In real time video processing, a trivial task is to detect the changes in multiple images on the same scene of a real time instant. The task is not only trivial but also very indispensable as it brings into play a great number of diversified focuses area application such as, remote sensing, surveillance, etc. The general processing steps and prime decision rules used in the advanced change detection algorithms, which are employed for the video surveillance in hardware implementation. The real time video surveillance includes the analytical and the background modeling techniques. In background modeling lot of efficient algorithms are there, from that ViBe method is one of the famous and emergent algorithm. The technique in ViBe is formulation background model which collect twenty background frames and then measure distance between current frame and background model. For that ViBe uses Euclidean distance (L2 norm). In our proposed method, distance can be measured by Manhattan distance (L1 norm), from this way the method achieves very less clock cycle and register memory used for an execution of a statement. Compared to the L2 norm and other methods, the performance of proposed method is improved by system speed and memory of the system. © 2016 Academic Publications, Ltd.