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.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

MLR-PV: Radical Nonlinear Image Steganography Approach with Regression Analysis Validation

  • Abhijit Sarkar,
  • Sabyasachi Samanta,
  • Soumen Ghosh,
  • Shyamalendu Kandar

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.