An Enhanced ROI-Based Image Steganography Method for Secure Data Hiding

Abstract

Image steganography is an important technique for secure communication that hides confidential information inside digital images without producing noticeable visual distortion. Traditional image steganography methods often suffer from poor robustness and low imperceptibility when hidden data is uniformly embedded throughout the image. To overcome these limitations, this research proposes an ROI-based adaptive image steganography method using edge detection and Least Significant Bit (LSB) embedding techniques.
In the proposed approach, Regions of Interest (ROI) are identified using edge-based analysis and texture characteristics to determine suitable embedding locations. Secret data is selectively embedded within ROI regions to improve security, payload capacity, and resistance against statistical steganalysis attacks. The embedding process minimizes visual distortion while maintaining high image quality.
Experimental evaluation was conducted using standard grayscale and color image datasets. Performance analysis was carried out using Peak Signal-to-Noise Ratio (PSNR), Mean Square Error (MSE), and Structural Similarity Index Measure (SSIM). The proposed ROI-based method achieved higher PSNR and SSIM values with lower MSE compared to conventional LSB and edge-based steganography techniques. The experimental results demonstrate that the proposed method provides improved imperceptibility, robustness, and secure data hiding performance while preserving image quality.
The proposed technique can be effectively applied in secure medical image transmission, confidential communication, digital watermarking, and military information security applications.