基于形心法的荞麦籽粒三轴尺寸亚像素测量方法研究
|
摘要
为给杂粮播种及收获机械,特别是排种器及凹板型孔等提供设计依据,以荞麦籽粒为研究对象,以亚像素形心定位技术为手段,在红色LED背光源下,采集包括标定物在内的荞麦籽粒群投影图像和侧视图像。经图像灰度化及二值化、图像去噪及边缘平滑、籽粒图像亚像素级形心定位及最小外接矩形绘制、图像标定等处理后,对荞麦籽粒的三轴尺寸在0.3像素级上进行精确测量。结果表明,该方法同时测量8粒荞麦籽粒平均耗时为14s,而利用游标卡尺对1粒籽粒进行人工测量,平均耗时约20s;籽粒长、宽、高三轴尺寸测量的平均相对误差分别为2.98%、2.64%和3.45%。为进一步验证该方法的准确性和适应性,继续测量燕麦、高粱、青稞、藜麦、薏米等5种杂粮籽粒的三轴尺寸,其长、宽、高的平均相对误差均在3%左右。研究结果一方面为荞麦等杂粮籽粒三轴尺寸提供一种高效、精准的测量技术,便于杂粮脱粒或清选等关键部件的设计,另一方面,也为类似小尺寸物体三轴尺寸的测量提供一种借鉴和参考。
In order to provide design basis for miscellaneous grains sowing and harvesting machinery,especially seeders and concave plate holes,the buckwheat kernel was used as the research object in this paper,and the sub-pixel centroid positioning technique was used as the method.Under the red LED backlight,the projection images and side view images of the buckwheat grain group were collected.The length,width,and height of each buckwheat kernel were measured precisely at 0.3 pixel level after image gray scale and binarization,image denoising and edge smoothing,sub-pixel centroid positioning and minimum external rectangle drawing and image calibration,and so on.The results showed that the average time-consuming of 8 grains of buckwheat was 14 s,however,the average time-consuming was 20 sfor one kernel using vernier caliper manually measurement.The average relative error of grain length,width and height were 2.98%,2.64% and 3.45%,respectively.In order to further verify the accuracy and adaptability of the method,we continuously measured the triaxial dimensions of five grains of oatmeal,sorghum,barley,quinoa and barley.The average relative error of length,width and height was about 3%.The results of the study provide an efficient and accurate on-line measurement technique for the three-axis grain size of buckwheat and other grains.On the other hand,it can also provide a reference for the measurement of the triaxial dimension of small objects.
In order to provide design basis for miscellaneous grains sowing and harvesting machinery,especially seeders and concave plate holes,the buckwheat kernel was used as the research object in this paper,and the sub-pixel centroid positioning technique was used as the method.Under the red LED backlight,the projection images and side view images of the buckwheat grain group were collected.The length,width,and height of each buckwheat kernel were measured precisely at 0.3 pixel level after image gray scale and binarization,image denoising and edge smoothing,sub-pixel centroid positioning and minimum external rectangle drawing and image calibration,and so on.The results showed that the average time-consuming of 8 grains of buckwheat was 14 s,however,the average time-consuming was 20 sfor one kernel using vernier caliper manually measurement.The average relative error of grain length,width and height were 2.98%,2.64% and 3.45%,respectively.In order to further verify the accuracy and adaptability of the method,we continuously measured the triaxial dimensions of five grains of oatmeal,sorghum,barley,quinoa and barley.The average relative error of length,width and height was about 3%.The results of the study provide an efficient and accurate on-line measurement technique for the three-axis grain size of buckwheat and other grains.On the other hand,it can also provide a reference for the measurement of the triaxial dimension of small objects.
引文
[1]梁晋,尤威,王晓光,等.三维复杂尺寸检测的研究与应用[J].中国工程机械学报,2017,15(1):67~71.Liang Jin,You Wei,Wang Xiaoguang,et al.The research and application of three dimensional complex size detection[J].Chinese Journal of Construction Machinery,2017,15(1):67~71.
[2]FlorkováZ,Jambor M.Quantification of aggregate surface texture based on three dimensional microscope measurement[J].Procedia engineering,2017,192:195~200.
