计算机工程 ›› 2012, Vol. 38 ›› Issue (21): 210-213.doi: 10.3969/j.issn.1000-3428.2012.21.056

• 图形图像处理 • 上一篇    下一篇

基于预测熵的JPEG图像无损数据隐藏

王巍巍1,童学锋1,宣国荣1,施云庆2   

  1. (1. 同济大学电子与信息工程学院,上海 201804;2. 新泽西理工学院电气与计算机工程系,新泽西 07102)
  • 收稿日期:2011-12-30 出版日期:2012-11-05 发布日期:2012-11-02
  • 作者简介:王巍巍(1987-),男,硕士研究生,主研方向:图像处理,模式识别;童学锋、宣国荣、施云庆,教授
  • 基金项目:
    国家自然科学基金资助项目(61175014);上海市教委科研创新基金资助项目(12ZZ033)

Lossless Data Hiding for JPEG Image Based on Forecast Entropy

WANG Wei-wei 1, TONG Xue-feng 1, XUAN Guo-rong 1, SHI Yun-qing 2   

  1. (1. School of Electronics and Information Engineering, Tongji University, Shanghai 201804, China; 2. Department of Electrical and Computer Engineering, New Jersey Institute of Technology, New Jersey 07102, USA)
  • Received:2011-12-30 Online:2012-11-05 Published:2012-11-02

摘要: 针对数据隐藏中的子块选择问题,提出一种基于预测熵的JPEG图像无损数据隐藏方法。利用直方图对法预测JPEG图像DCT系数块的熵,选择熵小的子块,并在中低频系数中嵌入数据。该方法可实现嵌入数据的盲提取和图像数据的无损恢复,且无需记录嵌入数据的子块位置。实验结果表明,与其他方法相比,该方法获得图像的峰值信噪比较高,数据隐藏性能较好。

关键词: 预测熵, JPEG图像, 无损数据隐, 直方图对, DCT系数, 中低频

Abstract: Aiming at the problem of choosing DCT coefficients blocks in data hiding, this paper proposes a lossless data hiding method for JPEG images based on minimum entropy. Utilizing the histogram pairs method, data are embedded into the low-middle frequency quantized coefficients of DCT blocks with the minimum forecast entropy. Without recording location map of selected blocks, hidden data can be extracted blindly and the original image can be reconstructed with no distortion. Experimental results show that the Peak Signal to Noise Ratio(PSNR) of this method is higher than other methods.

Key words: forecast entropy, JPEG image, lossless data hiding, histogram pairs, DCT coefficient, low-middle frequency

中图分类号: