摘要
图像置乱变换是信息隐藏和图像加密常用的方法,如何衡量一种置乱变换的好坏,是研究的热点和难点问题。基于置乱前后位置变化的方法可以作为一个衡量的标准,但是位置变化并不能代表像素点值的改变,而且把图像作为一个整体来考虑也会产生一些偏差。为了克服这个缺点,提出了一种基于各点相关性的图像置乱程度衡量方法,主要是通过比较置乱前后每个点与其相邻点相关性的差异来进行衡量。由Matlab仿真结果表明,该方法可以很好地衡量图像置乱次数与置乱效果的关系,而且与人类的视觉具有比较好的一致性。
Image scrambling transformation is a common method in information hiding and image eneryption.How to evaluate the advantage and disadvantage of scrambling transformation is the hot and difficult issue of research.The method that position change based on scrambling in front and back can be the standard of the measurement.But the positional change can not stand for the change of the pixel value,and regarding an image as a whole may produce deviation.In order to overcome the shortcoming,it proposes a evaluation method of image scrambling degree,which is based on point-relativity.It measures the relative differences of a point and adjacent point in scrambling front and back.As the simulation result of Madab,it shows that the method can measure the relationship between the times of image scrambling and scrambling degree,while it can be good consistency with human vision.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第8期134-136,共3页
Computer Engineering and Applications