为提高图像隐写率,在YASS算法的基础上,提出一种改进的图像隐写算法。分析不同图像分块离散余弦变换系数2_f.q的固有联系,借鉴整数规划的思想设计迭代算法,对已隐写的载体进行数据校正。使用98维马尔科夫通用图像检测方法进行实验,结果表明,该算法能减少YASS隐写过程中由JPEG图像压缩引入的误码,提高隐写率,同时可有效抵抗通用隐写检测。
In order to increase the steganography rate, this paper proposes a new scheme of steganography which is based on YASS algorithm. After discussing the inherent relationship between Discrete Cosine Transform(DCT) coefficients in different blocks, a method of iterative programming is proposed which is referenced to integer programming, and the carrier which has steganography has been corrected. In aspect of anti-detection, 98 degree's Markov steganalysis method is used to detect the stego images. Experimemal result shows that this optimization algorithm is proved to effectively decrease the errors caused by JPEG compression during YASS steganography. In this way, steganographic rate is increased to a great extent. This method is also proved to get the same outstanding performance in common anti-detection.