A reversible reduction method of medical images based on SLT-SVD

Abstract

The invention discloses a reversible reduction method of medical images based on SLT-SVD. This scheme transforms the whole original medical image from spatial domain to frequency domain to embed the watermark information. The watermark information is embedded without distinguishing ROI from RONI, avoiding the complete loss of watermark information after RONI is replaced, so as to identify the authenticity and integrity of the suspicious medical image shared from the network. Authentication of the source of medical images with copyright protection to ensure the correctness of suspicious medicalimages; integrity identification ensures that medical images are not distorted in the process of network transmission by detecting, locating and repairing the tampering of suspicious medical images.Moreover, the scheme realizes the reversible restoration of medical images without introducing a large amount of auxiliary information and without remembering the location information divided by ROI and RONI, which reduces the bandwidth of information transmission and reduces the security risks.

Description

technical field [0001] The invention belongs to the field of digital watermarking, and in particular relates to a method for reversibly restoring medical images based on SLT-SVD. Background technique [0002] With the development of Internet medical care, more and more sensitive and private medical images will be transmitted through open and interactive networks. While these medical images are conveniently shared, they also face new security challenges. On the one hand, medical images transmitted in the network may be unintentionally modified due to network instability; on the other hand, illegals can easily intercept and intentionally tamper with medical information. These unintentional and intentional modifications to medical images may lead to medical misdiagnosis, medical accidents, and endanger the health and even life of patients. Therefore, how to identify the integrity of the medical image content and ensure that the medical image content has not been modified is v...

AI Technical Summary

Problems solved by technology

[0005] In short, the above integrity identification algorithm has the following problems: 1) RONI cannot be reversibly restored, resulting in RONI distortion, which has a negative impact on the medical value of medical images; With RONI, in the attack environment, there will be a problem of the accuracy of ROI and RONI division. In order to solve this problem, it is necessary to remember the location information of ROI and RONI division, thus introducing a large amount of auxiliary information and increasing the information transmission bandwidth. and security risks; 3) Since ROI and RONI are spatially divided, RONI is easily replaced, resulting in complete loss of tamper detection information, tamper location information, and tamper repair information embedded in ROI generated based on ROI, making it impossible to complete 4) The algorithm cannot meet the requirements of robustness, reversibility and no need to carry a large amount of auxiliary information at the same time, and the practicability of the algorithm is poor

Images