This study aims to develop a system to instantly detect traffic condition by using computer vision with RFID technology for obtaining vehicles data. The system detects the presence of vehicles and also calculates the detected objects. The method of background subtraction on motion objects detection is proposed. By taking samples from the objects, background models are being set and extracting foreground areas from background subtraction and reducing the shadow in the foreground. A threshold method is presented to divide standard to the image and improve the detection effect. The experiment results indicate that the performance of the background subtraction method is more stable on dynamic objects detection. This advanced method can contribute the smart traffic control and analysis for smart city development.