Segmentation research between polygonal image and background
Qingdong Yan · Jisuanji gongcheng yu sheji · 2011
While a sampled image is a regular polygon or can be fitted a polygon with strong noise on the background,it is a great challenge to filter the noise with regular filter methods,furthermore,in which there is no any universality.A image segmentation tech-nology is put forward,which Ramer recursive subdivision arithmetic is employed to fit targets’ outline into a polygon,then Hough trans-form is adopted to detect this outline,solved the problem of selecting neighborhood in Hough transform and confirming outline point ac-cording to points of intersection by Hough detection.Consequently,case studies for images with quadrangle and hexagon are carried out,and the experimental results demonstrated the effectiveness of this method.