基于提升小波的自适应阈值边缘检测新算法

A New Adaptive Threshold Edge Detection Method Based on Lifting Wavelet

针对传统的单一边缘检测算法抗噪能力差、边缘不连续等缺点,本文提出采用两种算法相结合的方式来进行边缘检测。首先,对原始图像进行多层小波分解;对分解后的图像低频部分用提出的改进提升算法进行边缘检测,对高频部分用小波变换的局部模极大值算法检测边缘;通过将各层边缘信息按一定的融合规则融合起来得到一个组合边缘,最后细化图像边缘。实验证明,这种方法相对于传统小波分析有着计算量小,计算速度快和要求存储空间小等诸多优势,同时,也能做到不丢失图像信息,保证了边缘的连续性和封闭性,检测效果较好。

The image edge detection is one of the basic contents in image manipulation and analysis. Against the deficiency of low anti-noise capacity, dis-continuous edge of traditonal single edge detection algorithms, a new edge detection method combing two algorithms is proposed. Lifting wavelet is used to detect the edge through multi-level wavelet decomposition, thus the image edges of different resolution is obtained. Then these edges are combined as an integrated edge and an algorithm of thinning is adopted to thin the edge. Experimental results show that this method has a small amount of calculation, fast calculation speed and small storage space and many other advantages. This new algorithm is better than the traditional edge detection methods.