浏览全部资源
扫码关注微信
1.中国科学院 微电子研究所,北京 100094
2.中国科学院大学,北京 100049
[ "李 洋(1988-),男,河北邢台人,博士研究生,2006年于重庆大学获得学士学位,2010年于中国航空研究院获得硕士学位,主要从事光学精密测量及机器视觉方面的研究。E-mail: liyang@ime.ac.cn" ]
[ "周维虎(1962-),男,安徽无为人,博士,研究员,博士生导师,1983年、2000年于合肥工业大学分别获得学士、博士学位,现为中国科学院微电子研究所光电研发中心主任,主要从事光电检测、光电系统总体设计与集成测试、光电精密测量技术与仪器等方面的研究。E-mail:zhouweihu@ime.ac.cn" ]
收稿日期:2021-01-14,
修回日期:2021-03-16,
纸质出版日期:2021-08-15
移动端阅览
李洋,程智,周维虎等.面向工业复杂场景的合作靶标椭圆特征快速鲁棒检测[J].光学精密工程,2021,29(08):1910-1920.
LI Yang,CHENG Zhi,ZHOU Wei-hu,et al.A fast and robust method for detecting elliptical character of cooperative targets in industrial complex background[J].Optics and Precision Engineering,2021,29(08):1910-1920.
李洋,程智,周维虎等.面向工业复杂场景的合作靶标椭圆特征快速鲁棒检测[J].光学精密工程,2021,29(08):1910-1920. DOI: 10.37188/OPE.20212908.1910.
LI Yang,CHENG Zhi,ZHOU Wei-hu,et al.A fast and robust method for detecting elliptical character of cooperative targets in industrial complex background[J].Optics and Precision Engineering,2021,29(08):1910-1920. DOI: 10.37188/OPE.20212908.1910.
针对工业制造现场复杂场景下典型合作靶标识别需求,提出了一种椭圆特征快速鲁棒检测方法。首先对预处理图像进行边缘跟踪和弧段分割筛选分组,通过基于帕斯卡定理的邻接象限弧段匹配和基于关系矩阵的边缘弧段聚类,对具有椭圆匹配性的边缘弧段实现快速聚类。然后采用非迭代最小二乘拟合方法获取椭圆参数,并根据参数特征去伪,得到最终检测结果。本文通过采用更加严格的椭圆匹配约束和更加高效的弧段聚类方法,有效减少了参数拟合阶段对非椭圆特征的运算量,具有良好的检测效率和检测结果可靠性。实验结果表明,本文算法对检测距离、光照条件、目标方位、图像噪声等干扰因素具有显著的抑制作用,在复杂场景条件下能够得到可靠的检测结果。本文方法对640×480 pixel的实验图像检测时间为183.2 ms,快于同类椭圆检测算法,满足了工业测量仪器在现场复杂场景中对合作靶标快速搜索与识别的性能需求。
A robust ellipse detection method is proposed herein to meet the demands for recognizing typical cooperative targets in industrial complex backgrounds. First, edge segments are derived from discrete edge points of a preprocessed image by applying an edge tracing algorithm, which is followed by steps of segmentation, filtering, and grouping. Next, a Pascal’s theorem based matching method for arcs of adjacency quadrants and a relationship matrix based arc clustering method are proposed herein to realize fast clustering according to the elliptic matching quality. Then, a non-iterative least square fitting method is used to obtain the parameters of ellipses, and the final result is achieved by eliminating false ellipses based on their parameters. The proposed method performs well in terms of the detection efficiency and robustness owing to the application of a stricter elliptical matching constraint, and a more effective arc clustering method reduces the computational complexity of the parameter fitting step by eliminating non-elliptical datasets. The ellipse fitting efficiency increases, and a more exact and reliable result is acquired. Test results show that the proposed detection method is insensitive to interference factors such as distance, illumination, target’s orientation, and noise. The detection time for the proposed method is less than that for reference algorithms for the test image; the detection time is 183.2 ms for a 640×480 pixel image. The method shows great potential for industrial measuring instruments to search for collaboration targets in complex backgrounds.
王博 , 董登峰 , 周维虎 , 等 . 面向激光跟踪仪跟踪恢复的合作目标视觉检测 [J]. 光学 精密工程 , 2020 , 28 ( 2 ): 271 - 282 .
WANG B , DONG D F , ZHOU W H , et al . Visual detection of targetball for laser tracker target tracking recovery [J]. Opt. Precision Eng. , 2020 , 28 ( 2 ): 271 - 282 . (in Chinese)
余海涛 , 施光林 . 基于椭圆检测算法的位姿快速测量技术 [J]. 传感器与微系统 , 2020 , 39 ( 5 ): 122 - 124, 128 .
YU H T , SHI G L . Pose fast measurement technology based on ellipse detection algorithm [J]. Transducer and Microsystem Technologies , 2020 , 39 ( 5 ): 122 - 124, 128 . (in Chinese)
XU P , CHEUNG B C F , LI B . A complete, continuous, and minimal product of exponentials-based model for five-axis machine tools calibration with a single laser tracker, an R-test, or a double ball-bar [J]. Journal of Manufacturing Science and Engineering , 2019 , 141 ( 4 ): 041010 .
