This paper presents a thermal features’ relative angle method to separate vehicles for nighttime traffic. The objective of research is to identify and separate vehicles that have ambiguous thermal features. The features comprising of windscreen, engine heat and elsewhere, are depicted as different intensities in each thermal image. Moreover, the relation between the three features observed from the set-up camera can be illustrated as an angle of windscreen to heat, which differentiates one group of vehicles from another. There are three operating modules proposed in this research, comprising vehicle-thermal features detection, relative angle calculation and seperation, respectively. Initially, the detecting process finds a vehicle engine heat-feature by subtracting the road's intensity, selecting a suspected area to cover the vehicle’s windscreen feature. Secondly, the relative angle calculating model is implemented to find a relative angle. Eventually, the decision of vehicular group is made on the angle by the separating module. Experimentally, using three types of vehicle consisting of a car, van, and truck, the accuracy of separation is extremely high.