An optimized Vibe target detection algorithm based on gray distribution and Minkowski distance

Aiming at the ghost problem in Vibe algorithm, a fast optimization method based on gray histogram and Minkowski distance was proposed. Firstly, the pixels identified as foreground were labeled and the foreground blocks were connected by morphological processing. Then, the histogram of the current foreground block and the corresponding region of the specified frame were counted respectively. Finally, the similarity between the two gray-scale distributions was measured by Minkowski distance, and compared with the dynamic threshold to determine ghost area. In the target detection experiment, the algorithm could remove the ghosts in the Vibe, restrain the jitter disturbances in the background and reduce the false detection rate of the algorithm while preserving the fastness and real-time performance of the traditional Vibe algorithm, and the algorithm processing speed was between the original Vibe algorithm and mixed Gaussian method, the running time of each frame was 20ms. The experimental results show that this method is feasible for hardware system with real-time requirement.