基于四叉树分割的JPEG隐写分析



当前主要的隐写分析方法都是对整幅图像进行特征提取，而忽略了图像的内容差异。该文提出一种 基于四叉树分割的 JPEG 隐写分析方法，该方法根据图像块的纹理复杂度进行图像分割，对具有相同统计特性的子图像分别进行隐写检测特征的提取，并构造相应的分类器，通过加权融合得 到最终的检测结果。实验结果表明该方法具有良好的性能，尤其是在训练与测试图像的统计特性具有较大差异时，该算法的检测准确率提高更加明显。

-Based on quadtree segmentation JPEG steganalysis current major steganalysis methods are on the whole image feature extraction, while ignoring the differences in the content of the image. This paper presents a quadtree segmentation based on JPEG steganography analysis, the method according to the complexity of the image blocks of texture image segmentation of sub-images have the same statistical characteristics were steganography detection feature extraction, and configured accordingly classifier, to get the final test results by weighted fusion. Experimental results show that the method has good performance, especially with a large difference in the statistical properties of the training and test image when the algorithm to improve detection accuracy is more obvious.