| Autor: |
Qiang Wang, Hongbo Chen, Xiaorong Xu, Haiyan Liu |
| Předmět: |
|
| Zdroj: |
International Journal of Image & Graphics; Oct2005, Vol. 5 Issue 4, p715-727, 13p, 4 Color Photographs, 7 Diagrams, 2 Charts, 2 Graphs |
| Abstrakt: |
The heavy burden of computational complexity and massive storage requirement is the drawback of the standard Hough transform (SHT). To overcome the weakness of SHT, many modified approaches, for example, the probabilistic Hough transform (PHT), have been presented. However, a very important fact, which is that a line has its own width in a real digital image and the width of the line is uniform, was ignored by all of these modified algorithms of Hough transform. This phenomenon influenced the result of line detection. In this paper a new modified algorithm of Hough transform for line detection is proposed. In our algorithm, the fact mentioned above is fully considered and a strip-shaped area corresponding to the accumulate cells of HT is proposed. Experimental results have shown that our approach is efficient and promising, and the effect of detection is far better than the popular modified approaches. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
