基于方向导数和B样条小波的图像边缘检测  

Image edge detection based on directional derivative and cubic B-spline wavelet

作　　者：赵志刚[1] 杨应平[1] 蒋爱湘[1] 彭榆伟[1] 

机构地区：[1]武汉理工大学理学院,武汉430070

出　　处：《计算机工程与应用》2009年第23期176-178,189,共4页Computer Engineering and Applications

基　　金：国家自然科学基金No.60471024~~

摘　　要：根据图像边缘及噪声的多尺度传播特性和小波边缘检测的基本原理,提出了一种基于方向导数和三次B样条小波的边缘检测算法。该算法兼顾图像边缘的方向特征和小波基对称、线性相位的特点,较好地解决了边缘提取精度与噪声抑制能力之间的矛盾。通过计算机仿真对该算法进行验证,结果表明该算法不仅能准确地检测出图像边缘,而且能有效地抑制噪声,优于传统的边缘检测算法。According to the multi-scale propagation characteristic of image edge and noise as well as principle of edge detection using wavelet transforman,au algorithm based on cubic B-spline wavelet and directional derivative is presented.The algorithm takes not only edge direction of an image,but also symmetry and linear phase feature of the wavelet base into account,so well resolves the contradiction between the detection precision and the noise suppression ability.Computer simulation results show that this algorithm is superior to traditional edge detection algorithms with both precise image edge detection and effective noise restraining.

关 键 词：小波变换 三次B样条 方向导数 边缘检测 

分 类 号：TN911.37[电子电信—通信与信息系统]