基于齿轮局部图像的齿距机器视觉测量方法
|
摘要
针对工业对中小模数直齿圆柱齿距快速测量需求及视觉测量特点,提出一种基于齿廓图像边缘过渡带信息统计的单个齿距算法。该算法首先采用双阈值法提取齿轮齿廓边缘过渡带像素信息,然后根据齿轮渐开线几何关系,将过渡带像素信息逆向映射到基圆上,计算得到最优的齿廓边缘渐开线初始相位角,最后利用两条相邻同名齿廓初始相位角计算得出齿距。通过采用高精度量块组合边缘测量试验,验证了该算法的原理正确性和测量精度。结果表明,利用该算法视觉测量得到的相对位置最大偏差为0.002 1 mm,最大分散度为0.000 52 mm。对同一5级精度齿轮进行齿距测量,视觉齿轮测量仪和M&M3525齿轮测量中心测量的最大单个齿距偏差出现在相同齿距上,二者相差0.000 7 mm,其齿距累积总偏差相差0.001 mm,表明本齿轮齿距视觉测量方法可以满足5级精度直齿圆柱齿轮齿距的快速测量要求。
Aimed at the high efficient industial measurement demands of small and medium modulus spur gear pitch and the visional image feature,a new single pitch algorithm is proposed based on calculating the tooth image pixel information in gear edge transition zone. The algorithm first extracts the tooth image pixel information in gear edge transition zone with double threshold method. According to the involute geometric characteristics,this image pixel information is adversely mapped to the gear base circle. And the optimal initial phase angle of tooth profile edge involute is calculated. Finally,the gear pitch is obtained by using the initial phase angle of two adjacent corresponding tooth profile. The principle and measuring accuracy of the algorithm are verified by measuring experiments of combined edge obtained with high precision gauges. The results show that the maximum deviation of relative location using this algorithm vesical measurement is 0. 002 1 mm,and its maximum dispersity is 0. 000 52 mm. By measuring a same 5-level accuracy gear pitch,the gear maximum single pitch deviation obtained by visual measuring instrument occurs in the pitch same to that obtained by M&M3525 gear measuring center. The difference between two results is 0. 000 7 mm and the difference of total cumulative deviation obtained by two methods is 0. 001 mm. These experimental results indicate that the pitch visual measuring method can meet the requirement of high efficient measurement for 5-level accuracy vertical cylinder gear pitch.
Aimed at the high efficient industial measurement demands of small and medium modulus spur gear pitch and the visional image feature,a new single pitch algorithm is proposed based on calculating the tooth image pixel information in gear edge transition zone. The algorithm first extracts the tooth image pixel information in gear edge transition zone with double threshold method. According to the involute geometric characteristics,this image pixel information is adversely mapped to the gear base circle. And the optimal initial phase angle of tooth profile edge involute is calculated. Finally,the gear pitch is obtained by using the initial phase angle of two adjacent corresponding tooth profile. The principle and measuring accuracy of the algorithm are verified by measuring experiments of combined edge obtained with high precision gauges. The results show that the maximum deviation of relative location using this algorithm vesical measurement is 0. 002 1 mm,and its maximum dispersity is 0. 000 52 mm. By measuring a same 5-level accuracy gear pitch,the gear maximum single pitch deviation obtained by visual measuring instrument occurs in the pitch same to that obtained by M&M3525 gear measuring center. The difference between two results is 0. 000 7 mm and the difference of total cumulative deviation obtained by two methods is 0. 001 mm. These experimental results indicate that the pitch visual measuring method can meet the requirement of high efficient measurement for 5-level accuracy vertical cylinder gear pitch.
引文
[1]王江明,余烨,金强,等.多梯度融合的RGBD图像边缘检测[J].电子测量与仪器学报,2017,31(3):436-442.WANG J M,YU Y,JIN Q,et al.RGBD images edge detection basedon multi-gradient fusion[J].Journal of Electronic Measurement and Instrumentation,2017,31(3):436-442.
[2]GOCH G.Gear metrology[J].CIRP AnnalsManufacturing Technology,2003,52(2):659-695.
[3]凌四营,王立鼎,娄志峰,等.超精密齿距偏差测量中测点位置的误差补偿[J].仪器仪表学报,2014,35(3):691-696.LING S Y,WANG L D,LOU ZH F,et al.Error compensation of the measurement position in ultra precision gear pitch deviation measurement[J].Chinese Journal of Scientific Instrument,2014,35(3):691-696.
