MLR-PV: Radical Nonlinear Image Steganography Approach with Regression Analysis Validation
摘要
Steganography conceals information in apparently unimportant data. Steganography aims to hide information from observers. Due to the extensive use of internet and multimedia technologies, steganography’s scope of use has grown substantially. Current internet data transfer is insufficiently secure. The article proposes a non-linear embedding and pseudorandom key-based image steganography method for data security. This method is used for maintaining security, visual quality. Secret text had been hidden in the image’s edge intensity in a non-linear bit position manner. Parallel multiple bit embeddings in non-linear places have been compared for their effect on the major metrics. Furthermore, the validity of the suggested approach is proven using a Multiple Linear Regression based Prediction and Validation(MLR-PV) model analysis. Extensive evidence for a variety of outcomes has been presented, and an accuracy-analysis has been accomplished for the justification of proposed methodology using MLR-PV model with robustness analysis against attacks.