No abstract

No abstract

An exposure guided background subtraction (EGBS) model is proposed for smart cameras to handle illumination change due to auto-exposure in visual surveillance. To reduce false foreground pixels caused by auto-exposure, EGBS compensates background illumination directly utilizing the information generated by auto-exposure module without extra illumination change estimation. Hence, it is very preferable for smart camera without any extra hardware resources. Experimental results indicate the proposed model efficiently reduces false foreground pixels caused by auto-exposure.

No abstract