一种环状编码标记点的设计及解码算法
|
摘要
针对近景摄影测量中对编码标志点的精确定位和准确识别的要求,提出一种环状编码标记点的设计和识别算法。在传统环状编码标记点的基础上添加三个定位符,用于确定标志点的精确位置和增加标志点的数量。解码时先检测定位符坐标及其在标志点中的位置,然后对编码标志点进行透视变换以实现图像校正的目的,最后用提出的基于圆环扫描的方法进行解码。实验结果表明,该算法对任意旋转角度下的编码标志点均有较好的检测识别效果;当摄像机与标记平面的夹角小于65°时,其识别准确率可达99. 3%;在复杂背景情况下的平均识别准确率为97. 4%,误识别率为1. 25%,识别平均速率为2. 15 s/幅。
On the requests of coded targets in close range photogrammetry for precision location and accurate recognition,this paper proposed a designing and decoding algorithm of circular coded target. This algorithm added three locators on the basis of traditional circular coded target to achieve precision position and increase the number of targets. Firstly,the algorithm detected the locator coordinates and their positions in the targets. Then,it used the perspective transformation of the coded target to achieve image distortion correction. Finally,this paper used the method based on ring scan to achieve the decoding of the circular coded target. Experimental results show that the proposed algorithm has good detection and recognition results for all the coded targets which rotate any angle. When the angle between the camera and the marking plane is less than 65°,the recognition accuracy rate can reach 99. 3%,the average recognition accuracy rate is 97. 4%,false recognition rate is 1. 25%,and the recognition rate is 2. 15 s per picture under the complex background.
On the requests of coded targets in close range photogrammetry for precision location and accurate recognition,this paper proposed a designing and decoding algorithm of circular coded target. This algorithm added three locators on the basis of traditional circular coded target to achieve precision position and increase the number of targets. Firstly,the algorithm detected the locator coordinates and their positions in the targets. Then,it used the perspective transformation of the coded target to achieve image distortion correction. Finally,this paper used the method based on ring scan to achieve the decoding of the circular coded target. Experimental results show that the proposed algorithm has good detection and recognition results for all the coded targets which rotate any angle. When the angle between the camera and the marking plane is less than 65°,the recognition accuracy rate can reach 99. 3%,the average recognition accuracy rate is 97. 4%,false recognition rate is 1. 25%,and the recognition rate is 2. 15 s per picture under the complex background.
引文
[1]Moriyama T,Kochi N,Yamada M,et al.Automation target identification with the color-coded-target[C]//Proc of International Archives of Photogrammetry,Remote Sensing and Spatial Information Sciences.2008:39-44.
[2]杨剑,韩建栋,秦品乐.视觉测量中可纠错的编码点识别及提取[J].光学精密工程,2012,20(10):2293-2299.(Yang Jian,Han Jiandong,Qin Pingle.Correcting error on recognition of coded points for photogrammetry[J].Optics and Precision Engineering,2012,20(10):2293-2299.)
[3]Schneider C T,Sinnreic H K.Optical 3D measurement systems for quality control in industry[EB/OL].(2017-01-18).http://www.isprs.org/proceedings/XXIX/congress/part5/56_XXIX-part5.pdf.
[4]胡浩,梁晋,唐正宗,等.大视场多像机视频测量系统的全局标定[J].光学精密工程,2012,20(2):369-378.(Hu Hao,Liang Jin,Tang Zhengzong,et al.An accurate global calibration method for multi-camera videogrammetric system with large-scale field-of-view[J].Optics and Precision Engineering,2012,20(2):369-378.)
[5]解则晓,高翔,朱瑞新.环状编码标记点的高效提取与鲁棒识别算法[J].光电子·激光,2015,26(3):559-566.(Xie Zexiao,Gao Xiang,Zhu Ruixin.Efficient extraction and robust recognition algorithm of circular code target[J].Journal of Optoelectronics·Laser,2015,26(3):559-566.)
[6]宋丽梅,陈昌曼,陈卓,等.环状编码标记点的检测与识别[J].光学精密工程,2013,21(12):3239-3247.(Song Limei,Chen Chang man,Chen Zhuo,et al.Detection and recognition of cyclic coded targets[J].Optics and Precision Engineering,2013,21(12):3239-3247.)
[7]Hu Hao,Liang Jin,Xiao Zhenzhong,et al.A four-camera videogrammetric system for 3D motion measurement of deformable object[J].Optics and Lasers in Engineering,2012,50(5):800-811.
