International Scientific Journals
of Scientific Technical Union of Mechanical Engineering "Industry 4.0"

Steganography aims to conceal confidential information within digital media while minimizing perceptual and statistical distortions. Among image steganographic techniques, adaptive embedding methods have demonstrated superior resistance to steganalysis by exploiting image regions with high local complexity. One effective approach utilizes the Canny edge detection algorithm to identify suitable embedding locations. However, the effectiveness of such methods depends not only on embedding capacity but also on comprehensive security evaluation. This paper presents a systematic framework for assessing the security of an adaptive Canny-based steganographic method. Multiple objective metrics, including Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), Mean Squared Error (MSE), histogram similarity, entropy variation, correlation coefficients, and resistance to statistical steganalysis are analyzed. The proposed evaluation framework demonstrates that edge-based adaptive embedding maintains high visual fidelity while significantly reducing detectability compared with conventional least significant bit (LSB) techniques.

The paper examines the role of visual signs, interface hierarchy, and emotional engagement mechanisms in mobile application design. Icons, pictograms, maps, menus, landmark distribution, and achievement screens are analyzed as components of an integrated visual-interactive system. These elements do not merely transmit information; they guide attention, reduce cognitive load, support orientation, and influence the user’s subjective experience. The study is based on theoretical foundations from semiotics, cognitive psychology, visual search research, human-computer interaction, and gamification. The paper outlines principles for designing mobile interfaces that are understandable, accessible, navigationally coherent, and emotionally engaging.

The report examines three main groups of approaches: geometric methods, topological techniques, and transformation-based methods. Their robustness, data hiding capacity, visual imperceptibility, and computational complexity are evaluated through analysis.
The results show that geometric methods offer high capacity but are more vulnerable to attacks such as smoothing and model simplification. Topological techniques provide greater robustness but have limited capacity. Transformation-based methods achieve a good balance between robustness and quality, albeit at the cost of higher computational expense. The report emphasizes that the choice of approach depends on the specific application, such as copyright protection or secure communication, and provides guidance for selecting the optimal method according to security and performance requirements.