Simulation of Basis Pursuit Algorithm for Video Compressive Sensing Systems

Based on video compressive sensing systems, we have developed a target tracking-based algorithm. This algorithm not only achieves video compression but also enables target tracking during the compression process. To validate algorithm effectiveness, we conducted simulation experiments implemented through MATLAB/Octave code that processes video frames using sparse representation techniques. The simulation results demonstrate successful algorithm execution with key functions including frame segmentation, compressive measurements via random projection matrices, and ℓ1-minimization using basis pursuit solvers. Furthermore, we have explored practical applications in scenarios such as video surveillance and autonomous driving systems, where the algorithm efficiently processes video streams while maintaining tracking accuracy. Our ongoing research focuses on optimizing computational efficiency through iterative thresholding methods and implementing real-time processing capabilities for practical deployment in industrial and daily life applications.