Abstract: A new method based on multiple background model for moving target detection on the sea surface is proposed. It can detect the moving target on the sea surface when there is heavy sea clutter and the reflected light is strong. The monitored area is first partitioned into two regions using the statistical model-based change detection method, where the sea clutter in one region is heavier than the other. Then Weibull distribution is used to create the background model for pixels in the region having heavier sea clutter and Gaussian distribution is used for the other region. Finally the background model is used to detect moving target on the sea surface. Experimental results show that the proposed method can detect moving target effectively even when the sea clutter is heavy and the reflected light is strong.