Implementation Using the Least Significant Bit (LSB) and Pixel Value Differencing (PVD) Methods in Message Insertion in Digital Images

Abstract

The development of information technology increases the need for digital data protection against the risk of wiretapping, alteration, and unauthorized access. Steganography is one solution by hiding messages in digital images without significantly changing the visual appearance. This study aims to implement the Least Significant Bit (LSB) and Pixel Value Differencing (PVD) methods in the process of embedding and extracting SQL files in digital images and to compare the image quality and integrity of the resulting messages. This study uses an applied research type with a system implementation method. The research population is a digital image as a steganography medium, while the sample consists of 50 digital images in PNG and BMP formats and SQL files as secret messages. The research instrument is a web-based steganography application developed using Python and Streamlit. Data analysis was carried out through functional testing of the embedding and extraction processes, message integrity evaluation, and image quality measurement using Mean Square Error (MSE) and Peak Signal-to-Noise Ratio (PSNR). The results show that both methods produce high-quality stego images as indicated by high PSNR and low MSE values. The LSB method maintained message integrity throughout the test, while PVD still experienced character changes in some of the extracted results. It was concluded that the LSB method is more reliable than PVD for embedding SQL files into digital images because it is able to maintain image quality while preserving message integrity