Tetrolet变换方块效应改善算法  

Improved algorithm for reducing blocking artifacts of Tetrolet transform

作　　者：张凌晓[1] 刘克成[1] 李财莲[2] 

机构地区：[1]南阳理工学院计算机与信息工程学院,河南南阳473004 [2]海南省工业和信息化厅信息化产业处,海南海口570204

出　　处：《红外与激光工程》2014年第5期1679-1684,共6页Infrared and Laser Engineering

基　　金：河南省科技攻关项目(122102210563;132102210215)

摘　　要：针对Tetrolet变换算法对图像去噪后存在方块效应的缺陷,文中对Tetrolet变换算法进行了扩展和改进,并引入移位(Cycle Spinning)来有效消除Tetrolet变换算法中的方块效应。仿真结果表明,所提出的算法不仅能有效去除噪声,而且可得到更高的峰值信噪比,提高了图像的主客观质量。去噪后图像保留了原始图像的边缘和细节等局部特征,较为平滑,且方块效应得到了一定的改善和抑制,因而该算法是有效可行的。In order to reduce the blocking artifacts resulted from Ttetrolet transform algorithm, Tetrolet transform was improved and Cycle Spinning was employed to avoid the blocking artifacts in this paper. And thus the improved Tetrolet transform was introduced and performed for image denoising. Some numerical experiments show the effectiveness and superiority of our technique. Compared with typical Tetrolet transform, the denoised images by our method are smoother, and the blocking artifacts is improved to some extent and inhibition. At the same time it can preserve more significant information of original images, such as local features, including image edges and image details. In addition, the proposed method gives better performance in Peak Signal to Noise Ratio （PSNR） and improves the quality of subjective and objective of image. Experimental results show that the proposed algorithm is effective and feasible, and can perform better on most data sets.

关 键 词：改善算法 Tetrolet变换 移位 方块效应 图像去噪 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]