摘要
通过对图像拼接技术特点的分析,提出一种基于图像纹理特征分析和马尔科夫模型的改进的拼接图像检测算法。该算法计算图像DCT域上的马尔科夫转移概率矩阵,同时对图像进行纹理分析,得到两类特征共178维。为评估该检测算法的性能,提出了一个具体实现方案,提取了图片数据集的特征,使用支持向量机(Support Vector Machine,SVM)对特征数据进行训练与分类。实验表明,该方法取得了较好的分类效果。
Through analysis on the characteristic of image splicing, an improved splicing blind detection scheme based on image texture analysis and Markov model is proposed. It consists of two groups of features extracted from the Markov transition probability matrix of the image and the texture of the image, and 178 features are obtained. In order to evaluate the performance of this scheme, a concrete implementation of this scheme is presented, and the feature of the image data set extracted, and kernel-based support vector machine（SVM） is chosen as a classifier to train and classify the feature data. The experiment results show that the proposed scheme could achieve a fairly good classification result.
出处
《信息安全与通信保密》
2010年第1期76-78,共3页
Information Security and Communications Privacy
基金
国家自然科学基金项目(批准号:60772098
60772042)
教育部新世纪优秀人才支持计划项目(编号:NCET-0600393)
国家863项目(编号:2007AA01Z455)
2007年上海市曙光计划项目
关键词
灰度共生矩阵
马尔科夫过程
SVM分类
grey-level co-occurrence matrix
Markov process
SVM classification
