摘要
根据采用小波变换提取图像边缘的多尺度特性,对小波多尺度间的相关性进行了分析,提出多尺度小波变换相关算子,分析了相关算子的滤波特性,利用相关算子对图像的多尺度边缘信息进行了优化。既保留了图像中重要的细节边缘信息,又剔除大量的冗余边缘和虚假边缘,有效地提取出了图像的特征边缘。
According to multi-scale feature of image edge extraction based on wavelet transformation, the correlations between multi-scale wavelet transformation was analyzed, the correlation algorithm based on multi-scale wavelet transformation was educed, and the filtering performance of this algorithm was analyzed. The multi-scale image edges were optimized by using correlation algorithm, which can keep the important edge information in detail, eliminate redundant and illusory edge information, and also detect characteristic edge efficiently.
出处
《电光与控制》
北大核心
2009年第4期86-89,共4页
Electronics Optics & Control
关键词
图像处理
小波变换
多尺度
特征边缘
相关
image processing
wavelet transformation
multi-scale
characteristic edge
correlation
