摘要
边缘提取在图像处理中具有重要地位 ,Mallat等提出利用小波变换的局部极大值来表征信号奇异点 ,从而进行图像边缘的提取。但直接使用 Mallat算法进行边缘提取时需进行二抽取操作 ,图像将相应缩小。利用小波变换的多孔算法实现图像边缘提取 ,并提出一种更为简化的改进方法 ,实例测试验证了其有效性。
The edge detection is important in image processing. Mallat and other scholar had presented that the singularity of signal can be represented by the local maxima of its wavelet transform, thus we can detect the edges by wavelet transform. But the downsampling in Mallat algorithm will lessen the image, it is not suitable for the image edge detection. In this paper, the image edge can be detected through algorithm a′trous, and a simple implement method is presented. The validity of this method is proved by the experiment.
出处
《武汉理工大学学报》
CAS
CSCD
2004年第1期76-79,共4页
Journal of Wuhan University of Technology
基金
"十五"国防预先研究项目 (40 4 0 4 0 70 30 2 )
关键词
小波变换
边缘提取
多尺度分析
wavelet transform
edges detection
multiscale analysis
