基于相邻像素统计特性的LSB隐写分析技术

Steganalysis for LSB Steganography Based on Statistical Characteristic of Adjacent Pixels

基于对图像相邻像素灰度值的奇偶性及其大小关系的统计分析,提出了一种新的可靠检测空域LSB隐写的隐写分析算法。该算法通过统计相邻像素对中奇像素大于偶像素的相邻像素对数与偶像素大于奇像素的相邻像素对数,然后根据其比值来判断图像中隐秘信息的有无,并从理论上给予了证明。这一算法不仅可以实现对空域LSB隐写的可靠检测,还可以准确估计图像中隐密信息数据量的大小。算法实现简单、计算量小、检测速度快。实验结果表明,该方法可以获得优于GPC等隐写分析方法的检测性能,且适用于彩色图像。

Based on the statistical analysis on the oddity or evenness of the Gray-Scale values between adjacent pixels, a new steganalysis algorithm capable of a reliable detection of the spatial LSB steganography was proposed. The algorithm needs only to count the number of the adjacent pixel pairs in which the odd one is greater than the even one, and the number of the adjacent pixel pairs in which the even one is greater than the odd one, respectively. The ratio of which are then calculated to determine the existence of a stego-image in the carrier image, and it was proofed in theoretics. Not only can the algorithm reliably detect the existence of hidden messages embedded in images using LSB replacement, but also it can estimate the amount of hidden information exactly. Besides, the algorithm can be implemented conveniently and the computation speed is acceptable. Experimental comparison shows that the new algorithm has a better performance than the GPC （Gray-Level Plane Crossing） steganalysis method. And the new approach is also applicable for color images.