Image security enhancement to medical images by RDWT-DCT-Schur decomposition-based watermarking and its authentication using BRISK features

The growing use of the internet results in significant difficulties for the copyright protection of images. To resolve the copyright problem, a scheme is proposed that utilizes the advantages of Zigzag scanning (ZS), Schur decomposition (SD), Discrete cosine transform (DCT) and Redundant discrete wavelet transform (RDWT). In the proposed scheme, Henon map is applied on the watermarked image for its secure transfer and further enhancement of the security. Medical image security assurance is the need of the hour. Therefore, ultrasound images of the liver of six different patients are taken for experimental result analysis. The use of Singular value decomposition (SVD) is avoided here due to its false positive problem. RDWT is used to split the input ultrasound liver image into different sub-bands. Then DCT is applied on the suitable sub-band to improve the robustness of the presented work. The ZS is used to avoid artifacts which may occur at higher value of the gain factor. SD and RDWT are applied to the watermark logo (MNNIT logo) and selective pixel adding is used to embed the watermark logo into the host image. The authentication process is done by using binary robust invariant scalable keypoints (BRISK) features. The presented scheme perform satisfactorily against various attacks such as Histogram equalization, Region of interest (ROI) filtering, Cropping, Scaling, Rotation, Average and Median filtering attacks, Salt and Pepper, Gaussian noise attack, Flipping, etc. The proposed scheme shows better performance as compared to existing schemes in terms of various performance parameters such as Peak signal-to-noise ratio (PSNR), Structural similarity index measurement (SSIM) and Normalized correlation coefficient (NCC).

Automated summary

Due to the growing use of the internet, copyright protection of images faces significant difficulties. To address this issue, a scheme that combines Zigzag scanning, Schur decomposition, Discrete cosine transform, and Redundant discrete wavelet transform is proposed. The scheme also applies Henon map for secure transfer and enhanced security. Experimental analysis is conducted using ultrasound liver images of different patients, and the scheme demonstrates satisfactory performance against various attacks.