Intelligent Smart Car Parking System Using Deep Learning and Internet of Things

Abstract

Parking problems are more acute in contemporary cities due to the rising urban population and the exponential growth of cars in the city. Traditional parking operations can be sub-optimal, may not offer real-time insight, and can lead to higher congestion, fuel usage, environmental impacts and driver frustration. While some of the existing parking solutions are partially automated using IoT, many of them are based on a hardware-based sensor deployment or on traditional Optical Character Recognition (OCR) number plate recognition techniques that struggle with varying environmental conditions and accuracy. This paper proposes an intelligent smart car parking system, which utilizes Internet of Things (IoT), Deep Learning, Computer Vision, and predictive analysis to improve the parking management process. The proposed framework further developed a previously designed IoT based prototype to include YOLO based Automatic Number Plate Recognition (ANPR), camera-based parking occupancy detection and machine learning based parking demand forecasting. The system automates the identification of vehicles, parking space allocation, monitoring of vehicle parking and exit management with less human involvement. The embedding of predictive features allows for proactive decision making and optimization of the use of parking resources. The proposed framework\'s effectiveness in smart city contexts and ITS applications is confirmed by the improvements in various metrics, including scalability, reliability, recognition performance, and operational efficiency, shown through comparative evaluation.

Introduction

Rapid urbanization has significantly increased the number of vehicles in cities, creating major parking challenges such as traffic congestion, fuel wastage, pollution, and driver frustration caused by the search for available parking spaces. Traditional parking systems rely on manual processes or limited automation, offering insufficient real-time guidance. Although IoT-based smart parking systems improve monitoring and management, many depend on dedicated sensors for each parking space, resulting in high installation, maintenance, and scalability costs. Similarly, conventional OCR-based Automatic Number Plate Recognition (ANPR) systems perform poorly under challenging conditions such as poor lighting, motion blur, shadows, and adverse weather.

To address these limitations, the proposed Phase-2 Intelligent Smart Car Parking System (ISCPS) integrates IoT, YOLO-based Automatic Number Plate Recognition (ANPR), computer vision-based parking occupancy detection, and predictive analytics into a unified framework. Compared with the earlier Phase-1 system, which combined IoT with OCR for basic entry and exit management, the new approach replaces OCR with YOLO for more robust vehicle recognition and eliminates hardware-intensive parking sensors by using camera-based occupancy detection. Historical parking data are analyzed to forecast future parking demand, enabling proactive parking management and improved resource allocation.

The proposed architecture consists of four layers: sensing, edge processing, cloud services, and the user interface. Cameras capture vehicle images for license plate recognition, verify vehicle information with the parking database, automatically assign parking slots, monitor occupancy in real time, and update records when vehicles exit. The implementation was developed in Python using OpenCV for image processing and SQLite for database management, following a sense–decide–act operational model.

Experimental evaluation demonstrates significant improvements over the previous Phase-1 system. YOLO-based ANPR provides greater robustness under varying environmental conditions, camera-based occupancy monitoring reduces hardware complexity while improving scalability, and predictive analytics transforms parking management from reactive to proactive by forecasting parking demand. Overall, the proposed system enhances operational efficiency, reduces maintenance costs, minimizes human intervention, and provides a scalable, intelligent parking management solution suitable for modern Intelligent Transportation Systems (ITS) and smart city applications.

