Edge detection is one of the most active research topics in the field of image processing and computer visions. Traditional image edge detection methods are sensitive to noise. Aiming at the shortcomings of the conventional methods, it is presented the method of edge detection based on dyadic wavelet transform. The original image is decomposed by dyadic wavelet at first, then the low-frequency sub-image is enhanced by histogram equalization, the enhanced low-frequency sub-image is detected using dyadic wavelet transform modulus maximum edge detection method, and the edge of original image is obtained finally. The results show that the method is much better than using traditional image edge detection methods. It is also better than using dyadic wavelet transform modulus maximum edge detection method directly.%边缘检测是图像处理和计算机视觉领域最活跃的研究课题之一.传统边缘检测方法对噪声非常敏感,针对该问题在传统边缘检测算法分析的基础上,提出了一种基于二进小波变换的图像边缘检测方法.首先,对原图像进行二进小波分解,然后对低频子图像用直方图均衡化来进行增强,对增强后的低频子图像用二进小波变换模极大值点方法进行边缘检测得到边缘图像.实验结果表明,这种边缘检测方法明显优于对原图像直接使用传统边缘检测算子或二进小波变换模极大值点的边缘检测方法.
展开▼
