| 支持向量机在图像隐秘检测中的应用 |
| |
| 引用本文： | 杨晓元,王志刚,王育民.支持向量机在图像隐秘检测中的应用[J].西安电子科技大学学报,2005,32(3):457-459. |
| |
| 作者姓名： | 杨晓元 王志刚 王育民 |
| |
| 作者单位： | [1]武警工程学院电子技术系网络与信息安全重点实验室，陕西西安710086／／西安电子科技大学综合业务网国家重点实验室，陕西西安710071 [2]武警工程学院电子技术系网络与信息安全重点实验室，陕西西安710086 [3]西安电子科技大学综合业务网国家重点实验室，陕西西安710071 |
| |
| 基金项目： | 国家自然科学基金资助项目(60073052)，信息安全教育部重点实验室资助课题，武警部队科研项目(wj200406) |
| |
| 摘 要： | 研究了支持向量机的学习算法，提出了基于支持向量机的图像隐秘检测算法，选取了两种隐秘软件F5r11和Jsteg4．1进行了大量的隐秘检测实验．通过实验发现，二次规划函数中惩罚因子C的选取对识别率影响较大，给出了不同C值之下的检测结果．实验结果证明，该算法的识别率较Fisher线性判别算法有了明显提高．
|
| 关 键 词： | 图像隐秘检测 支持向量机 统计学习理论 模式分类 |
| 文章编号： | 1001-2400(2005)03-0457-03 |
The application of support vector machines in detection of images steganography |
| |
| YANG Xiao-yuan.The application of support vector machines in detection of images steganography[J].Journal of Xidian University,2005,32(3):457-459. |
| |
| Authors: | YANG Xiao-yuan |
| |
| Affiliation: | (1. Network and Information Security Key Lab., Electronics Dept. of Eng. College of the Armed Police Forces, Xidian Univ., Xi′an 710086， China; 2. State Key Lab. of Integrated Service Networks, Xidian Univ., Xi′an 710071， China) ; |
| |
| Abstract: | The Support Vector Machines is a new machines learning algorithm based on the Statistical Learning Theory. It has found important applications in pattern discrimination of high dimensions Character vectors. The algorithm of Support Vector Macnines has been studied in this paper, and the detection algorithm of steganography based on the Support Vector Machines has been put forward. For this detection algorithm, we select two kinds of steganography software F5r11 and Jsteg4.1 to make a large number of experiments. In quadratic programming, we find that the penalty gene C is important to experimental results, and therefore, we show different detection results of different C's, and the results show that the detection rate of this algorithm improves evidently compared with the Fisher Linear Discrimination. |
| |
| Keywords: | detection of image steganography support vector machines statistical learning theory pattern class |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《西安电子科技大学学报》浏览原始摘要信息 |
|
点击此处可从《西安电子科技大学学报》下载全文 |
