Compressive imaging: An application for surveillance systems

In this paper, we proposed an application of compressive imaging system to the problem of wide-area video surveillance system. A motion target detection algorithm in video using compressive image data is developed. For background subtracted compressive images, Kullback-Leibler divergence is applied in compressed image field to detect motion objects which are not part of the background model. Foreground image retrieval from underdetermined measurements using total variance optimization algorithm are explored. Simulation results show that compared with the traditional optical system, compressive imaging system can dramatically reduce sampling costs, energy consumption and alleviate communication and storage burdens with comparable image performance. Low dimensional compressed imaging representation is sufficient to determine spatial motion targets.