Title :
Design and Application of All Phase Pyramidal Directional Filter Bank
Author :
Li, Li ; Hou, Zhengxin ; Wang, Chengyou ; He, Fan
Abstract :
Based on the essence of contourlet and all phase DCT sequency filtering theory, a novel all phase pyramidal directional filter bank is developed and applied to image denoising. In all phase datum space, pure 2-D directional filters are designed directly from its spectrum characteristics, so the directional filter banks can be constructed flexibly without rotation and resampling of image so as to preserve more image details and decrease the computational complexity. In addition, with the all phase directional filter bank proposed, horizontal and vertical directional subbands are divided, which can not be obtained by contourlet and other directional filter banks. It is shown in our experimental results that the proposed approach is superior to both wavelet and contourlet for most of the images and has especially better visual quality for images with many horizontal and vertical edges.
Keywords :
computational complexity; discrete cosine transforms; filtering theory; image denoising; DCT sequency filtering theory; computational complexity; image denoising; pyramidal directional filter bank; visual quality; Computer networks; Design engineering; Discrete cosine transforms; Educational technology; Filter bank; Frequency; Image denoising; Information management; Space technology; Wavelet transforms;
Conference_Titel :
Networked Computing and Advanced Information Management, 2008. NCM '08. Fourth International Conference on
Conference_Location :
Gyeongju
Print_ISBN :
978-0-7695-3322-3
DOI :
10.1109/NCM.2008.85