Vehicle object observation using position based local gradient model

The conception of traffic activities in urban framework is the major endeavor in the study of traffic troubles. This examination is chiefly processed on the maximization of traffic flow, and the minimization of traffic accidents, congestions and pollution. In common, network representations of transport systems are understood to be fixed, but these representations do not permit an exact recreation of seriously packed urban road networks. For this rationale, traffic engineers happened to judge some substitute models such as simple object recognition and localization of road vehicles based on the position and orientation of vehicle image data. But the drawback of the approach is that if the shape of the vehicle and its pose varies in multiple junction coordination, the model based recognition is an inefficient one. To enhance the process of traffic flow mechanism, in this paper, we are going to implement a surveillance image object recognition and localization using improved local gradient model. At first, the vehicle images are taken by camera which is taken on the junction board of the traffic coordination. Based on location, position, angle and height of the camera to be fixed on the junction board, the vehicle images are obtained clearly in an aerial view and extracted the entire road for identifying the vehicle densities and postures using ray traced templates. An experimental evaluation is carried out to estimate the performance of the proposed vehicle object observation using position based local gradient model in terms of vehicle density, acquired image clarity, pose recovery and compared with an existing model based on simple object recognition and localization.