可信JPEG XR图像采集研究与实现

Research and realization of credible JPEG XR image acquisition

文章在原始图像中加入定位信息和时间信息并编码为JPEG XR图像,采用半脆弱水印技术实现图像的鲁棒性认证和内容认证,构造可信的JPEG XR图像采集系统。由CMOS采集原始图像,借助GPS全球定位系统获取定位信息和时间信息添加在该图像合适位置,被编码为JPEG XR频率模式码流结构,提取码流中的DC系数和LP系数构造基于图像内容的二值特征信息作为水印,置乱后嵌入到JPEG XR解码后的图像对应像素的Y分量的最低有效位中,然后再对含有水印的图像进行JPEG XR编码实现水印嵌入;通过比较基于图像的二值信息和嵌入在图像的二值水印信息进行篡改定位;在S3C2410A上移植定制并优化的JPEGXR编解码算法和半脆弱水印算法并进行测试。

An original image is added with location information and time information and then encoded as a JPEG XR image, and the robust authentication and content authentication of the image are achieved by the semi-fragile digital watermarking technology to construct a credible JPEG XR image collection system. The right location of the original image acquired by CMOS is added with location information and time information by the global positioning system（GPS）, which is encoded as JPEG XR frequency mode code stream structure. Then the DC coefficients and the LP coefficient of the code stream are used to construct the information of the image-based content as binaxy digital watermarking, which are scrambled and embedded in the Y component of the least significant bit of correspond- ing pixels of the decoded JPEG XR image, and the image is encoded by JPEG XR to realize the watermark embedding. By comparing the binary digital information of the image-based content with the binary digital information of the embedded watermark, the tamper positioning is made. The JPEO XR code and semi-fragile watermarking algorithms are customized and optimized based on S3C2410A, then ported on 83C2410A and tested.