首页 | 本学科首页   官方微博 | 高级检索  
     

基于压缩量化与邻域空间LBP算子的图像哈希算法
引用本文:王彦超.基于压缩量化与邻域空间LBP算子的图像哈希算法[J].包装工程,2017,38(21):191-198.
作者姓名:王彦超
作者单位:平顶山教育学院,平顶山,467000
基金项目:河南省科技计划(102102210416);河南省软科学研究计划(152400410323)
摘    要:目的为了解决哈希算法的感知鲁棒性与伪造检测能力不高的问题,提出基于特征压缩机制与邻域空间局部二值模式的紧凑图像哈希算法。方法首先利用2D线性插值技术,对输入图像进行预处理;嵌入Ring分割技术,将其变为二次图像;再利用Gabor滤波技术对其完成过滤;考虑到图像的颜色特征与其内在的空间关系,基于局部二值模式LBP设计邻域空间LBP算子,提取滤波图像的特征;构建特征压缩量化准则,输出紧凑的哈希二值数组;迭代Logistic映射,输出随机序列,通过量化每个序列值输出密钥流,通过构建动态引擎设计分段异加密模型,实现紧凑哈希序列的加密,获取图像哈希;最后计算原始哈希序列与待检测哈希序列的Hamming距离,实现图像信息的安全认证。结果与已有的哈希生成机制相比,文中算法所输出的哈希序列更紧凑,对旋转、伽马校正等篡改操作具有更好的感知鲁棒。结论所提哈希技术具备较高的安全性,在包装图标检索、信息检测等领域具有较好的价值。

关 键 词:图像哈希  特征压缩量化规则  邻域空间局部二值模式  分段异加密  Gabor滤波  决策阈值
收稿时间:2017/3/13 0:00:00
修稿时间:2017/11/10 0:00:00

Image Hashing Algorithm Based on Compression Quantization and Neighborhood Space LBP Operator
WANG Yan-chao.Image Hashing Algorithm Based on Compression Quantization and Neighborhood Space LBP Operator[J].Packaging Engineering,2017,38(21):191-198.
Authors:WANG Yan-chao
Affiliation:Pingdingshan Institute of Education, Pingdingshan 467000, China
Abstract:The work aims to solve the problem of low perceptual robustness and forgery detection ability of Hash algorithm and propose the compact image Hash algorithm based on the feature compression mechanism and local binary pattern (LBP) in the neighborhood space. Firstly, the 2D liner interpolation was introduced to preprocess the input image and such image was changed into a secondary image with the Ring segmentation technology. Then, the Gabor filter technique was used to filter the image. LBP operator in the neighborhood space was designed based on LBP by considering the color feature and the intrinsic spatial relationship of image for extracting the feature of the filter image. The feature compression quantization rule was constructed to output the compact Hash binary array. The Logistic mapping was iterated to output the random sequence. The key stream was generated by quantifying each sequence value to design the segment diffusion model by constructing the dynamic engine, so as to realize the encryption of compact Hash sequence and obtain the image Hash. Finally, the Hamming distance between the original Hash sequence and the Hash sequence to be detected was calculated, and the security authentication of the image information was realized. The test results showed that, compared with the existing Hash generation mechanism, the proposed algorithm was more compact and more robust to rotation, gamma correction and other tampering operations. The proposed Hash technique has higher security and better value in the fields of packaging icon retrieval and information detection, etc.
Keywords:image Hash  feature compression quantization rule  local binary pattern in neighborhood space  segment
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《包装工程》浏览原始摘要信息
点击此处可从《包装工程》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号