[8]宋凤菲.彩色图像灰度化及其效果的客观评价方法研究[D].泉州:华侨大学,2014.(Song Fengfei.Color images to grayscale conversions and its objective evaluation method research[D].Quanzhou:Huaqiao University,2014.)
[9]Hou Jiali,Li Baokui.An improved algorithm for horizon detection based on OSTU[C]//Proc of the 7th International Conference on Intelligent Human-Machine Systems and Cybernetics.Washington DC:IEEE Computer Society,2015:414-417.
[10]程增木,王悦宏,程琳琳.基于逆透视变换的道路检测技术[J].电子科学技术,2016,3(2):113-116.(Cheng Zengmu,Wang Yuehong,Cheng Linlin.Technology of road detection based on inverse perspective transformation[J].Electronic Science&Technology,2016,3(2):113-116.)
[11]周玲,张丽艳,郑建冬,等.近景摄影测量中标记点的自动检测[J].应用科学学报,2007,25(3):288-294.(Zhou Ling,Zhang Liyan,Zheng Jiandong,et al.Automated reference point detection in close range phoyogrammetry[J].Journal of Applied Sciences,2007,25(3):288-294.)
[12]Doring T,Meysel F,Reulke R.Autonomous calibration of moving line scanners with coded photogrammetric targets recognition[C]//Proc of ISPRS Commission V Symposium on Image Engineering and Vision Metrology.2006:84-89.
[2]杨剑,韩建栋,秦品乐.视觉测量中可纠错的编码点识别及提取[J].光学精密工程,2012,20(10):2293-2299.(Yang Jian,Han Jiandong,Qin Pingle.Correcting error on recognition of coded points for photogrammetry[J].Optics and Precision Engineering,2012,20(10):2293-2299.)
[3]Schneider C T,Sinnreic H K.Optical 3D measurement systems for quality control in industry[EB/OL].(2017-01-18).http://www.isprs.org/proceedings/XXIX/congress/part5/56_XXIX-part5.pdf.
[4]胡浩,梁晋,唐正宗,等.大视场多像机视频测量系统的全局标定[J].光学精密工程,2012,20(2):369-378.(Hu Hao,Liang Jin,Tang Zhengzong,et al.An accurate global calibration method for multi-camera videogrammetric system with large-scale field-of-view[J].Optics and Precision Engineering,2012,20(2):369-378.)
[5]解则晓,高翔,朱瑞新.环状编码标记点的高效提取与鲁棒识别算法[J].光电子·激光,2015,26(3):559-566.(Xie Zexiao,Gao Xiang,Zhu Ruixin.Efficient extraction and robust recognition algorithm of circular code target[J].Journal of Optoelectronics·Laser,2015,26(3):559-566.)
[6]宋丽梅,陈昌曼,陈卓,等.环状编码标记点的检测与识别[J].光学精密工程,2013,21(12):3239-3247.(Song Limei,Chen Chang man,Chen Zhuo,et al.Detection and recognition of cyclic coded targets[J].Optics and Precision Engineering,2013,21(12):3239-3247.)
[7]Hu Hao,Liang Jin,Xiao Zhenzhong,et al.A four-camera videogrammetric system for 3D motion measurement of deformable object[J].Optics and Lasers in Engineering,2012,50(5):800-811.
[8]宋凤菲.彩色图像灰度化及其效果的客观评价方法研究[D].泉州:华侨大学,2014.(Song Fengfei.Color images to grayscale conversions and its objective evaluation method research[D].Quanzhou:Huaqiao University,2014.)
[9]Hou Jiali,Li Baokui.An improved algorithm for horizon detection based on OSTU[C]//Proc of the 7th International Conference on Intelligent Human-Machine Systems and Cybernetics.Washington DC:IEEE Computer Society,2015:414-417.
[10]程增木,王悦宏,程琳琳.基于逆透视变换的道路检测技术[J].电子科学技术,2016,3(2):113-116.(Cheng Zengmu,Wang Yuehong,Cheng Linlin.Technology of road detection based on inverse perspective transformation[J].Electronic Science&Technology,2016,3(2):113-116.)
[11]周玲,张丽艳,郑建冬,等.近景摄影测量中标记点的自动检测[J].应用科学学报,2007,25(3):288-294.(Zhou Ling,Zhang Liyan,Zheng Jiandong,et al.Automated reference point detection in close range phoyogrammetry[J].Journal of Applied Sciences,2007,25(3):288-294.)
[12]Doring T,Meysel F,Reulke R.Autonomous calibration of moving line scanners with coded photogrammetric targets recognition[C]//Proc of ISPRS Commission V Symposium on Image Engineering and Vision Metrology.2006:84-89.