摘要
针对目前几种数字图像清晰度评价方法的局限性,研究了图像的总有界变差,理论上证明了图像总有界变差与图像清晰度之间的关系.在图像恢复过程中,使图像的总有界变差达到极值,可以使所恢复的图像更清晰.通过计算机仿真,比较了清晰图像与模糊图像的总有界变差,并验证了有关结果.
In order to compensate for some limitations of the existed clarity evaluation function （CEF） of the image definition, a total bounded variation （TBV） of image is studied. It is proved that there is some relationships between the TVB of image and the digital image definition. Furthermore, the clearer image from image restoration process can be acquired when the criterion based on TVB of the image is used. Finally, the computer simulation is provided for the validation of the relation obtained.
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2009年第B04期120-122,139,共4页
Journal of Beijing University of Posts and Telecommunications
基金
江苏省科技重大项目(BG2006028)
关键词
图像清晰度
评价函数
图像恢复
总有界变差
image definition
evalution funtion
image restoration
total bounded variation
