Adaptive digital medical image watermarking approach in 2D-wavelet domain using speed-up robust feature method

Abstract

In the recent trends, the utilisation of digital images had got their importance in many fields. One of the major fields which had the more prominence utilisation of digital images is health care. The patient's information is stored in digital images for maintaining privacy and there is a need for preserving the content of the images. Watermarking serves well for protecting digital images. In this paper, Adaptive Digital Medical image watermarking approach is proposed. This proposed method utilises the 2D-Wavelet domain for converting the medical image in to number of sub bands. The interesting points are identified using the Speed-Up Robust Feature (SURF) method. The scaling parameter is calculated for the digital images using the bipolar sigmoid function. The control parameter is introduced to adjust the scaling parameter and it plays a crucial role in deciding the strength of the watermark. Experimental evaluation is carried out using three parameters, Peak Signal to Noise Ratio (PSNR), Structure SImilarity Measure (SSIM) and Normalised Correlation Coefficient (NCC). The experimental results proved that the proposed method had superior performance in both visible and invisible watermarking.