[3]单紫薇.基于机器视觉测量的齿轮图像边界提取算法研究[D].沈阳:沈阳工业大学,2017.Shan Ziwei.Research of gear image boundary extraction algorithm based on machine vision measurement[D].Shenyang:Shenyang University of Technology,2017.
[4]Shi Tian,Kong Jianyi,Wang Xingdong,et al.Improved Roberts operator for detecting surface defects of heavy rails with superior precision and efficiency[J].High Technology Letters,2016,22(2):207~214.
[5]Kumar B M,Ratnam M M.Machine vision method for non-contact measurement of surface roughness of a rotating workpiece[J].Sensor review,2015,35(1):10~19.
[6]潘震,方宇,茅健,等.基于关节臂式坐标测量机的转向架构架尺寸测量方法[J].上海工程技术大学学报,2016,30(4):311~315.Pan Zhen,Fang Yu,Mao Jian,et al.Dimensional measurement method of bogie frame based on articulated arm coordinate measuring machines[J].Journal of Shanghai University of Engineering Science,2016,30(4):311~315.
[7]师蒙蒙,韩焱,王鉴,等.基于图像的大型物体尺寸的三维测量方法研究[J].计算机测量与控制,2011,19(10):2351~2353.Shi Mengmeng,Han Yan,Wang Jian,et al.Research on 3D large object size based on measurement method of image[J].Computer Measurement&Control,2011,19(10):2351~2353.
[8]刘刚.基于阵列相机的图像测量技术研究[D].北京:北方工业大学,2017.Liu Gang.Image measurement technology research based on the array camera[D].Beijing:North China University of Technology,2017.
[9]王俊杰.基于图像的尺寸测量算法研究[D].郑州:郑州大学,2013.Wang Junjie.Algorithm study on dimension measurement based on image[D].Zhengzhou:Zhengzhou University,2013.
[10]Escamilla G A P,Kobayashi F,Otani Y.Three-dimensional surface measurement based on the projected defocused pattern technique using imaging fiber optics[J].Optics Communications,2017,390:57~60.
[11]Jiang Y,Akkus A,Roperto R,et al.Measurement of Jintegral in CAD/CAM dental ceramics and composite resin by digital image correlation[J].Journal of the mechanical behavior of biomedical materials,2016,62:240~246.
[12]何雪明,李咏平,武美萍,等.基于三坐标测量机双参数向自适应测量自由曲面[J].中国机械工程,2017,28(5):525~531.He Xueming,Li Yongping,Wu Meiping,et al.Adaptive measurement of free-form surfaces based on CMM double parameter directions[J].China Mechanical Engineering,2017,28(5):525~531.
[13]李钊宝,汪新新,贾茜媛.基于机器视觉的工件尺寸高精度测量方法[J].计算机测量与控制,2015,23(6):1866~1870.Li Zhaobao,Wang Xinxin,Jia Xiyuan.A method of high-precision machine vision measurement of size of workpiece[J].Computer Measurement&Control,2015,23(6):1866~1870.
[14]Shuxia G,Jiancheng Z,Xiaofeng J,et al.Mini milling cutter measurement based on machine vision[J].Procedia Engineering,2011,15:1807~1811.
[15]Gao Z,Zhang Q,Su Y,et al.Accuracy evaluation of optical distortion calibration by digital image correlation[J].Optics and Lasers in Engineering,2017,98:143~152.
[16]Xiaobin L,Yushun W,Lihong F.Monitoring lettuce growth using K-means color image segmentation and principal component analysis method[J].Transactions of the Chinese Society of Agricultural Engineering,2016,32(12):179~186.
[17]李晓斌,郭玉明,崔清亮,等.用图像法分析茄子在冻干过程中的水分动态运移规律[J].农业工程学报,2016,32(1):304~311.Li Xiaobin,Guo Yuming,Cui Qingliang,et al.Moisture diffusion and transfer dynamic analysis of eggplant during vacuum freeze-drying based on image processing technique[J].Transactions of the Chinese Society of Agricultural Engineering,2016,32(1):304~311.
