A low cost foreground object detection architecture design with Multi-model Background Maintenance algorithm

This paper proposed an architecture design for a low cost foreground object detection based on Multi-model Background Maintenance (MBM) algorithm. The MBM algorithm framework basically contains two principal features. The principal features consist of static and dynamic pixels to represent the characteristic of background. Under this framework, a pure time-varying background image is maintained and learned using the statistical information of the multiple Gaussian distribution with principal features. In the MBM architecture, look-up table based Gaussian density function architecture is proposed. Three look-up tables are used for exponential and division of the Gaussian density function. The characteristic of Gaussian density function is also used to enormously reduce table size in a low cost consideration. The total gate count of the foreground object detection architecture design is about 14.4K gate with TSMC 0.18 μm technology. The maximum operation frequency of our design is up to 100MHz.