摘要
根据压缩图像中图像块平滑特性的不同,提出了一种基于平滑因子的压缩图像质量盲评估方法。通过统计图像中为零的一维DCT交流系数的比例,构造了压缩程度因子以及质量评估体系,可有效区分图像的压缩程度。实验表明该方法对压缩图像有较好的评价效果,同时在对低通滤波图像的质量评估中性能优于利用块边缘像素差异的方法。
According to the different smoothness characteristics of compressed images blocks, a blind compressed image quality assessment method is proposed. The system of compression degree and quality assessment is constructed through the ratio of zero one-dimension AC coefficients, which can effectively identify the degrees of compression. Experiments show that the proposed method has a good performance for compressed images, and has a better performance for the low-pass filtered images than the methods using block edge pixel differences in the simulation.
出处
《电视技术》
北大核心
2013年第2期91-94,共4页
Video Engineering
基金
国家自然科学基金项目(61103181)
上海自然科学基金项目(11ZR1413200)
上海市教委创新基金项目(11YZ10)
关键词
质量评估
无参估计
压缩图像
image quality assessment
no-reference assessment
compressed image