岳颖 , 李丛 , 张勇 , 等 . 全站仪测量系统在船舶建造中的应用 [J]. 测绘科学技术学报 , 2020 , 37 ( 3 ): 258 - 263 .
YUE Y , LI C , ZHANG Y , et al . Application of total station measurement system in ship construction [J]. Journal of Geomatics Science and Technology , 2020 , 37 ( 3 ): 258 - 263 . (in Chinese)
解则晓 , 王晓敏 . 平面标定靶标标记点的圆心提取 [J]. 光学 精密工程 , 2019 , 27 ( 2 ): 440 - 449 .
XIE Z X , WANG X M . Research on extraction algorithm of projected circular centers of marked points on the planar calibration targets [J]. Opt. Precision Eng. , 2019 , 27 ( 2 ): 440 - 449 . (in Chinese)
翟永立 , 丁雷 , 裴浩东 . 基于改进轮廓提取的Hough变换椭圆检测方法 [J]. 现代电子技术 , 2019 , 42 ( 6 ): 34 - 37 .
ZHAI Y L , DING L , PEI H D . A Hough transform ellipse detection method based on improved contour extraction [J]. Modern Electronics Technique , 2019 , 42 ( 6 ): 34 - 37 . (in Chinese)
MCLAUGHLIN R A . Randomized Hough Transform: Improved ellipse detection with comparison [J]. Pattern Recognition Letters , 1998 , 19 ( 3/4 ): 299 - 305 .
XU L , OJA E , KULTANEN P . A new curve detection method: Randomized Hough transform (RHT) [J]. Pattern Recognition Letters , 1990 , 11 ( 5 ): 331 - 338 .
GUIL N , ZAPATA E L . Lower order circle and ellipse Hough transform [J]. Pattern Recognition , 1997 , 30 ( 10 ): 1729 - 1744 .
CHIA A Y S , LEUNG M K H , ENG H L , et al . Ellipse detection with Hough transform in one dimensional parametric space [C]. 2007 IEEE International Conference on Image Processing . 1619,2007 , San Antonio , TX, USA . IEEE , 2007 , V: 333 - 336 .
邹荣 , 赵稼宸 , 凌俊 , 等 . 基于Hough投票空间的椭圆图像特征亚像素提取方法 [J]. 光学技术 , 2016 , 42 ( 2 ): 141 - 145 .
ZOU R , ZHAO J CH , LING J , et al . Sub-pixel ellipse feature extraction method based on Hough voting space [J]. Optical Technique , 2016 , 42 ( 2 ): 141 - 145 . (in Chinese)
PRASAD D K , LEUNG M K H , CHO S Y . Edge curvature and convexity based ellipse detection method [J]. Pattern Recognition , 2012 , 45 ( 9 ): 3204 - 3221 .
FORNACIARI M , PRATI A , CUCCHIARA R . A fast and effective ellipse detector for embedded vision applications [J]. Pattern Recognition , 2014 , 47 ( 11 ): 3693 - 3708 .
WANG Z P , CHEN D R , GONG J L , et al . Fast high-precision ellipse detection method [J]. Pattern Recognition , 2021 , 111 : 107741 .
LIU Y , XIE Z W , LIU H . Fast and robust ellipse detector based on edge following method [J]. IET Image Processing , 2019 , 13 ( 13 ): 2409 - 2419 .
MENG C , LI Z X , BAI X Z , et al . Arc adjacency matrix-based fast ellipse detection [J]. IEEE Transactions on Image Processing , 2020 , 29 : 4406 - 4420 .
JIA Q , FAN X , LUO Z X , et al . A fast ellipse detector using projective invariant pruning [J]. IEEE Transactions on Image Processing , 2017 , 26 ( 8 ): 3665 - 3679 .
徐波 , 冯山 . 基于邻域关系矩阵的属性约简算法 [J]. 小型微型计算机系统 , 2019 , 40 ( 8 ): 1595 - 1600 .
XU B , FENG SH . Attribute reduction algorithm based on neighborhood relationship matrix [J]. Journal of Chinese Computer Systems , 2019 , 40 ( 8 ): 1595 - 1600 . (in Chinese)
PRASAD D K , LEUNG M K H , QUEK C . ElliFit: an unconstrained, non-iterative, least squares based geometric Ellipse Fitting method [J]. Pattern Recognition , 2013 , 46 ( 5 ): 1449 - 1465 .
初广丽 , 王延杰 , 邸男 , 等 . 复杂场景中航天器靶标的快速识别 [J]. 光学 精密工程 , 2016 , 24 ( 4 ): 865 - 872 .
CHU G L , WANG Y J , DI N , et al . Fast recognition of aircraft target in complex scenes [J]. Opt. Precision Eng. , 2016 , 24 ( 4 ): 865 - 872 . (in Chinese)
吴斌 , 叶东 , 张鑫 , 等 . 非合作目标姿态测量的嵌入式算法 [J]. 光学 精密工程 , 2016 , 24 ( 11 ): 2804 - 2813 .
WU B , YE D , ZHANG X , et al . Embedded algorithm for relative pose measurement between non-cooperative targets [J]. Opt. Precision Eng. , 2016 , 24 ( 11 ): 2804 - 2813 . (in Chinese)
0
浏览量
534
下载量
2
CSCD
关联资源
相关文章
相关作者
相关机构