基于区域分割与形态学相融合的图像边缘检测  

The Study of Image Edge Detection Based on Segmentation and Morphological Methods

作　　者：姜静[1] 李纲[1] 

机构地区：[1]周口师范学院计算机科学与技术学院,河南周口466001

出　　处：《平顶山学院学报》2014年第5期84-86,共3页Journal of Pingdingshan University

基　　金：河南省科技厅攻关项目(122400450356)

摘　　要：图像边缘检测是图像分析的基础,但现有图像边缘检测方法各有特点,有一定的局限性,因此还需要进一步的探索和研究.根据区域分割和形态学腐蚀的特点,利用基于区域的四叉树分裂合并法对图像不同区域设置不同阈值进行区域分割,对于分布比较单一的内容区域阈值设置一般比较低,然后把分割区域进行腐蚀,从而提取出边缘信息.最后对此方法进行实例仿真,并与Prewitt算子和Canny算子进行比较,可见此算法对于内容比较单一的图像具有较高的检测精度和较强的抗干扰能力.Existing edge detection methods have their limitations and shortcomings,and need further im-provement and development. This paper uses quadtree region splitting and merging method to segment the image, and different images set different thresholds value for regional segmentation. Then it uses the corrosion method of morphological to extract image edge. At last it uses canny operator and prewitt operator to extract image edge of the image contents of a single distribution. Compared with edge extraction of canny operator and prewitt operator, this algorithm has strong anti - interference ability and high detection accuracy.

关 键 词：边缘检测 四叉树分裂合并 腐蚀运算 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]