直线提取的预处理研究

Investigation on Preprocessing of Line Extraction

传统的直线提取算法对标准的直线比较有效,但对非典型直线并不适用。现实应用中图像由于受到噪声污染,图像中的直线会变得模糊不清,一般的直线提取方法无法提取到足够的直线信息。为了准确提取出受污染图像中足够的直线信息,以车牌和甲骨拓片字形为例,提出了一种结合分形几何与霍夫变换来提取非典型直线的方法。该方法采用基于分形几何的方法对非典型直线进行计算机辅助复原,通过计算非典型直线边缘的分形维数,对其边缘进行压缩变换,平滑非典型直线边缘因污染所造成的凹凸形态,恢复非典型直线的原貌,并利用Hough变换提取其中的直线。实验结果表明,经平滑后的非典型直线中可提取更多的直线信息,分形几何与霍夫变换的组合是有效的直线提取方法。

Although traditional line extraction algorithms are effective for standard lines, they are not applicable to atypical lines. The lines in image are blurry since there is noise in many applications, and sufficient information cannot be acquired with the general method of line extraction. To extract sufficient line information in noised image accurately, a method for extracting atypical lines by combining fractal ge- ometry and Hough transform has been proposed, with car plate and Oracle character shape as the examples. It adopts the approach based on fractal geometry to realize computer-assisted restoration for atypical lines. After computation of fractal dimensions of atypical lines edges, compressing and transforming of their edges, the shape of concaves or convexes for atypical lines since noise is smoothed and their o- riginal shape are restored, extraction of straight lines by Hough transform. The experimental results show that more information has been extracted from the smoothed atypical lines and that the method of fractal geometry with Hough transform is effective.