Control System Image Analysis of Unmanned Vehicles Based on Computer Vision Technology

The article shows the urgency of control system development by unmanned vehicles based on computer vision technologies. These technologies are used in such tasks as a car location determination, the detection of obstacles and a suitable parking place determination. These tasks are resource intensive and must be performed in real time. Therefore, it is important to develop effective models, methods and tools that ensure the achievement of the required time and the accuracy of use in unmanned vehicle control systems. At that, the choice of image representation model is important. In this paper, they considered the model of energy characteristics. On its basis, it is possible to develop the descriptions of images to highlight and analyze their characteristic features, also for the isolating of contours, regions, and special points. The effectiveness of the proposed approach to image analysis is conditioned by the fact that the objects (road signs, road marking or car numbers) under consideration that have to be revealed and identified are characterized by the relevant features. Besides, the use of wavelet transformations allows one to perform the same basic operations to solve a set of tasks in onboard unmanned vehicle systems, also for the tasks of image primary processing, segmentation, description, recognition and compression. The application of this approach will allow to reduce the time for all procedure performance and to reduce the requirements to compute the resources of the vehicle on-board system.