Image Foreground Extraction by ICA-PSO and Hilbert-Schmidt Independence Criterion

Foreground extraction from an image, even by using the reference background is still a challenging problem due to the temporal variations that might occur for the test image, such as background illumination changes and object movements. In this paper, a novel structure has been presented to address this problem using a modified version of ICA algorithm which leverages the Hilbert-Schmidt Independence Criterion (HSIC) instead of the common objective functions. Moreover, the unmixing matrix elements of ICA are extracted through a Particle Swarm (PSO) evolutionary algorithm in a much faster way. The experimental results clearly show the outperformance of the proposed structure over the original works being cited among the references, using Wallflower dataset.