Unmanned vehicles are a promising technology that improves the efficiency of logistics traffic. The first step here is the perception of the environment through the systems installed in it. Computer vision (CV) allows the latter to acquire meaningful information from videos. Thanks to the development of deep learning and neural networks, it is possible to enter the information from a camera as an image string. In this research, the source of information is given by a single lens camera, which is a cheap approach for moving objects. This work discusses a method based on the results of applying a heat map approach, providing real-time operation.