| 摘要: |
| 目的 从颜色这个角度研究新的全参考图像质量评价方法。方法 首先在Lab色空间内基于颜色信息提取2个图像特征，即颜色和对比度特征。然后将颜色特征和图像梯度相结合，对比度特征和图像局部范围相结合，分别提出不同的图像质量评估方法。结果 采用该领域内4种常用的参数来评估文中算法和14种其他经典算法的表现，结果显示文中提出的2种基于图像彩色信息的算法总体上优于其他算法。结论 从颜色计算显著特征出发来研究图像质量评价方法是一个有效的途径。 |
| 关键词: 全参考图像质量评价方法 彩色显著特征 对比度显著特征 图像局部范围 |
| DOI： |
| 分类号:TP391 |
| 基金项目: |
|
| Color Image Quality Assessment Based on Color Visual Saliency Feature |
|
WANG Xiao-hong, KUANG Sheng-kun
|
|
University of Shanghai for Science and Technology, Shanghai 200093, China
|
| Abstract: |
| The work aims to study new full reference image quality assessment methods in view of color. First, two image features like color and contrast saliency were extracted from Lab color space based on color information. Then color feature was combined with gradient map and contrast ratio was combined with local scope of image to propose different image quality assessment methods respectively. The mentioned algorithms and 14 other classic algorithms were evaluated by four common indexes. The assessment result showed that the proposed two new algorithms were generally better than all other ones. In conclusion, it is an effective way to study image quality assessment from aspect of color visual saliency feature. |
| Key words: full reference image quality assessment color saliency feature contrast saliency feature local scope of image |
