Abstract: This paper proposes a new method for moving objects (foreground) detection with non-stationary background using background subtraction. While background subtraction has traditionally worked well for stationary backgrounds, the same cannot be implied for a nonstationary viewing sensor. To a limited extent, motion compensation for non-stationary backgrounds can be applied, but in practice, it is difficult to realize the motion compensation to sufficient accuracy and the background subtraction algorithm will fail for a moving scene. The problem is further compounded when the moving target to be detected/tracked is small, since the pixel error in motion compensating the background will subsume the small target. A Spatial Distribution of Gaussians (SDG) model is proposed to deal with moving object detection having motion compensation which are only approximately extracted. The distribution of each background pixel is temporally and spatially modeled; a pixel in the current frame is then classified based on this statistical model. The emphasis of this approach is on the robust detection of moving objects even with approximately accurate motion compensation, noise, or environmental changes. Test cases involving the detection of small moving objects with a highly textured background and a pan-tilt tracking system are demonstrated successfully.