基于Lambert光照模型的图像真伪盲鉴别算法  

Blind Identification for Image Authenticity Based on Lambert Illumination Model

作　　者：陈海鹏[1,2] 申铉京[1,2] 吕颖达[1,2] 金玉善[1,2] 

机构地区：[1]吉林大学计算机科学与技术学院长春130012 [2]符号计算与知识工程教育部重点实验室吉林大学长春1300121

出　　处：《计算机研究与发展》2011年第7期1237-1245,共9页Journal of Computer Research and Development

基　　金：国家自然科学基金项目(60973090);吉林省科技发展计划基金项目(20050305);吉林大学研究生创新基金项目(20111063)

摘　　要：依据Lambert光照模型,提出了基于Lambert光照模型的图像真伪盲鉴别算法.针对无限光源图像和局部光源图像,根据其实际的光强度与计算出的光强度之间的误差函数以及光源对光线的约束函数,分别提出了相应的算法用以计算图像中不同区域的光源方向,并根据这些不同区域的光源方向是否一致判定图像是否被篡改,从而对图像真伪进行盲鉴别.实验结果及分析表明,针对这2种图像各自提出的光源方向计算方法,能够准确地计算出图像中不同区域的光源方向,从而有效地鉴别出图像的真伪.Based on Lambert illumination model, blind identification algorithm for image authenticity is proposed. For an image, there is an error function between its actual light intensity and calculated light intensity; and for different light source models, there are different constraint functions of light source to light beam. Therefore, for infinite light source images and local light source images, a corresponding algorithm is presented, which calculates the light source direction of different areas in an image according to the error function and the corresponding constraint function. And the image tampering can be determined by the inconsistency in light source direction of different areas in an image. That is to say, if light source directions of different areas in an image are inconsistent, it can be considered that the image has been tampered;on the other hand, if light source directions of different areas in an image are consistent, it can be considered that the image has not been tampered. This has realized blind identification for image authenticity. Experimental results show that the light source direction of different areas in an image could be calculated accurately by the proposed corresponding algorithms for the two kinds of images mentioned above. Thus, the image authenticity could be identified effectively.

关 键 词：Lambert光照模型 无限光源图像 局部光源图像 光源方向 图像真伪盲鉴别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]