| 基于小波变换的屋脊边缘亚像素检测 |
| |
| 引用本文： | 丁兴号,邓善熙. 基于小波变换的屋脊边缘亚像素检测[J]. 哈尔滨工业大学学报, 2004, 36(11): 1480-1482,1530 |
| |
| 作者姓名： | 丁兴号 邓善熙 |
| |
| 作者单位： | 厦门大学,电子工程系,福建,厦门,361005;合肥工业大学,仪器仪表学院,安徽,合肥,230009 |
| |
| 基金项目： | 安徽省"十五"科技攻关重点项目(01022010). |
| |
| 摘 要： | 工业图像检测中,许多待检目标边缘是屋脊边缘,而现有亚像素边缘检测算法都是假设边缘模型为阶跃边缘,并且现有算法在定位精度、抗噪性能和计算速度方面还不尽人意．针对上述问题,提出一种使用小波变换系数的屋脊边缘亚像素检测算法,将小波分析和概率统计理论引入到亚像素边缘检测问题中,该方法具有较好的定位精确度、抗噪能力和较快的计算速度．实验证明该方法可达到1／60像素的精确度。
|
| 关 键 词： | 小波变换 屋脊边缘 亚像素 边缘检测 |
| 文章编号： | 0367-6234(2004)11-1480-03 |
| 修稿时间： | 2003-05-21 |
Sub-pixel roof edge detection based on wavelet transform |
| |
| DING Xing-hao,DENG Shan-xi. Sub-pixel roof edge detection based on wavelet transform[J]. Journal of Harbin Institute of Technology, 2004, 36(11): 1480-1482,1530 |
| |
| Authors: | DING Xing-hao DENG Shan-xi |
| |
| Affiliation: | DING Xing-hao~1,DENG Shan-xi~2 |
| |
| Abstract: | In the traditional algorithm for sub-pixel edge detection, the model of edges on the objects images is supposed as a step edge. However, the edges of many industrial objects inspected are roof edge. For improving the traditional algorithm in edge location accuracy, ability of anti-noise and the computation speed, a new sub-pixel roof edge detection algorithm is presented, based on wavelet transform coefficient and probability theory. This new algorithm has been shown with better edge location accuracy, anti-noise ability and low computation cost. The experiment showed that the measurement precision reached 1/60 pixel. |
| |
| Keywords: | wavelet transform roof edge sub-pixel edge detection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
