| 基于Blob算法的标记圆检测技术研究 |
| |
| 引用本文： | 段康容,;刘先勇.基于Blob算法的标记圆检测技术研究[J].传感器世界,2014(8):7-10. |
| |
| 作者姓名： | 段康容 ;刘先勇 |
| |
| 作者单位： | [1]西南科技大学工程技术中心,四川绵阳621010; [2]绵阳铁牛科技有限公司,四川绵阳621010 |
| |
| 基金项目： | 四川省科技支撑计划基金项目（项目编号：2013GZX0154） |
| |
| 摘 要： | 大型物体的三维测量中,大多采用基于标记圆的拼接,标记圆检测的正确性和定位精度决定了拼接的精度。在二值化图像上运用Blob分析得到标记圆的轮廓信息,综合使用标记圆灰度特性、尺寸特征、圆形度和误差准则来筛选标记圆,圆心定位采用具有亚象素定位精度的最小二乘拟合法。实验表明该方法抗干扰强,标记圆的识别率可达98%。
|
| 关 键 词： | 三维测量 标记圆 Blob算法 亚象素定位 |
Research on reference point detection based on Blob algorithm |
| |
| Affiliation: | DUAN Kang-rong, LIU Xian-yong ( 1 The Center of Engineering and Technology. Southwest University of Science and Teehnology, Mianvang 621010, China ; 2 Mianyang TieNiu technology Co., Ltd. Mianyang 621010,China) |
| |
| Abstract: | Registration based on reference circles is usually used in 3-D measurement of large objects. The detection accuracy and the precision of reference circles determine the accuracy of registration. The Blob analysis is used to get contour information of circles from the binary images. The circles are searched after those ensemble characteristics： grayscale, circle size, circular degree and error criterion. Circle sub-pixel localization uses the least squares fitting method. Experiments show that the method is robust to noise and the correct decoding rate is up to 98%. |
| |
| Keywords: | 3-D measurement reference points BIob algorithm Sub-pixel localization |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
