¹ The College of Information Engineering, Xiangtan University, Hunan 411105, China
² School of Computer, National University of Defense Technology, Hunan 410073, China
³ Key Laboratory of Intelligent Computing & Information Processing of Ministry of Education, Xiangtan University,
⁴ School of Mathematics and Computational Science, Xiangtan University, Hunan 411105, China
⁵ National Computer Network Emergency Response Technical Team/Coordination Center of China,Beijing 100029, China
⁶ Department of Information and Computer Engineering, Ajou University, Suwon, 443749, Korea

---

Abstract—In this paper, an algorithm named best wavelet packet tree decomposition (BWPTD) is proposed for image compression. In order to obtain better sparse representation of image, best wavelet packet basis is introduced to decompose image signal in the algorithm. Experimental results show that BWPTD is better than single layer wavelet decompression (SLWD) and original compressed sensing (OCS) in peak signal to noise ratio (PSNR) by 2db and 8db, respectively. In addition, the reconstruction time of BWPTD is only half as that of SLWD and OCS.

Index Terms—compressed sensing, image processing, wavelet packet, matching pursuit algorithms

[PDF]

Cite: Zhetao Li, Jingxiong Xie, Zuguo Yu, Dengbiao Tu, and YoungJune Choi, " Compressed Sensing Based on Best Wavelet Packet Basis for Image Processing," Journal of Computers vol. 8, no. 8, pp. 1947-1950, 2013.