Medical Image Watermarking using Spread Spectrum and Compressive Sensing Techniques

In the current era, almost everyone has used digital media to exchange information. One example of the use of digital media in the medical field is the Electronic Patient Record (EPR). However, the use of digital media in EPR raises concerns about the authenticity and security aspects of patient data. The authenticity of medical images is one of the important data and must be considered. One way that can be done to minimize the possibility of various threats is to use a watermark. Another problem that arises is the size of the medical image. This study proposes a watermarking system for medical images using the Compressive Sensing method based on Singular Value Decomposition (SVD) which can be used to reduce the size of medical images. Compressive Sensing includes lossless compression which is a way to compress the size of a data without losing the original information. The medical image watermarking process is divided into two processes, namely the insertion process and the extraction process. The insertion process is carried out using the Spread Spectrum method with a Gaussian distributed PN code key, resulting in a watermarked image. While the extraction process is carried out by reconstructing the image using the Orthogonal Matching Pursuit (OMP) method, SVD reconstruction. The experimental results obtained BER values with an average of 0 when the Gaussian noise attack was carried out. PSNR value range is 30-38 dB and compression ratio is 0.2656.