摘要： 针对家庭数字照片的特点以及应用范围，提出了带有衰减因子的Robert微分算子与动态的自适应阈值相结合的边缘检测方法，并利用了边缘检测后边缘点的方向信息，作为Hough变换的方向角，可以较快提取出边缘线段，从而通过边缘跟踪获得无噪声点的相似区域，这为进一步提取图像的颜色特征或形状特征提供了良好的基础。

关键词: 图像分割, 边缘检测, 自适应阈值, Hough变换

Abstract: Considering the speciality and the application range of the home digital images, an image segmentation approach based on edge detection is proposed in this paper. It introduces a Robert differential operator with an attenuation factor to obtain a gray edge image, then a white- black binary image is obtained by a dynamical automatic adapting threshold. The edge points' direction information is applied to the Hough transform, which speeds up the process of extracting edge lines. After that, the similar regions which have less yawp point can be extracted by these edge lines using the egde tracking. From these regions, people can extract the color feature and the shape feature with better performance.

中图分类号: 

• TP391.41