156.(in Chinese)
18 谢利民.基于近场照明的光度立体视觉算法研究[D].武汉:华中科技大学,2015.XIE Limin.Research on photometric stereo algorithm based on near-field lighting[D].Wuhan:Huazhong University of Science&Technology,2015.(in Chinese)
19 程岳.基于光度立体的高质量表面重建研究[D].杭州:浙江大学,2013.CHENG Yue.High quality surface reconstruction using photometric stereo[D].Hangzhou:Zhejiang University,2013.(in Chinese)
20 芦碧波,李玉静,王玉琨,等.亮度分区自适应对数色调映射算法[J].计算机应用研究,2018,35(9):1-8.LU Bibo,LI Yujing,WANG Yukun,et al.Adaptive logarithmic tone mapping arithmetic with luminance regionalization[J].Application Research of Computers,2018,35(9):1-8.(in Chinese)
21 郭惠楠,曹剑中,王华,等.高动态范围数字相机sRGB色彩空间颜色管理[J].红外与激光工程,2014,43(增刊):238-242.GUO Huinan,CAO Jianzhong,WANG Hua,et al.Color management of sRGB color space for HDR digital camera[J].Infrared and Laser Engineering,2014,43(Supp.):238-242.(in Chinese)
22 朱磊.色域扩展技术的研究与实现[D].上海:上海交通大学,2008.ZHU Lei.Design and implementation of gamut extension techniques[J].Shanghai:Shanghai Jiao Tong University,2008.(in Chinese)
23 刘志平,石林英.最小二乘法原理及其Matlab实现[J].中国西部科技,2008(17):33-34.LIU Zhiping,SHI Linying.The principle of least square algorithm and its achievement by Matlab[J].Science and Technology of West China,2008(17):33-34.(in Chinese)
2 韦泽秀,梁银丽,周茂娟,等.水肥组合对日光温室黄瓜叶片生长和产量的影响[J].农业工程学报,2010,26(3):69-74.WEI Zexiu,LIANG Yinli,ZHOU Maojuan,et al.Physiological characteristics of leaf growth and yield of cucumber under different watering and fertilizer coupling treatments in greenhouse[J].Transactions of the CSAE,2010,26(3):69-74.(in Chinese)
3 郝金魁,张西群,齐新,等.工厂化育苗技术现状与发展对策[J].江苏农业科学,2012,40(1):349-351.HAO Jinkui,ZHANG Xiqun,QI Xin,et al.Present situation and development countermeasures of factory nursery technology[J].Jiangsu Agricultural Sciences,2012,40(1):349-351.(in Chinese)
4 HUANG Kuoyi.Application of artificial neural network for detecting Phalaenopsis seedling diseases using color and texture features[J].Computers and Electronics in Agriculture,2007,57:3-11.
5 张彦娥,李民赞,张喜杰,等.基于计算机视觉技术的温室黄瓜叶片营养信息检测[J].农业工程学报,2005,21(8):102-105.ZHANG Yan'e,LI Minzan,ZHANG Xijie,et al.Nutrition information extraction of the cucumber leaves in the greenhouse based on computer vision technology[J].Transactions of the CSAE,2005,21(8):102-105.(in Chinese)
6 张雷,贺虎,武传宇.蔬菜嫁接机器人嫁接苗特征参数的视觉测量方法[J].农业工程学报,2015,31(9):32-38.ZHANG Lei,HE Hu,WU Chuanyu.Vision method for measuring grafted seedling properties of vegetable grafted robot[J].Transactions of the CSAE,2015,31(9):32-38.(in Chinese)
7 杨意,初麒,杨艳丽,等.基于机器视觉的白掌组培苗在线分级方法[J].农业工程学报,2016,32(8):33-40.YANG Yi,CHU Qi,YANG Yanli,et al.Online grading method for tissue culture seedlings of Spathiphyllum floribundum based on machine vision[J].Transactions of the CSAE,2016,32(8):33-40.(in Chinese)
8 FLORA Yuhui,LAI Tsungcheng,LIU Tingyu.An automated growth measurement system for leafy vegetables[J].Biosystems Engineering,2014,117:43-50.
9 GISELSSON T M,MIDTIBY H S.Seedling discrimination with shape features derived from a distance transform[J].Sensors,2013,13:5585-5602.