[4]王静,王海亮,向茂生,等.基于非极大值抑制的圆目标亚像素中心定位[J].仪器仪表学报,2012,33(7):1460-1468.WANG J,WANG H L,XIANG M SH,et al.Subpixelaccuracy central location of circle target based on nonmaximum suppression[J].Chinese Journal of Scientific Instrument,2012,33(7):1460-1468.
[5]李智勇,余小萍,姜真,等.基于蛙眼检测模型的双阈值判定目标跟踪方法[J].电子测量与仪器学报,2015,29(2):187-195.LI ZH Y,YU X P,JIANG ZH,et al.Dual-threshold determination tracking method based on frog visual detection model[J].Journal of Electronic Measurement and Instrumentation,2015,29(2):187-195.
[6]陈鑫元,李筠,杨海马,等.自适应阈值图像二值化及形态学处理的FPGA实现[J].电子测量技术,2016,39(7):67-71.CHEN X Y,LI J,YANG H M,et al.Adaptive threshold binarization and morphological image processing besed on FPGA[J].Electronic Measurement Technology,2016,39(7):67-71.
[7]贺忠海,王宝光,廖怡白,等.利用曲线拟合方法的亚像素提取算法[J].仪器仪表学报,2003,24(2):195-197.HE ZH H,WANG B G,LIAO Y B,et al.Sub-pixel extraction algorithmusing curve fitting method[J].Journal of Scientific Instrument,2003,24(2):195-197.
[8]田子怡,李云红.基于多尺度Gabor滤波器的角点检测[J].国外电子测量技术,2016,35(7):75-79,84.TIAN ZI Y,LI Y H.Corner detection using multi-scale Gabor filters[J].Foreign Electronic Measurement Technology,2016,35(7):75-79,84.
[9]舒彬.基于小波变换的边缘检测算法研究[J].自动化与仪器仪表,2016(8):163-164.SHU B.Research on edge detection algorithm based on Wavelet Transform[J].Automation&Instrumentation,2016(8):163-164.
[10]汤洁,石照耀.齿轮双面啮合多维测量系统的动态性[J].北京工业大学学报,2012,38(11):1633-1637.TANG J,SHI ZH Y.Dynamic characteristics of double flank gear rolling test with many degrees of freedom[J].Journal of Beijing University of Technology,2012,38(11):1633-1637.
[11]段振云,王宁,赵文辉,等.基于点阵标定板的视觉测量系统标定方法研究[J].光学学报,2016,36(5):143-151.DUAN ZH Y,WANG N,ZHAO W H,ZHAO W H.Study on calibration method based on lattice calibration plate invision measurement system[J].Acta Optica Sinica,2016,36(5):143-151.
[12]王宁,段振云,赵文辉,等.视觉检测中齿轮外轮廓分段方法研究[J].组合机床与自动化加工技术,2016(4):117-120.WANG N,DUAN ZH Y,ZHAO W H,et al.Researchon method of gear outer contour segmentation in vision measu-rement[J].Modular Machine Tool&Automatic Manufacturing Technique,2016(4):117-120.
[13]段振云,王宁,赵文珍,等.基于高斯积分曲面拟合的亚像素边缘定位算法[J].仪器仪表学报,2017,38(1):219-225.DUAN ZH Y,WANG N,ZHAO W ZH,et al.Sub-pixel edge location algorithm based on Gauss integral curved surface fitting[J].Chinese Journal of Scientific Instrument,2017,38(1):219-225.
[14]王笑一,石照耀,林家春.基于全齿廓信息的齿距偏差快速测量方法[J].仪器仪表学报,2016,37(10):2202-2210.WANG X Y,SHI ZH Y,LIN J CH.Fast measurement method for pitch deviation based on full toothprofile information[J].Chinese Journal of Scientific Instrument,2016,37(10):2202-2210.
[15]SCHNEIDER S C,RUPITSCH S J,ZAGAR B G.Signal processing for laser speckle strain-measurement techniques[J].IEEE Transactions on Instrumentation and Measurement,2007,56(6):2681-2687.
[16]王静,王海亮,向茂生,等.基于非极大值抑制的圆目标亚像素中心定位[J].仪器仪表学报,2012,33(7):1460-1468.WANG J,WANG H L,XIANG M SH,et al.Sub-pixel accuracy central location of circle target based on nonmaximum suppression[J].Chinese Journal of Scientific Instrument,2012,33(7):1460-1468.
[2]GOCH G.Gear metrology[J].CIRP AnnalsManufacturing Technology,2003,52(2):659-695.
[3]凌四营,王立鼎,娄志峰,等.超精密齿距偏差测量中测点位置的误差补偿[J].仪器仪表学报,2014,35(3):691-696.LING S Y,WANG L D,LOU ZH F,et al.Error compensation of the measurement position in ultra precision gear pitch deviation measurement[J].Chinese Journal of Scientific Instrument,2014,35(3):691-696.
