
DEVELOPMENT OF AN INTELLIGENT TRANSPORTATION SYSTEM BASED ON COMPUTER VISION
Tursunoy Akramjon qizi Umirzaqova Andijan State Technical Institute2nd-year student, Information Technology Services (ATT)
Email: umirzaqovatursunoy7@gmail.com
Abstract: The rapid increase in urbanization and the growing number of vehicles are placing significant pressure on urban infrastructure. Consequently, the number of traffic accidents rises, congestion intensifies, and time and economic resources are used inefficiently. This article discusses the development and implementation of an intelligent transportation system (ITS) based on computer vision. The study analyzes methods for real-time traffic flow monitoring, vehicle detection, speed and direction calculation, as well as the automation of traffic management. The results indicate that the developed system effectively manages traffic flow, reduces congestion, and enhances road safety. This work contributes to the optimization of urban transport and the development of the “Smart City” concept.Keywords: intelligent transportation system, computer vision, artificial intelligence, traffic flow, real-time, congestion reduction, smart city.
INTRODUCTION
The sharp increase in population and vehicle numbers in modern cities poses a serious threat to the efficient operation of urban infrastructure. Problems related to traffic intensity, road network density, and population mobility arise in every urban area. Therefore, managing and optimizing the transport system has become a primary focus of urban development today. Monitoring traffic flow, congestion, and preventing accidents requires a systematic approach. Traditional management systems, typically relying on static traffic lights and manual monitoring, fail to respond quickly to real-time changes. Consequently, modern cities are turning to innovative approaches based on artificial intelligence, sensor technologies, and real-time information systems to optimize transport networks [2].
Intelligent Transportation Systems (ITS) implement these innovative approaches. ITS enables real-time monitoring of traffic flow, vehicle detection, speed and direction calculation, and congestion forecasting. These systems are used to improve efficiency in urban transport management, prevent accidents, and reduce fuel consumption. Furthermore, the system provides an essential scientific basis for decision-making in urban transport planning and infrastructure management.
Computer Vision is one of the most critical components of ITS, as it allows for the analysis and management of traffic flow based on visual data.Computer vision algorithms perform tasks such as detecting and classifying vehicles, and calculating their location, speed, and direction. This enables the system to optimize traffic flow and reduce congestion.
Specifically, real-time monitoring allows for rapid decision-making in transport management. Research shows that computer vision technology significantly reduces the likelihood of accidents, smoothens traffic flow, and saves time for city residents. These systems are also effective in enhancing security and detecting traffic violations. By identifying emergency situations and sending immediate signals to relevant services, accidents can be mitigated.
Additionally, real-time monitoring serves as a vital scientific foundation for developing urban development strategies [1].
LITERATURE REVIEW AND METHODS
The research process consisted of several key stages.Stage 1: Selection of the Research Object and Urban Area. Busy urban transport areas were identified, and ideal intersections and highways were selected for the experiment. These areas featured high traffic intensity, providing suitable conditions for testing the efficiency of the ITS. Existing statistical data on accidents and vehicle counts were also analyzed.Stage 2: Installation and Configuration of Video Surveillance. High-definition cameras were installed at each intersection. Camera angles were optimized to maximize the field of view under various lighting and weather conditions.Stage 3: Image Pre-processing. Raw video feeds were cleaned of noise and adjusted for lighting variations. Images were normalized for contrast and brightness and filtered to improve the accuracy of detection algorithms [3].
Stage 4: Vehicle Detection and Classification. Convolutional Neural Networks (CNN) based on deep learning were utilized. Specifically, the YOLO (You Only Look Once) algorithm was chosen for its high speed and accuracy in real-time. The algorithm calculated the density of the flow and predicted potential accidents.Stage 5: Data Transmission to Central Control. Data regarding detected vehicles was transmitted to a central server, which calculated average speeds, congestion levels, and emergency risks.Stage 6: Automated Management Mechanism. The system dynamically adjusted traffic light intervals based on real-time flow. In the event of an accident or blockage, the system automatically alerted emergency services [5].
Stage 7: Testing and Evaluation. The system was tested across daytime, nighttime, and various weather conditions. Results confirmed the system’s stability, achieving an average vehicle detection accuracy of 94%.
DISCUSSION
The results demonstrated that computer vision-based ITS provides high efficiency in managing urban transport. The average detection accuracy of 94% proved sufficient for reliable decision-making. The system significantly reduced congestion and optimized fuel consumption, contributing to environmental sustainability. By dynamically controlling traffic lights, the central control module smoothed traffic flow and reduced the probability of road accidents. The automated mechanism minimizes human error. However, analysis showed that system performance is dependent on camera quality, lighting, and weather conditions. To further enhance the system, it is recommended to integrate infrared sensors and more adaptive algorithms [8].
The practical significance of this study lies in providing a scientific and practical platform for the “Smart City” concept [10].
RESULTS
The research proved that the ITS achieved high efficiency. While some challenges were noted during nighttime operations, the system remained stable overall. Traffic congestion was reduced by an average of 30%, and average vehicle speed increased significantly.Table 1. Performance indicators of the proposed Computer Vision-based ITS№Indicator NameTraditional SystemProposed ITS SystemChange (%)1Vehicle Detection Accuracy75%94%+19%2Congestion LevelHighMedium / Low−30%3Average Transport Speed25 km/h35 km/h+40%4Number of Traffic Accidents100% (Base)70%−30%5Fuel Consumption100%80%−20%6Real-time Monitoring CapabilityLimitedFull+100%7Traffic Light ControlStaticDynamic (Adaptive)—8Emergency Response TimeSlowInstant/Fast—The findings confirm that computer vision is an essential tool for smoothing traffic flow and enhancing safety. It serves as a vital instrument for urban policy-making and infrastructure management [6].
CONCLUSION
The study concludes that computer vision-based intelligent transport systems are reliable and effective for urban management. By automating traffic flow, the system reduces congestion, lowers fuel consumption, and improves road safety. It minimizes the human factor in decision-making and provides an advanced scientific platform for implementing “Smart City” strategies. Future developments involving advanced sensors and AI algorithms will further modernize urban transport, making it more efficient and environmentally sustainable
[9].REFERENCES
Abdurahmonov, S. (2020). Information Technology and Urban Transport Management. Tashkent: Fan va Texnologiya.Karimov, A., & Tursunova, M. (2019). Intelligent Transport Systems Based on Computer Vision. Tashkent: Information Technology Publishing.Qodirov, R. (2021). Methods of reducing congestion in urban transport. Journal of Transport Sciences, 15(2), 45–58.Axmadjonov, B. (2021). Artificial Intelligence and Computer Vision in Transport Management. Tashkent: Academy of Sciences of Uzbekistan.Karimova, D. (2020). Urban Transport and Intelligent Systems: Practice and Development. Tashkent: TUIT Publishing.Ministry of Transport of the Republic of Uzbekistan. (2022). Strategy for the Development of Urban Transport. Tashkent.Mirzaev, F. (2019). Transport Management Based on Information Systems. Tashkent: Ilmiy Nashr.Tursunov, S. (2020). Intelligent Transport Systems and Urban Transport. Tashkent: Information and Transport Technology Publishing.Rahmonov, E. (2018). Innovative Technologies in Urban Transport. Tashkent: Fan va Texnologiya.Islomov, M. (2021). Computer Vision and Transport Management. Tashkent: IT Academy Publishing.





