| |
| 科研成果 |
|
|
 |
|
| 论文题目： |
A hybrid method for ellipse detection in industrial images |
| 第一作者： |
Chen SL(陈松林);Xia RB(夏仁波);Zhao JB(赵吉宾);Chen YL(陈月玲);Hu MB(胡茂邦) |
| 参与作者： |
|
| 联系作者： |
|
| 发表刊物： |
Pattern Recognition |
| 发表年度： |
2017 |
| 卷,期,页： |
68,,82-98 |
| 论文出处： |
|
| 第一作者所在部门： |
|
| 论文编号： |
|
| 论文摘要： |
Many ellipse detection methods have been proposed for detecting ellipses in images. However, they are unsuitable for industrial images due to low signal-to-noise ratios (SNR). This paper presents an ellipse detection method combining the advantages of Hough transform(HT) based methods and the advantages of edge following methods, which is capable of detecting fragmented ellipses and is both computational and memory efficient. Our method works in two steps. In the first step, an edge following method is proposed to quickly and accurately extract the majority of ellipses. For ellipses missed in the first step, candidate regions where each may contain one missed ellipse are extracted in the second step using cluster analysis, and then a HT based method is performed on these regions to extract the missed ellipses. This can not only guarantee the accuracy of the HT based method, but also save the memory and computation time. We test the performance of our method using both synthetic images and low SNR industrial images. Experimental results demonstrate that the proposed method performs far better than existing methods in terms of recall, precision, F-measure, and reliability. Especially in term of reliability, our method has achieved a very high value close to 1 while the reliabilities of state-of-the art methods are almost less than 0.5. |
| 论文全文： |
|
| 其他备注： |
|
| 附件下载： |
|
|
|
