Target tracking algorithm for pedestrians movement based on kernel-correlation filtering

Abstract

Objective To reduce the effects of light changes, scale changes, local occlusion and other factors during target tracking, a kernel-correlation filtering (KCF) target tracking algorithm is introduced, which introduces the target block model.Results Comparative experiments of multiple mainstream algorithms on multiple data sets. Experimental results show that the algorithm has the highest accuracy and success rate, which are 11.89% and 15.24% higher than the KCF algorithm, respectively, indicating that the algorithm proposed in this paper possesses a more sensitive response to changes in illumination. Among them, factors such as scale change and local occlusion are more robust.