Abstract:

This study explores spatial domain video steganography, a technique for concealing information by modifying pixel color values across consecutive video frames. The effectiveness of this approach relies on embedding these modifications within the inherent noise present in video content, originating from inter-frame differences, compression artifacts, and processing variations. To inform and improve steganographic techniques, this research analyzes the statistical properties of naturally occurring pixel color differences in frames.

A mathematical model for video files was developed, enabling the definition of a function to calculate color differences between consecutive frames. Using this function, a formal description of a color difference histogram was created, providing a distinctive metric to characterize video files based on the frequency of color variations between frames. The study’s theoretical contributions include a formalized approach to color difference analysis in videos, offering a framework for examining color variations in video content.

Practically, this model can aid in the development and refinement of spatial domain steganography techniques that leverage pixel color differences in the RGB space for encoding information. Future research will apply the model to diverse video files, examining color profiles and motion dynamics to validate its applicability and reveal insights into natural color variation patterns across different types of video content.