A Secure Image Steganography Scheme Based on Kirsch Edge Detection and Adaptive Variable LSB Substitution

Abstract

Secure exchange of confidential data over open communication networks requires image data-hiding methods that preserve imperceptibility while providing sufficient payload and key-controlled access. This paper presents a Kirsch Edge Detector based Variable Data Hiding (KED-VDH) technique for spatial-domain image steganography. The proposed method first isolates complex regions of a cover image using the Kirsch edge detector and forms a binary edge pattern at threshold tau = 1200. Unlike fixed least significant bit (LSB) embedding, the method uses an 8-bit stego key to vary the number of message bits embedded in each selected edge pixel. Experiments were conducted on ten USC-SIPI cover images: Airport, Boat, House, Lena, Mandrill, Peppers, Sailboat, Splash, Stream Bridge, and Tank. The proposed key-controlled method was evaluated by hiding capacity, peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM). The highest average payload key achieved a reported hiding-capacity value of 4.3598 with an average PSNR of 41.9979 dB and an average SSIM of 0.9669. The lowest payload key produced an average PSNR of 67.0174 dB and SSIM of 1.0000. Results show that the proposed variable KED-DH approach provides a controllable trade-off between embedding payload and visual quality while keeping PSNR above 30 dB for all tested images.