10 YU Zhenghong,CAO Zhiguo,WU Xi.Automatic image-based detection technology for two critical growth stages of maize:emergence and three-leaf stage[J].Agricultural and Forest Meteorology,2013,174:68-84.
11 TONG J H,LI J B,JIANG H Y.Machine vision techniques for the evaluation of seedling quality based on leaf area[J].Biosystems Enginering,2013,115:369-379.
12 孙国祥,汪小旵,何国敏.基于边缘链码信息的番茄苗重叠叶面分割算法[J].农业工程学报,2010,26(12):206-211.SUN Guoxiang,WANG Xiaochan,HE Guomin.Segmentation algorithm of overlapping tomato seedling leaves based on edge chaincode information[J].Transactions of the CSAE,2010,26(12):206-211.(in Chinese)
13 JULIO C Pastrana,THOMAS Rath.Novel image processing approach for solving the overlapping problem in agriculture[J].Biosystems Enginering,2013,115:106-115.
14 张新,赵燕东,郑力嘉,等.基于三维机器视觉的植物叶片萎蔫预测模型[J/OL].农业机械学报,2014,45(9):260-267.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20140942&flag=1&journal_id=jcsam.DOI:10.6041/j.issn.100-1298.2014.09.042.ZHANG Xin,ZHAO Yandong,ZHENG Lijia,et al.Prediction model of plant leaf wilting using 3-D machine vision[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2014,45(9):260-267.(in Chinese)
15 冯青春,刘新南,姜凯,等.基于线结构光视觉的穴盘苗外形参数在线测量系统研制及试验[J].农业工程学报,2013,29(21):143-149.FENG Qingchun,LIU Xinnan,JIANG Kai,et al.Development and experiment on system for tray-seedling on-line measurement based on line structured-light vision[J].Transactions of the CSAE,2013,29(21):143-149.(in Chinese)
16 王传宇,赵明,阎建河,等.基于双目立体视觉技术的玉米叶片三维重建[J].农业工程学报,2010,26(4):198-202.WANG Chuanyu,ZHAO Ming,YAN Jianhe,et al.Three-dimensional reconstruction of maize leaves based on binocular stereovision system[J].Transactions of the CSAE,2010,26(4):198-202.(in Chinese)
17 于合龙,黄浦,苏红宏,等.基于双目视觉的植物叶片三维形态与光照度同步测量[J].农业工程学报,2016,32(10):149-156.YU Helong,HUANG Pu,SU Honghong,et al.Synchronous measurement of 3D morphology and illuminance of plant leaves based on binocular vision[J].Transactions of the CSAE,2016,32(10):149-156.(in Chinese)
18 谢利民.基于近场照明的光度立体视觉算法研究[D].武汉:华中科技大学,2015.XIE Limin.Research on photometric stereo algorithm based on near-field lighting[D].Wuhan:Huazhong University of Science&Technology,2015.(in Chinese)
19 程岳.基于光度立体的高质量表面重建研究[D].杭州:浙江大学,2013.CHENG Yue.High quality surface reconstruction using photometric stereo[D].Hangzhou:Zhejiang University,2013.(in Chinese)
20 芦碧波,李玉静,王玉琨,等.亮度分区自适应对数色调映射算法[J].计算机应用研究,2018,35(9):1-8.LU Bibo,LI Yujing,WANG Yukun,et al.Adaptive logarithmic tone mapping arithmetic with luminance regionalization[J].Application Research of Computers,2018,35(9):1-8.(in Chinese)
21 郭惠楠,曹剑中,王华,等.高动态范围数字相机sRGB色彩空间颜色管理[J].红外与激光工程,2014,43(增刊):238-242.GUO Huinan,CAO Jianzhong,WANG Hua,et al.Color management of sRGB color space for HDR digital camera[J].Infrared and Laser Engineering,2014,43(Supp.):238-242.(in Chinese)
22 朱磊.色域扩展技术的研究与实现[D].上海:上海交通大学,2008.ZHU Lei.Design and implementation of gamut extension techniques[J].Shanghai:Shanghai Jiao Tong University,2008.(in Chinese)
23 刘志平,石林英.最小二乘法原理及其Matlab实现[J].中国西部科技,2008(17):33-34.LIU Zhiping,SHI Linying.The principle of least square algorithm and its achievement by Matlab[J].Science and Technology of West China,2008(17):33-34.(in Chinese)