Identifying moving objects from a video sequence is a fundamental and critical task in many computer-vision applications. For surveillance video captured by static camera

the common approach is to perform background subtraction

which identifies moving objects from the portion of a video frame that differs significantly from a background model. However

there are many challenges in developing a good background subtraction algorithm for many factors such as changes in illumination should be considered in constructing a background model. This paper firstly expounds the universal process of detecting objects with background subtraction

all typical background modeling algorithms and their merits. Then their whole characteristics are summarized and the performances of some algorithms are compared based on experiments and other literatures. Finally

key issues and directions of future study in this area are pointed out.