A novel particle swarm tracking system based on chromatic co-occurrence matrices

Abstract

Moving object tracking in video sequences becomes an active research field due to its application in various domains. This work proposes a PSO algorithm based on new chromatic co-occurrence matrices descriptor in order to track objects under a dynamic environment. The use of the co-occurrence matrices will give us the capability to exploit the information about the texture of the target objects. The qualitative and quantitative studies on newest benchmark demonstrate that the obtained results are very competitive in comparison with many recent state-of-the-art methods.