Vehicles are an integral part of everyday life. Since drivers depend on these transport means to perform their activities, it is important to investigate how they behave inside the vehicles in order to facilitate their task and avoid any undesired consequences (e.g. congestion, accidents). Different components of the driving environment have an influence on the driving behavior and the main objective of this thesis is to identify the points where the driving behavior changes due to these driving factors. From the driving conditions the traffic flow and its composition, the road geometry, the traffic signs and the traffic lights are selected and are analyzed through the use of vehicle-based systems. The actual situation that occurs in the road is determined from cameras mounted on the vehicle. For the identification of the driving behavior a simple car-following model is used, the Helly model. The first part of this study consists of the estimation of the parameters of the car-following model, while the second part consists of the identification of the points where these parameters change due to the exogenous driving factors. The estimation of the parameters is achieved through an optimization procedure which uses a moving estimation window and searches for constant sets of parameters in that time window. The estimation model performs well only under car-following situations. When the car-following task is achieved, the change of lanes, the traffic lights and the changes in the traffic conditions are found to have an influence in the parameters of the Helly model. In contrast, the composition of traffic and the existence of on-ramps and off-ramps in the road do not influence the driving behavior. The proposed system captures the intra-heterogeneity in driver's behavior when performing the same action under the same circumstances. A future real-time application which would identify the driving behavior and the action points would be a valuable tool for the traffic management operations because the traffic flow phenomena could be recognized and predicted directly and more accurately. Another useful application could be found in traffic safety operations where the automated vehicles would be able to provide the right information in the appropriate phase.