Conclusion

The need for intelligent parking management technology to overcome the constraints of traditional parking systems has grown in speed and scale with the demand for efficient urban mobility solutions. In the case of the paper, an intelligent smart car parking framework that leveraged Internet of Things (IoT), Deep learning, Computer vision, and predictive analytics was presented in order to enhance the parking operations in modern urban scenario. The proposed Phase-2 architecture is a great improvement to the IoT–OCR based prototype developed in Phase-1. The implementation of Automatic Number Plate Recognition (ANPR) using YOLO algorithm improved the robustness and reliability of the system for vehicle identification in different environments, overcoming some drawbacks of the traditional OCR system. The implementation of computer vision based occupancy monitoring led to the elimination of the requirement for large platforms of sensors and instead it proposed to monitor multiple parking spaces using the camera infrastructure. Such change improved the scalability, system complexity and maintainability. Predictive analytics was also an important feature that was introduced in the proposed model. The system went from a reactive approach to an operational model to a proactive approach to a decision support model, using historical occupancy and forecasting. This foresight enables better use of the parking resources, optimal allocation of the parking spaces, and congestion minimization during peak periods. From a communication perspective, low-power wide-area communication technologies did deliver energy efficiency benefits, and added to the potential viability of deploying in large quantities with long service lives. Furthermore, the proposed system was hierarchical in its system design and allowed for flexibility, modularity, and adaptability by integrating the different functions of sensing, processing, communication, and application into one system. The comparative study helped the researchers come to the conclusion that the proposed Phase-2 system\'s recognition ability, operational efficiency, scalability, reliability, and intelligent decision was better than Phase-1 system. The results suggest that the proposed framework has a potential scope of being implemented in the smart city transportation infrastructure and is practical and effective in intelligent parking management. The proposed framework shows good performance, but there is scope for further development that will increase its capability and applicability. Autonomous vehicle systems can be incorporated with future intelligent parking systems to enable fully automated parking with reduced user interaction. By implementing blockchain mechanisms, these systems can further enhance the transparency, security, and trust in transactions, including the secure reservation system and automated payment processing. In addition, dynamic pricing models can be used to optimize parking use and behaviour based on the real-time demands and booking situations. Furthermore, having advanced urban data analytics and large-scale deployment can offer useful information on long-term transportation planning and sustainable mobility initiatives. As new technology becomes a part of ITS, its role is crucial in solving the increasing mobility issues of modern cities.

