Robust Image Steganography Based on Hybrid Edge Detection

Abstract

Steganography is a technique for hiding digital messages into other digital media, such as images, audio, video, and others so that the messages can be disguised and not directly visible to the human vision. There are two kinds of steganography by domain, i.e., spatial domain and frequency domain [7]. Least Significant Bit (LSB) is a popular method in spatial domains [8]. LSB is not a new method. However, this method has many advantages, such as simple algorithm and the quality of imperceptibility of relatively good stego-image [6, 9]. This makes the LSB still open for more development and further research. This paper presents the image hiding technique using Least Significant Bit (LSB) steganography based on hybrid edge detection that maximized the data embedding capacity of the cover image. The performance of the system is evaluated for hybrid edge detection using pairs of various edge detection techniques such as Canny, Kirsch, Sobel, and Prewitt. The results of the proposed steganography approach are estimated for different payload capacities based on Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR), Structural Similarity Index Metrics (SSIM), and Universal Image Quality Index (UQI). It is examined that the proposed approach performs better than the existing state of arts for image steganography.