Survey of Methods in the Spatial Domain Image Steganography based Imperceptibility and Payload Capacity

Abstract

Image steganography is a technique to hide messages in an image container with the aim of securing the message. Steganography is widely investigated in the spatial domain of images, with the most widely used methods being Least Significant Bit (LSB), Pixel Value Differencing (PVD) and Chinese Remainder Theorem (CRT). Some important aspects of steganography techniques are imperceptibility, payload capacity, data integrity, data security, and robustness, where each of these methods has advantages and disadvantages in each of these aspects. The imperceptibility aspect is strongly influenced by payload capacity, where the greater the payload, the lower the imperceptibility quality and vice versa. This research aims to compare the three methods based on the two most important aspects of the embedding process, namely imperceptibility and payload capacity. Whereas in the extraction process data integrity tests are carried out. Based on the results of testing, it has been proven that the LSB method has the best imperceptibility quality, while for the PVD capacity aspect it becomes the best. But after further testing, message embedding using 2-bits LSB can overcome the problem of overcapacity and can be compared with PVD with the same size of payload capacity. After further testing, it turns out that LSB has a good imperceptibility quality compared to PVD with when tested with the same maximum payload. Whereas for data integrity testing in the extraction process, all methods can be extracted perfectly without losing data bits. Then it can be concluded that the LSB method has the best imperceptibility quality, and by using the 2-bits LSB, capacity the LSB method can offset the payload capacity of the PVD method, whereas the imperceptibility quality in the LSB method is still superior.