Journal of Shanghai University >
Arc contour detection method based on discrete curvature characteristics
Received date: 2016-01-28
Online published: 2017-10-30
Detection of object shape is the basis of autonomous environment understanding for robots. An object turns into connected domains after color layering and multi-scale filtering segmentation, and therefore it can be easily analyzed. In view of this, a method based on discrete curvature characteristics is proposed to detect the arc shape of objects. This method is based on extraction of object contour, and detection of lines and features. Interferential lines are illuminated, and the arc features are detected according to the discrete curvature changes of the remaining contour points. A system of robot operation is established for experiments. Average precision of arc contour detection is 90.6% and the handling time is 0.75 s. The result shows effectiveness of the proposed method in detecting arc contour of objects.
Key words： environment understanding; object shape; service robot; arc detection
ZHANG Xudong, ZHAO Qijie . Arc contour detection method based on discrete curvature characteristics[J]. Journal of Shanghai University, 2017 , 23(5) : 702 -713 . DOI: 10.12066/j.issn.1007-2861.1764
[1] Uckermann A, Eibrechter C, Haschke R, et al. Real-time hierarchical scene segmentation and classification [C]// 2014 14th IEEE-RAS International Conference on Humanoid Robots
(Humanoids). 2014.
[2] Borges P V K, Moghadam P. Combining motion and appearance for scene segmentation[C]// 2014 IEEE International Conference on Robotics and Automation (ICRA). 2014:
1028-1035.
[3] Li J, Jin L, Fei S, et al. Robust urban road image segmentation [C]// 11th IEEEWorld Congress on Intelligent Control and Automation (WCICA). 2014: 2923-2928.
[4] 庄严, 卢希彬, 李云辉. 移动机器人基于三维激光测距的室内场景认知[J]. 自动化学报, 2011, 37(10): 1232-1240.
[5] 李平, 杨宁, 钟绍山. 基于RFID 和DR 技术的室内组合定位技术[J]. 上海大学学报(自然科学版), 2014, 20(4): 498-504.
[6] 武丽, 陈军, 张海瑞. 基于傅里叶描述子的形状特征提取的讨论与研究[J]. 中国科技信息, 2013(7): 51.
[7] Zhou L, Hospital M T. Beyond local appearance: category recognition from pairwise interactions of simple features [C]// IEEE Conference on Computer Vision and Pattern Recognition.
2007: 1-8.
[8] 黄东晋, 张倩, 吴冏, 等. 基于角点及局部特征的饮料瓶识别方法[J]. 上海大学学报(自然科学版), 2011, 17(3): 238-242.
[9] Stark M, Goesele M, Schiele B. A shape-based object class model for knowledge transfer [J]. IEEE International Conference on Computer Vision, 2009, 30(2): 373-380.
[10] 吴桐树, 张瑞林, 邹敏. 基于Freeman 链码的B 样条曲线轮廓拟合[J]. 计算机系统应用, 2014, 23(8): 130-134.
[11] 王文豪, 周泓, 严云洋. 一种基于连通区域的轮廓提取方法[J]. 计算机工程与科学, 2011(6): 67-71.
[12] 董晶, 杨夏, 于起峰. 基于边缘连接的快速直线段检测算法[J]. 光学学报, 2013(3): 213-220.
[13] 胡立华, 张继福. 基于图像的四边形自动检测方法[J]. 小型微型计算机系统, 2014, 35(8): 1911-1915.
[14] 范怡, 傅继武. 基于中点提取的椭圆检测算法[J]. 计算机应用, 2011, 31(10): 2705-2707.
[15] 师雪超, 孙振忠, 卢盛林, 等. 基于机器视觉的弧形件检测算法[J]. 计算机应用, 2010, 30(7): 1841-1843.
[16] Zhang X, Zhao Q, Meng Q X, et al. A new scene segmentation method based on color information for mobile robot in indoor environment [C]// The International Conference on Wearable Sensor and Robot (ICWSR2015). 2015: 1867-1874.
[17] 段汝娇, 赵伟, 黄松岭, 等. 一种基于改进Hough 变换的直线快速检测算法[J]. 仪器仪表学报, 2011(12): 2774-2780.
[18] Shi J, Tomasi C. Good features to track [C]// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 1994: 593-600.
[19] 何鹏, 高峰, 魏厚敏. 基于Catmull-Rom样条曲线的弯曲车道线检测研究[J]. 汽车工程学报, 2015(4): 276-281.
[20] 蒋海波, 张云伟, 徐会杰, 等. 农田作业机器人视觉导航目标曲线检测算法研究[J]. 昆明理工大学学报(理工版), 2008, 33(4): 91-96.
/
| 〈 |
|
〉 |
