基于梯度改进 Hough 变换的油菜籽图像分割方法

摘要

提出了一种基于梯度改进的Hough 变换的图像分割算法，用于提高油菜籽的图像分割质量。先将原始图像转换成灰度图像，然后通过改进的Hough 变换找到油菜籽粒的圆心，连接两个相邻的籽粒的圆心，将垂直于圆心连线的切线作为两个籽粒的分界限，从而实现油菜籽粒的分割。采用一些通用的分割算法与本方法对同一图像的分割效果进行比较，结果表明本方法的分割效果优于其它方法。%A method for improving the quality of rapeseed image segmentation based on the gradient im-provement of Hough transform is proposed. At first,the original image is converted to gray image,then the center of rapeseed is found by the improved Hough transform,the connecting line of the center of the two adjacent rapeseed is found,and the tangent that perpendicular to the connecting line of the center is con-sidered to be the boundary between the two rapeseed to separate it. Compared the segmenting effects of the same image,the result by above method is superior to those by some common image segmentation methods.