基于无损数字水印的医学图像安全技术研究
发布时间：2018-08-04 19:11
【摘要】：网络技术的高速发展与医疗数字影像传输及通信(DICOM)标准的建立,在促进数字医疗影像信息的交流的同时,也带来了医学数据在传输中可能出现的丢失或被篡改等信息安全问题。因此,对医学数据的真实性及完整性进行认证是十分重要的。数字水印技术作为一种可将隐私信息直接嵌入载体信息中的新兴信息安全技术,被广泛应用于数字信息保护和认证等领域。为能够对医学图像的完整性进行认证,且不引入任何失真,设计出一种能无损恢复原始图像,并能对图像进行完整性认证的无损水印算法具有重要的意义。 本文主要成果性工作包括： (1)提出一种将压扩技术、差值扩展技术与纠错码技术结合的水印算法。该算法利用了图像中相邻图像间的灰度值相关性,通过对相邻像素的差值扩展来嵌入信息。并使用压扩技术实现对图像失真的控制,及提高差值扩展过程中得到的二值位置图压缩率,从而达到增大水印嵌入容量的目的。同时使用纠错码的纠错特性提高图像整体的自修复能力。 (2)在基于区域的医学图像无损水印算法上进行改进,增添篡改恢复功能。该算法先确定感兴趣区域,并将其作为认证区域,获取认证信息作为水印信息的一份子；在感兴趣区域外(非感兴趣区域)自动选取嵌入区域；通过对相邻像素进行差值扩展,实现非感兴趣区域内水印的嵌入；提取水印后,通过认证信息的比较实现篡改定位检测,通过纠错码技术实现篡改恢复的功能。最终,嵌入水印后图像失真水平可通过输入阈值进行调节。与前一种算法相比无需嵌入压缩后的标识二值图像,水印容量不再受二值图像压缩率影响,并在一定程度上能实现水印容量的提高,且该方法能将嵌入水印引起的失真限制在感兴趣区域外,嵌入水印后的图像也能直接用于医学诊断或其他医学应用,方便了嵌入水印后医学图像的实际使用。
[Abstract]:The rapid development of network technology and the establishment of (DICOM) standard for medical digital image transmission and communication not only promote the exchange of digital medical image information, but also bring about information security problems such as the loss or tampering of medical data in transmission. Therefore, it is very important to authenticate the authenticity and integrity of medical data. As a new information security technology which can embed privacy information directly into carrier information, digital watermarking technology has been widely used in digital information protection and authentication. In order to authenticate the integrity of medical image without any distortion, it is of great significance to design a lossless watermarking algorithm that can restore the original image without loss and can authenticate the integrity of the image. The main achievements of this paper are as follows: (1) A watermarking algorithm combining compression spread, difference spread and error correction code is proposed. The algorithm makes use of the gray correlation between adjacent images and embeds information by extending the difference between adjacent pixels. The compression spread technique is used to control the image distortion and to increase the compression ratio of the binary position map in the process of the difference expansion, so as to increase the embedding capacity of the watermark. At the same time, the error-correcting characteristics of error-correcting codes are used to improve the overall self-repairing ability of the image. (2) to improve the lossless watermarking algorithm of medical image based on region, and to add the tampering recovery function. The algorithm first determines the region of interest and takes it as the authentication region to obtain the authentication information as a part of the watermark information; automatically selects the embedded region outside the region of interest (not the region of interest); and extends the difference between adjacent pixels. After extracting the watermark, the tamper location detection is realized by comparing the authentication information, and the tampering recovery function is realized by error correction code technology. Finally, the image distortion level after embedding watermark can be adjusted by input threshold. Compared with the former algorithm, the watermark capacity is no longer affected by the compression ratio of the binary image, and the watermark capacity can be improved to a certain extent. The method can limit the distortion caused by the embedded watermark to the region of interest, and the embedded image can also be directly used in medical diagnosis or other medical applications, thus facilitating the practical use of the medical image after embedding the watermark.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318
本文编号：2164866
[Abstract]:The rapid development of network technology and the establishment of (DICOM) standard for medical digital image transmission and communication not only promote the exchange of digital medical image information, but also bring about information security problems such as the loss or tampering of medical data in transmission. Therefore, it is very important to authenticate the authenticity and integrity of medical data. As a new information security technology which can embed privacy information directly into carrier information, digital watermarking technology has been widely used in digital information protection and authentication. In order to authenticate the integrity of medical image without any distortion, it is of great significance to design a lossless watermarking algorithm that can restore the original image without loss and can authenticate the integrity of the image. The main achievements of this paper are as follows: (1) A watermarking algorithm combining compression spread, difference spread and error correction code is proposed. The algorithm makes use of the gray correlation between adjacent images and embeds information by extending the difference between adjacent pixels. The compression spread technique is used to control the image distortion and to increase the compression ratio of the binary position map in the process of the difference expansion, so as to increase the embedding capacity of the watermark. At the same time, the error-correcting characteristics of error-correcting codes are used to improve the overall self-repairing ability of the image. (2) to improve the lossless watermarking algorithm of medical image based on region, and to add the tampering recovery function. The algorithm first determines the region of interest and takes it as the authentication region to obtain the authentication information as a part of the watermark information; automatically selects the embedded region outside the region of interest (not the region of interest); and extends the difference between adjacent pixels. After extracting the watermark, the tamper location detection is realized by comparing the authentication information, and the tampering recovery function is realized by error correction code technology. Finally, the image distortion level after embedding watermark can be adjusted by input threshold. Compared with the former algorithm, the watermark capacity is no longer affected by the compression ratio of the binary image, and the watermark capacity can be improved to a certain extent. The method can limit the distortion caused by the embedded watermark to the region of interest, and the embedded image can also be directly used in medical diagnosis or other medical applications, thus facilitating the practical use of the medical image after embedding the watermark.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318
【参考文献】
相关期刊论文 前5条
1 吕林涛;张玺;;改进的大容量可逆数字水印嵌入算法[J];计算机工程;2011年22期
2 全海英,杨源,张歆东,郭树旭,刘景鑫;DICOM数据集与DCM文件格式[J];计算机应用;2001年S1期
3 肖国强，钟华;逐步优化的二值图像压缩算法研究[J];计算机应用研究;1994年06期
4 阙大顺;陈碧;陈智翔;;可逆数字水印在医学图像中的应用研究[J];武汉理工大学学报(信息与管理工程版);2008年06期
5 钟晓燕;卜祥磊;刘文军;冯前进;;数字水印在医学影像存储与传输系统中的研究与应用[J];医疗装备;2010年06期
相关博士学位论文 前2条
1 翁韶伟;数字图像的高容量可逆水印的研究[D];北京交通大学;2009年
2 武丹;可逆数字水印技术的研究[D];浙江大学;2009年
，本文编号：2164866
本文链接：https://www.wllwen.com/yixuelunwen/swyx/2164866.html
