一种融合小波变换和数学形态学的图像边缘检测算法

摘要

Considering the disadvantages in the traditional image edge detection methods, such as emergence of noises edges and inaccurate edge location, this paper proposes a fusion edge detection algorithm based on the combination of wavelet transform with the mathematical morphology algorithm to solve the above problems.The image is divided by wavelets, then the algorithm of the wavelet module maximum coefficients combining with the cross-band and cross-scale correlation is applied to the high frequency part of the image , so noises can be suppressed effectively.The mathematical morphology algorithm is applied to the low frequency part of the image, so weak edges can be detected with high quality and accurate location.Finally, the two images mentioned above are fused.The results show that the proposed algorithm can spppress noises efficiently, improve the accurancy of image edges and guarentree accurate location.%针对传统图像边缘检测方法中出现噪边、边缘定位不精确等缺点,提出一种融合小波变换和数学形态学的图像边缘检测算法.先将图像进行小波分解,高频部分利用小波模极大值噪声抑制和尺度相关子带关联算法进行边缘检测,可以很好地减弱噪声;低频部分利用修正的形态算子进行形态学边缘检测,能够检测出弱边且定位准确;最后对两种方法得到的边缘图像进行融合.实验结果表明,该算法能有效抑制噪声,提高边缘精度并且定位准确.