摘要
普通Hough圆检测计算量大,且实际图像易受光照和背景不均匀等条件的影响,针对这些情况提出了一种将待检测圆图像先作边缘检测后,再对边缘图像进行粗匹配和精匹配,以实现圆检测的方法.粗匹配使用改进的SSDA算法进行,精匹配利用圆的两条不平行弦的中垂线相交于圆心的几何性质,对经典Hough变换加以改进,提高了检测速度.
It consumes a lot of time when using Hough circle transformation to do circle detection. The image usually is influenced by illumination, uneven background and so on. A new idea of circle detection is proposed to detect the edge of circle first in image preprocess and deal with the edge image with rough matching and precise matching. Rough matching is accomplished by improved SSDA . Precise matching improved Hough circle transformation by using the characteristics of circle that perpendicular bisectors of two unparallel chords of a circle intersect at the center of the circle. The algorithm increases the performance of circle detection.
出处
《应用科技》
CAS
2006年第6期74-76,共3页
Applied Science and Technology
