| 基于注意力选择与感觉容量的图像质量评价算法 |
| |
| 引用本文： | 郑江云.基于注意力选择与感觉容量的图像质量评价算法[J].安庆师范学院学报(自然科学版),2014(3):74-77. |
| |
| 作者姓名： | 郑江云 |
| |
| 作者单位： | 安庆师范学院物理与电气工程学院,安徽安庆246133 |
| |
| 基金项目： | 安徽省高等学校省级自然科学研究基金重点项目（KJ2010A226）资助. |
| |
| 摘 要： | 图像质量评价是用可计算的模型模拟人的主观判断。根据人眼对图像高低频失真的感觉容量不同及注意力选择特性,本文提出一种新的图像质量评价算法。首先,将图像的高低频信号分别选用视觉显著图的系数和小波变换的近似系数表示;然后,采用不同方法分别计算图像高频和低频失真量,两种失真量的乘积就代表图像质量评价。这种客观评价算法与LIVE图库中DMOS的线性相关性达到0.906 4以上,明显优于PSNR和SSIM等算法。
|
| 关 键 词： | 图像质量评价 注意力选择 感觉容量 视觉显著图(VSM) |
An Algorithm of Image Quality Assessment Based on Attention Selection and Sense Capacity |
| |
| ZHENG Jiang-yun.An Algorithm of Image Quality Assessment Based on Attention Selection and Sense Capacity[J].Journal of Anqing Teachers College(Natural Science Edition),2014(3):74-77. |
| |
| Authors: | ZHENG Jiang-yun |
| |
| Institution: | ZHENG Jiang-yun (School of Physics and Electronic Engineering, Anqing Teachers College, Anqing 246133 ,China) |
| |
| Abstract: | Image quality assessment simulates human subjective opinion with calculated model. Based on different sensitivities in the frequency domain and attention selection of human visual system （HVS）, a new algorithm of image quality assessment is proposed. First, high and low frequency signals of image are expressed by coefficients of visual saliency map and wavelet approximation coefficients. Then, low frequency error and high frequency error are calculated by different means, the product of two errors is adopted as objective evaluation of quality. The new method is validated with subjective quality scores on LIVE database, linear correlation coefficient reaches 0. 9064. Experimental results show that the performance of the new method is superior to the algorithms of PSNR and SSIM. |
| |
| Keywords: | image quality assessment attention selection sense capacity visual saliency map（VSM） |
| 本文献已被 CNKI 维普 等数据库收录！ |