References

[1] “Adak, R., Kumbhar, A., Pathare, R., and Gowda, S., “Automatic Number Plate Recognition with YOLOv3-CNN,” 2022. https://arxiv.org/abs/2211.05229” [2] “Adelantado, F., Vilajosana, X., Tuset-Peiro, P., Martinez, B., Melia-Segui, J., and Watteyne, T., “Understanding the Limits of LoRaWAN,” IEEE Communications Magazine, vol. 55, no. 9, pp. 34–40, 2017. https://www.researchgate.net/publication/312627168_Understanding_the_limits_of_LoRaWAN” [3] “Agarwal, V., and Bansal, G., “Automatic Number Plate Detection and Recognition Using YOLO World,” Computers & Electrical Engineering, 2024. https://www.researchgate.net/publication/384119658_Automatic_number_plate_detection_and_recognition_using_YOLO_world” [4] “Al Mamun, A., Hasib, A., Mussa, A. S. M., et al., “IoT-Enabled Smart Car Parking System Through Integrated Sensors and Mobile Applications,” arXiv, 2024. https://arxiv.org/abs/2412.10774” [5] “Al-Turjman, F., and Malekloo, A., “Smart Parking Systems: Reviewing the Literature, Architecture and Ways Forward,” Smart Cities, vol. 4, no. 2, pp. 623–642, 2021. DOI: https://doi.org/10.3390/smartcities4020032” [6] “Amato, G., Carrara, F., Falchi, F., et al., “Deep Learning for Decentralized Parking Lot Occupancy Detection.” https://www.researchgate.net/publication/309540140_Deep_Learning_for_Decentralized_Parking_Lot_Occupancy_Detection” [7] “Atzori, L., Iera, A., and Morabito, G., “The Internet of Things: A Survey,” Computer Networks, vol. 54, no. 15, pp. 2787–2805, 2010. https://www.researchgate.net/publication/222571757_The_Internet_of_Things_A_Survey” [8] Barriga, J. J., Sulca, J., León, J. L., et al., “Smart Parking: A Literature Review from the Technological Perspective,” Applied Sciences, vol. 9, no. 21, p. 4569, 2019. DOI: https://doi.org/10.3390/app9214569 [9] “Bochkovskiy, A., Wang, C. Y., and Liao, H. Y. M., “YOLOv4: Optimal Speed and Accuracy of Object Detection,” 2020. https://arxiv.org/abs/2004.10934” [10] “Box, G. E. P., and Jenkins, G. M., Time Series Analysis: Forecasting and Control. https://repo.darmajaya.ac.id/4781/1/Time%20Series%20Analysis_%20Forecasting%20and%20Control%20%28%20PDFDrive%20%29.pdf” [11] “Bradski, G., “The OpenCV Library,” Dr. Dobb\'s Journal of Software Tools, 2000. https://www.researchgate.net/publication/233950935_The_Opencv_Library” [12] “Brownlee, J., Deep Learning for Time Series Forecasting. https://www.inf.u-szeged.hu/~korosig/teach/books/Jason%20Brownlee%20-%20Deep%20Learning%20for%20Time%20Series%20Forecasting%20-%20Predict%20the%20Future%20with%20MLPs,%20CNNs%20and%20LSTMs%20in%20Python%20(2018).pdf” [13] “Centenaro, M., Vangelista, L., Zanella, A., and Zorzi, M., “Long-Range Communications in Unlicensed Bands: The Rising Stars in the IoT and Smart City Scenarios,” IEEE Wireless Communications, vol. 23, no. 5, pp. 60–67, 2016. https://www.researchgate.net/publication/282603371_Long-Range_Communications_in_Unlicensed_Bands_the_Rising_Stars_in_the_IoT_and_Smart_City_Scenarios” [14] “Geng, Y., and Cassandras, C., “A New Smart Parking System Infrastructure and Implementation.” https://www.sciencedirect.com/science/article/pii/S1877042812043042” [15] “Goodfellow, I., Bengio, Y., and Courville, A., Deep Learning. Cambridge, MA, USA: MIT Press, 2016. https://www.researchgate.net/publication/320703571_Ian_Goodfellow_Yoshua_Bengio_and_Aaron_Courville_Deep_learning_The_MIT_Press_2016_800_pp_ISBN_0262035618” [16] “Gubbi, J., Buyya, R., Marusic, S., and Palaniswami, M., “Internet of Things (IoT): A Vision, Architectural Elements, and Future Directions,” Future Generation Computer Systems, vol. 29, no. 7, pp. 1645–1660, 2013. https://www.researchgate.net/publication/228095891_Internet_of_Things_IoT_A_Vision_Architectural_Elements_and_FutureDirections” [17] “Hsieh, C. F., Lin, C. Z., Li, Z. Z., and Cho, C. H., “Automatic Vehicle License Plate Recognition Based on YOLOv4 for Smart Parking Management System,” IEEE GCCE, 2022. DOI: 10.1109/GCCE56475.2022.10014165 https://www.researchgate.net/publication/367262526_Automatic_Vehicle_License_Plate_Recognition_Based_on_YOLO_v4_for_Smart_Parking_Management_System” [18] “Hyndman, R., and