The paper provides a new insight into perception problem of autonomous transport. This study is dedicated to solve sensor fusion problem in intelligent transport system by performing tasks of object detection, real-time recognition and also determine motion parameters using solo camera as sensor. Study deals with the simulation model for a computer vision based sensor which can be very economical. A simulation model of novel system structure for recognition, speed estimation process using solo camera sensor and CNN algorithm with training is proposed in this paper. The entire computational complexity is supposed to be significantly lower in comparison with actual experimentation. From this point of view, the simulation of sensor system can be used for optimization and better results of object detection and recognition.The computer model and simulation results along with the workability of the system is presented in the paper.