Real-time robust background subtraction under rapidly changing illumination conditions

Fast robust background subtraction under sudden lighting changes is a challenging problem in many applications. In this paper, we propose a real-time approach, which combines the Eigenbackground and Statistical Illumination method to address this issue. The first algorithm is used to reconstruct the background frame, while the latter improves the foreground segmentation. In addition, we introduce an online spatial likelihood model by detecting reliable background pixels. Extensive quantitative experiments illustrate our approach consistently achieves significantly higher precision at high recall rates, compared to several state-of-the-art illumination invariant background subtraction methods.

Graphical abstractFigure optionsDownload full-size imageDownload high-quality image (72 K)Download as PowerPoint slideHighlights► Robust background subtraction under rapidly varying illumination conditions ► Eigenbackground based Statistical Illumination (ESI) ► Effective online spatial likelihood model without training ► 10–15% higher recall on challenging datasets than state-of-the-art ► Real-time 18 frames per second at a resolution of 320 × 240 on a CPU