A new spatio-temporal background–foreground bimodal for motion segmentation and detection in urban traffic scenes

Automatic vehicle detection in urban traffic surveillance is an important and urgent issue, since it provides necessaryinformation for further processing. Conventional techniques utilize either motion segmentation or appearance-baseddetection, which involves either poor adaptation or high computation. The complexity of urban traffic scenarios lies in slowmotion temporarily stopped or parked vehicles, dynamic background, and sudden illumination variations. In this paper, anew motion segmentation technique is proposed based on spatio-temporal background–foreground bimodal. The temporalbackground information is modeled using a weighted sigma–delta estimation, cumulative frame differencing is used tomodel the foreground pixels, and the spatial correlation between neighboring pixels is utilized to combine both backgroundand foreground models. The median of consecutive frame difference adapts sudden illumination variation, update back-ground model, and reinitialize foreground model. Comparative experimental results for typical urban traffic sequencesshow that the proposed technique achieves robust and accurate segmentation, which improves adaptation, reduce falsedetection, and satisfy real-time requirements.

KeywordsMotion segmentation-Background subtraction-Cumulative frame differencing-Sigma–delta filter-Vehicle detection