Recently, video surveillance systems have been greatly implemented due to the increasing demand for public security, real-time monitoring and crime prevention. Instead of applying manual monitoring by person in charge (operator), automatic detection is applied, in order to help operator and focus only on suspicious abnormal events. Background construction is the base of object detection and tracking for the machine vision system. Traditional background modeling methods often require complicated computations and are sensitive to illumination changes. This thesis proposes a new block-based background modeling method based on two distinct approaches; multiscale color and multiscale texture description, which fully utilizes the color and texture characteristics of each incoming frame. The proposed method is quite efficient and is capable of resisting illumination changes and shadow disturbance. Experimental results show that our method is suitable for real-world scenes and real-time applications.