[18]李晓飞,梅树立,李丽.基于Meyer窗函数剪切波变换的蝗虫切片图像去噪[J].农业机械学报,2016,47(S1):449~456.Li Xiaofei,Mei Shuli,Li Li.Image denoising of locust slices based on Meyer window function shearlet transform[J].Transactions of the Chinese Society for Agricultural Machinery,2016,47(S1):449~456.
[19]Liu J,Liu R,Wang Y,et al.Image denoising searching similar blocks along edge directions[J].Signal Processing:Image Communication,2017,57:33~45.
[20]Zhang X.Image denoising using local Wiener filter and its method noise[J].Optik-International Journal for Light and Electron Optics,2016,127(17):6821~6828.
[21]Oliveira P C,Moura J P,Fernandes L F,et al.A non-destructive method based on digital image processing for calculate the vigor and the vegetative expression of vines[J].Computers and Electronics in Agriculture,2016,124:289~294.
[22]Guo Y,Zhu W,Jiao P,et al.Multi-object extraction from topview group-housed pig images based on adaptive partitioning and multilevel thresholding segmentation[J].Biosystems Engineering,2015,135:54~60.
[23]于起峰,尚洋.摄像测量学原理与应用研究[M].北京:科学出版社,2009.
[24]张法全,王国富,曾庆宁,等.利用重心原理的图像目标最小外接矩形快速算法[J].红外与激光工程,2013,42(5):1382~1387.Zhang Faquan,Wang Guofu,Zeng Qingning,et al.New algorithm for minimum enclosing rectangle of the object in the image region based on center-of-gravity principle[J].Infrared and Laser Engineering,2013,42(5):1382~1387.
[25]李晓斌.运用图像处理技术在线监测真空冻干果蔬含水率[D].太谷:山西农业大学,2013.Li Xiaobin.On-line monitoring of moisture ratio for fruits and vegetables during vacuum freeze-drying based on image processing technique[D].Taigu:Shanxi Agricultural University,2013.
[2]FlorkováZ,Jambor M.Quantification of aggregate surface texture based on three dimensional microscope measurement[J].Procedia engineering,2017,192:195~200.
[3]单紫薇.基于机器视觉测量的齿轮图像边界提取算法研究[D].沈阳:沈阳工业大学,2017.Shan Ziwei.Research of gear image boundary extraction algorithm based on machine vision measurement[D].Shenyang:Shenyang University of Technology,2017.
[4]Shi Tian,Kong Jianyi,Wang Xingdong,et al.Improved Roberts operator for detecting surface defects of heavy rails with superior precision and efficiency[J].High Technology Letters,2016,22(2):207~214.
[5]Kumar B M,Ratnam M M.Machine vision method for non-contact measurement of surface roughness of a rotating workpiece[J].Sensor review,2015,35(1):10~19.
[6]潘震,方宇,茅健,等.基于关节臂式坐标测量机的转向架构架尺寸测量方法[J].上海工程技术大学学报,2016,30(4):311~315.Pan Zhen,Fang Yu,Mao Jian,et al.Dimensional measurement method of bogie frame based on articulated arm coordinate measuring machines[J].Journal of Shanghai University of Engineering Science,2016,30(4):311~315.
[7]师蒙蒙,韩焱,王鉴,等.基于图像的大型物体尺寸的三维测量方法研究[J].计算机测量与控制,2011,19(10):2351~2353.Shi Mengmeng,Han Yan,Wang Jian,et al.Research on 3D large object size based on measurement method of image[J].Computer Measurement&Control,2011,19(10):2351~2353.
[8]刘刚.基于阵列相机的图像测量技术研究[D].北京:北方工业大学,2017.Liu Gang.Image measurement technology research based on the array camera[D].Beijing:North China University of Technology,2017.
[9]王俊杰.基于图像的尺寸测量算法研究[D].郑州:郑州大学,2013.Wang Junjie.Algorithm study on dimension measurement based on image[D].Zhengzhou:Zhengzhou University,2013.
