Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (15): 25-28.

• 博士论坛 • Previous Articles     Next Articles

Real color image edge detection based on wavelet multi-scale analysis

XIAO Feng1,3,ZHOU Mingquan2,GENG Guohua1   

  1. 1.Institute of Visualization Technology,Northwest University,Xi’an 710127,China
    2.School of Information Science and Technology,Beijing Normal University,Beijing 100875,China
    3.School of Computer Science and Engineering,Xi’an Technological University,Xi’an 710032,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-05-21 Published:2011-05-21


肖 锋1,3,周明全2,耿国华1   

  1. 1.西北大学 可视化技术研究所,西安 710127
    2.北京师范大学 信息科学与技术学院,北京 100875
    3.西安工业大学 计算机科学与工程学院,西安 710032

Abstract: Modern image edge detection method requires to provide more scale edge information by making full use of the color image.A wavelet multi-scale color images edge detection algorithm is put forward.To extract true color image edge information,multi-scale wavelet feature template is taken to implement the three components(R,G,B)filter polishing output and perform the polished image gradient vector extensions(Vector Expand Gradient,VEG).Experiments show the algorithm can effectively extract the true color image edge information from fine lines to bold lines in different scales.

Key words: real color image, edge detection, wavelet transform, Vector Expand Gradient(VEG), multi-scale, polishing

摘要: 现代图像的边缘检测方法需要在充分利用图像中的色彩信息基础上能够提供多尺度下的边缘信息。提出了小波多尺度下彩色图像边缘提取算法,利用小波多尺度特征模板对R、G、B三个分量滤波磨光输出,并对磨光后的图像进行向量扩展梯度(Vector Expand Gradient,VEG),提取边缘。实验结果表明:算法能有效地在不同尺度下提取真彩色图像由细到粗的轮廓边缘信息。

关键词: 真彩图像, 边缘检测, 小波变换, 向量扩展梯度, 多尺度, 磨光