8方向自适应权值Sobel边缘检测算子  

Adaptive Weighted Sobel Edge Detection Operator with Eight Directions

作　　者：武亚沙 张有会[1,2] 邱志宇[1,2] 

机构地区：[1]河北师范大学数学与信息科学学院,河北石家庄050024 [2]河北省计算数学与应用重点实验室,河北石家庄050024

出　　处：《河北师范大学学报（自然科学版）》2015年第4期294-299,共6页Journal of Hebei Normal University：Natural Science

基　　金：国家自然科学基金(61170107);河北省教育厅自然科学研究项目(Q2012041)

摘　　要：针对传统Sobel边缘检测算子对图像边缘检测存在漏检和噪声敏感的缺点,提出一种改进的Sobel算子.该算子重新构造出一种8方向5×5检测模板,并且在模板卷积过程中应用自适应权值的算法,增强了该算子的抗噪性.实验表明,改进的Sobel算子保留了传统Sobel算子的优点,增加了其他方向的边缘检测,降低了漏检的概率,提高了抗噪性.To overcome the shortcomings of classical Sobel edge detection operator, such as missed detection of image edges and sensitive to noise, this paper proposes an improved edge detection operator. The new operator uses 5 × 5 templates in 8 directions and convolution involving adaptive weights, and it en hances noise immunity. Experimental results show that the new Sobel operator maintains advantages of the classical Sobel operator, enhances the edge detection of other directions,reduces the undetected probability and improves anti-noise performance.

关 键 词：图像处理 边缘检测 SOBEL算子 8方向 自适应权值 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]