[4]王静,王海亮,向茂生,等.基于非极大值抑制的圆目标亚像素中心定位[J].仪器仪表学报,2012,33(7):1460-1468.WANG J,WANG H L,XIANG M SH,et al.Subpixelaccuracy central location of circle target based on nonmaximum suppression[J].Chinese Journal of Scientific Instrument,2012,33(7):1460-1468.
[5]李智勇,余小萍,姜真,等.基于蛙眼检测模型的双阈值判定目标跟踪方法[J].电子测量与仪器学报,2015,29(2):187-195.LI ZH Y,YU X P,JIANG ZH,et al.Dual-threshold determination tracking method based on frog visual detection model[J].Journal of Electronic Measurement and Instrumentation,2015,29(2):187-195.
[6]陈鑫元,李筠,杨海马,等.自适应阈值图像二值化及形态学处理的FPGA实现[J].电子测量技术,2016,39(7):67-71.CHEN X Y,LI J,YANG H M,et al.Adaptive threshold binarization and morphological image processing besed on FPGA[J].Electronic Measurement Technology,2016,39(7):67-71.
[7]贺忠海,王宝光,廖怡白,等.利用曲线拟合方法的亚像素提取算法[J].仪器仪表学报,2003,24(2):195-197.HE ZH H,WANG B G,LIAO Y B,et al.Sub-pixel extraction algorithmusing curve fitting method[J].Journal of Scientific Instrument,2003,24(2):195-197.
[8]田子怡,李云红.基于多尺度Gabor滤波器的角点检测[J].国外电子测量技术,2016,35(7):75-79,84.TIAN ZI Y,LI Y H.Corner detection using multi-scale Gabor filters[J].Foreign Electronic Measurement Technology,2016,35(7):75-79,84.
[9]舒彬.基于小波变换的边缘检测算法研究[J].自动化与仪器仪表,2016(8):163-164.SHU B.Research on edge detection algorithm based on Wavelet Transform[J].Automation&Instrumentation,2016(8):163-164.
[10]汤洁,石照耀.齿轮双面啮合多维测量系统的动态性[J].北京工业大学学报,2012,38(11):1633-1637.TANG J,SHI ZH Y.Dynamic characteristics of double flank gear rolling test with many degrees of freedom[J].Journal of Beijing University of Technology,2012,38(11):1633-1637.
[11]段振云,王宁,赵文辉,等.基于点阵标定板的视觉测量系统标定方法研究[J].光学学报,2016,36(5):143-151.DUAN ZH Y,WANG N,ZHAO W H,ZHAO W H.Study on calibration method based on lattice calibration plate invision measurement system[J].Acta Optica Sinica,2016,36(5):143-151.
[12]王宁,段振云,赵文辉,等.视觉检测中齿轮外轮廓分段方法研究[J].组合机床与自动化加工技术,2016(4):117-120.WANG N,DUAN ZH Y,ZHAO W H,et al.Researchon method of gear outer contour segmentation in vision measu-rement[J].Modular Machine Tool&Automatic Manufacturing Technique,2016(4):117-120.
[13]段振云,王宁,赵文珍,等.基于高斯积分曲面拟合的亚像素边缘定位算法[J].仪器仪表学报,2017,38(1):219-225.DUAN ZH Y,WANG N,ZHAO W ZH,et al.Sub-pixel edge location algorithm based on Gauss integral curved surface fitting[J].Chinese Journal of Scientific Instrument,2017,38(1):219-225.
[14]王笑一,石照耀,林家春.基于全齿廓信息的齿距偏差快速测量方法[J].仪器仪表学报,2016,37(10):2202-2210.WANG X Y,SHI ZH Y,LIN J CH.Fast measurement method for pitch deviation based on full toothprofile information[J].Chinese Journal of Scientific Instrument,2016,37(10):2202-2210.
[15]SCHNEIDER S C,RUPITSCH S J,ZAGAR B G.Signal processing for laser speckle strain-measurement techniques[J].IEEE Transactions on Instrumentation and Measurement,2007,56(6):2681-2687.
[16]王静,王海亮,向茂生,等.基于非极大值抑制的圆目标亚像素中心定位[J].仪器仪表学报,2012,33(7):1460-1468.WANG J,WANG H L,XIANG M SH,et al.Sub-pixel accuracy central location of circle target based on nonmaximum suppression[J].Chinese Journal of Scientific Instrument,2012,33(7):1460-1468.