Vehicle driver identification system based on biometric data

Abstract

The article discusses the problems associated with monitoring the observance of work and rest regimes by taxi drivers, which, together with an increase in the number of taxi services, lead to an increase in the number of road traffic accidents. The most common cause of these accidents, experts call significant excess of working time, leading to fatigue of drivers. It is quite difficult to control work and rest regimes of drivers in the current conditions. One of the effective ways to solve the problem, according to the authors, is a hardware- software complex that works on the basis of computer vision algorithms, which can identify the driver’s identity by recognizing his face, send information to supervisor and integrate into car control systems. The main goal of this article is to reduce the accident rate of taxis and car-sharing vehicles by introducing a face recognition system in them. The scientific novelty of the work lies in the use of new methods of computer vision in solving the problem of reducing the accident rate of automobile transport. The system works as follows: using a video camera while using a car, the system periodically captures and recognizes the image of the driver’s face. If the result is negative, the supervisor receives a notification of non-compliance. The article provides an overview of the functionality of the prototype of the proposed system created by the authors. The system prototype consists of five conditional modules: a face recognition module; module for storing information in a log file and application interaction with a web browser; a module for sending a signal to the serial port for interaction with the hardware (mechanical) part of the project using a programmable microcontroller; a module associated with the operation of the microcontroller and peripheral actuators; multi-platform application wrapper module. System modules are written in Python and are based on the OpenCV computer vision library. The system has the ability to send statistical information about driver identification to the network. It is also possible to interact with the hardware and mechanical components of the car, for example, with a central lock. System modules are written in Python and are based on the OpenCV computer vision library. Such a product will be in demand by companies providing taxi services, car sharing, as well as owners and users of corporate vehicle parks in various industries, for example, in the oil and gas industry.

Keywords

• accident rate
• automobiles
• time tracking
• driver
• face recognition
• computer vision