TY  - EJOU
AU  - Boateng, Isaac Asare 
AU  - Amoah, Lord 
AU  - Adewale, Isogun Toluwalase 

TI  - Pairwise Reversible Data Hiding for Medical Images with Contrast Enhancement
T2  - Journal of Information Hiding and Privacy Protection

PY  - 2024
VL  - 6
IS  - 1
SN  - 2637-4226

AB  - Contrast enhancement in medical images has been vital since the prevalence of image representations in healthcare. In this research, the PRDHMCE (pairwise reversible data hiding for medical images with contrast enhancement) algorithm is proposed as an automatic contrast enhancement (CE) method for medical images based on region of interest (ROI) and non-region of interest (NROI). The PRDHMCE algorithm strategically enhances the ROI after segmentation using histogram stretching and data embedding. An initial histogram evaluation compares histogram bins with their neighbours to select the bin with the maximum pixel count. The selected bin is set as the point for contrast stretching with enhancement and secret data embedding in the ROI. The remaining data is embedded in the NROI while reducing image distortions. Experimental results show the effectiveness of PRDHMCE in optimally improving image contrast and increasing embedding capacity compared with existing methods based on qualitative and objective metrics such as peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), relative contrast error (RCE), relative mean brightness error (RMBE) and mean opinion score (MOS). Additionally, PRDHMCE recovers medical images fully without data loss.
KW  - Target histogram stretching; pairwise histogram embedding; reversible data hiding

DO  - 10.32604/jihpp.2024.051354