基于QDCT马尔科夫方法的彩色图像拼接检测
发布时间：2019-02-18 18:25
【摘要】：提出一种基于四元数离散余弦变换(quaternion discrete cosine transform,QDCT)的拼接检测方案.该方案充分利用了彩色图像三通道之间的相关性,减少了图像彩色信息的损失.将彩色图像进行QDCT得到变换系数,然后使用四元数马尔科夫方法提取特征,最后用支持向量机检测拼接图像.使用图像库CASIA TIDE v1.0和CASIA TIDE v2.0进行实验,所得准确率分别达到98.75%和96.78%,表明该方法优于大部分现有方法.
[Abstract]:A splicing detection scheme based on quaternion discrete cosine transform (quaternion discrete cosine transform,QDCT) is proposed. This scheme makes full use of the correlation between the three channels of color images and reduces the loss of color information. The transform coefficients are obtained by QDCT, then the features are extracted by quaternion Markov method, and finally the stitched images are detected by support vector machine (SVM). The experimental results using image library CASIA TIDE v1.0 and CASIA TIDE v2.0 show that the accuracy of this method is 98.75% and 96.78% respectively, which indicates that this method is superior to most existing methods.
【作者单位】： 南京信息工程大学计算机与软件学院;
【基金】：国家自然科学基金(No.61772281,No.61272421,No.61232016,No.61402235,No.61502241) 江苏省自然科学基金(No.BK20141006) 江苏高校优势学科建设工程项目 大气环境与装备技术协同创新中心基金资助
【分类号】：TP391.41
[Abstract]:A splicing detection scheme based on quaternion discrete cosine transform (quaternion discrete cosine transform,QDCT) is proposed. This scheme makes full use of the correlation between the three channels of color images and reduces the loss of color information. The transform coefficients are obtained by QDCT, then the features are extracted by quaternion Markov method, and finally the stitched images are detected by support vector machine (SVM). The experimental results using image library CASIA TIDE v1.0 and CASIA TIDE v2.0 show that the accuracy of this method is 98.75% and 96.78% respectively, which indicates that this method is superior to most existing methods.
【作者单位】： 南京信息工程大学计算机与软件学院;
【基金】：国家自然科学基金(No.61772281,No.61272421,No.61232016,No.61402235,No.61502241) 江苏省自然科学基金(No.BK20141006) 江苏高校优势学科建设工程项目 大气环境与装备技术协同创新中心基金资助
【分类号】：TP391.41
【相似文献】
相关期刊论文 前10条
1 金圣华;周玮;;马尔科夫蒙特卡洛在视网膜血管分割中的应用[J];长沙大学学报;2011年05期
2 曹容菲;张美霞;王醒策;武仲科;周明全;田l，
本文编号：2426078
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2426078.html