[10]Escamilla G A P,Kobayashi F,Otani Y.Three-dimensional surface measurement based on the projected defocused pattern technique using imaging fiber optics[J].Optics Communications,2017,390:57~60.
[11]Jiang Y,Akkus A,Roperto R,et al.Measurement of Jintegral in CAD/CAM dental ceramics and composite resin by digital image correlation[J].Journal of the mechanical behavior of biomedical materials,2016,62:240~246.
[12]何雪明,李咏平,武美萍,等.基于三坐标测量机双参数向自适应测量自由曲面[J].中国机械工程,2017,28(5):525~531.He Xueming,Li Yongping,Wu Meiping,et al.Adaptive measurement of free-form surfaces based on CMM double parameter directions[J].China Mechanical Engineering,2017,28(5):525~531.
[13]李钊宝,汪新新,贾茜媛.基于机器视觉的工件尺寸高精度测量方法[J].计算机测量与控制,2015,23(6):1866~1870.Li Zhaobao,Wang Xinxin,Jia Xiyuan.A method of high-precision machine vision measurement of size of workpiece[J].Computer Measurement&Control,2015,23(6):1866~1870.
[14]Shuxia G,Jiancheng Z,Xiaofeng J,et al.Mini milling cutter measurement based on machine vision[J].Procedia Engineering,2011,15:1807~1811.
[15]Gao Z,Zhang Q,Su Y,et al.Accuracy evaluation of optical distortion calibration by digital image correlation[J].Optics and Lasers in Engineering,2017,98:143~152.
[16]Xiaobin L,Yushun W,Lihong F.Monitoring lettuce growth using K-means color image segmentation and principal component analysis method[J].Transactions of the Chinese Society of Agricultural Engineering,2016,32(12):179~186.
[17]李晓斌,郭玉明,崔清亮,等.用图像法分析茄子在冻干过程中的水分动态运移规律[J].农业工程学报,2016,32(1):304~311.Li Xiaobin,Guo Yuming,Cui Qingliang,et al.Moisture diffusion and transfer dynamic analysis of eggplant during vacuum freeze-drying based on image processing technique[J].Transactions of the Chinese Society of Agricultural Engineering,2016,32(1):304~311.
[18]李晓飞,梅树立,李丽.基于Meyer窗函数剪切波变换的蝗虫切片图像去噪[J].农业机械学报,2016,47(S1):449~456.Li Xiaofei,Mei Shuli,Li Li.Image denoising of locust slices based on Meyer window function shearlet transform[J].Transactions of the Chinese Society for Agricultural Machinery,2016,47(S1):449~456.
[19]Liu J,Liu R,Wang Y,et al.Image denoising searching similar blocks along edge directions[J].Signal Processing:Image Communication,2017,57:33~45.
[20]Zhang X.Image denoising using local Wiener filter and its method noise[J].Optik-International Journal for Light and Electron Optics,2016,127(17):6821~6828.
[21]Oliveira P C,Moura J P,Fernandes L F,et al.A non-destructive method based on digital image processing for calculate the vigor and the vegetative expression of vines[J].Computers and Electronics in Agriculture,2016,124:289~294.
[22]Guo Y,Zhu W,Jiao P,et al.Multi-object extraction from topview group-housed pig images based on adaptive partitioning and multilevel thresholding segmentation[J].Biosystems Engineering,2015,135:54~60.
[23]于起峰,尚洋.摄像测量学原理与应用研究[M].北京:科学出版社,2009.
[24]张法全,王国富,曾庆宁,等.利用重心原理的图像目标最小外接矩形快速算法[J].红外与激光工程,2013,42(5):1382~1387.Zhang Faquan,Wang Guofu,Zeng Qingning,et al.New algorithm for minimum enclosing rectangle of the object in the image region based on center-of-gravity principle[J].Infrared and Laser Engineering,2013,42(5):1382~1387.
[25]李晓斌.运用图像处理技术在线监测真空冻干果蔬含水率[D].太谷:山西农业大学,2013.Li Xiaobin.On-line monitoring of moisture ratio for fruits and vegetables during vacuum freeze-drying based on image processing technique[D].Taigu:Shanxi Agricultural University,2013.