Athanasopoulos, G., Forecasting: Principles and Practice. https://otexts.com/fpp3/” [19] “Jocher, G., et al., “YOLOv5 Documentation and Implementation.” https://github.com/ultralytics/yolov5” [20] “Kotb, A. O., Shen, Y. C., Zhu, X., and Huang, Y., “iParker: A New Smart Car-Parking System Based on Dynamic Resource Allocation and Pricing.” https://livrepository.liverpool.ac.uk/3003393/1/iParker%20-%20A%20New%20Smart%20Car-Parking%20System%20Based%20on%20Dynamic%20Resource%20Allocation%20and%20Pricing.pdf” [21] “Laroca, R., Severo, E., Zanlorensi, L., et al., “A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector,” 2018. https://arxiv.org/abs/1802.09567” [22] “Laroca, R., Zanlorensi, L., Gonçalves, G., et al., “An Efficient and Layout-Independent Automatic License Plate Recognition System Based on the YOLO Detector,” 2019. https://arxiv.org/abs/1909.01754” [23] “LeCun, Y., Bengio, Y., and Hinton, G., “Deep Learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015. https://www.researchgate.net/publication/277411157_Deep_Learning” [24] “Lin, T., Rivano, H., and Le Mouël, F., “A Survey of Smart Parking Solutions,” IEEE Transactions on Intelligent Transportation Systems. https://ieeexplore.ieee.org/document/7895130” [25] “Mahmood, Z., Haneef, O., Muhammad, N., et al., “Towards a Fully Automated Car Parking System.” https://ietresearch.onlinelibrary.wiley.com/doi/full/10.1049/iet-its.2018.5021” [26] “Malkar, N., Taklikar, P., Borkar, M., et al., “Survey Paper on an IoT-Based Smart Parking System.” https://www.academia.edu/44800096/Survey_Paper_on_An_IoT_Based_Smart_Parking_System” [27] “Mastronarde, N., et al., “Vision-Based Parking Occupancy Detection: A Survey.” https://www.researchgate.net/publication/361043919_Review_of_Research_on_Vision-Based_Parking_Space_Detection_Method” [28] “Mathur, S., Kaul, S., et al., “ParkNet: Drive-by Sensing of Road-Side Parking Statistics.” https://www.researchgate.net/publication/221234396_ParkNet_Drive-by_Sensing_of_Road-Side_Parking_Statistics” [29] “Ogás, M. G. D., Fabregat, R., and Aciar, S., “Survey of Smart Parking Systems,” Applied Sciences, vol. 10, no. 11, p. 3872, 2020. DOI: https://doi.org/10.3390/app10113872” [30] “Paidi, V., Fleyeh, H., and Håkansson, J., “Vision-Based Smart Parking Applications.” https://ietresearch.onlinelibrary.wiley.com/doi/full/10.1049/iet-its.2017.0406” [31] “Paidi, V., Fleyeh, H., Håkansson, J., and Nyberg, R., “Smart Parking Sensors, Technologies and Applications for Open Parking Lots: A Review.” https://ietresearch.onlinelibrary.wiley.com/doi/10.1049/iet-its.2017.0406” [32] “Paudel, S., Vechione, M., and Gurbuz, O., “Predicting University Campus Parking Demand Using Machine Learning Models,” Transportation Research Record: Journal of the Transportation Research Board, vol. 2678, no. 6, pp. 14–26, 2024. DOI: 10.1177/03611981231193417 https://journals.sagepub.com/doi/10.1177/03611981231193417?utm_source=chatgpt.com” [33] “Redmon, J., and Farhadi, A., “YOLOv3: An Incremental Improvement,” 2018. https://arxiv.org/abs/1804.02767” [34] “Redmon, J., Divvala, S., Girshick, R., and Farhadi, A., “You Only Look Once: Unified Real-Time Object Detection,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. https://arxiv.org/abs/1506.02640” [35] “Russell, S. J., and Norvig, P., Artificial Intelligence: A Modern Approach, 4th ed. Upper Saddle River, NJ, USA: Pearson Education, 2020. https://people.engr.tamu.edu/guni/csce625/slides/AI.pdf” [36] “Szeliski, R., Computer Vision: Algorithms and Applications, 2nd ed. Cham, Switzerland: Springer, 2022. https://www.researchgate.net/publication/357543576_Computer_Vision_Algorithms_and_Applications” [37] “Ultralytics, “YOLOv8 Official Documentation.” https://docs.ultralytics.com” [38] “Valipour, S., Siam, M., Jagersand, M., and Ray, N., “Deep Learning-Based Vehicle Parking Occupancy Detection.” https://www.researchgate.net/publication/393132053_Deep_Learning-Based_Vehicle_Parking_Occupancy_Detection” [39] “Zanella, A., Bui, N., Castellani, A., Vangelista, L., and Zorzi, M., “Internet of Things for Smart Cities,” IEEE Internet of Things Journal, vol. 1, no. 1, pp. 22–32, 2014. https://ieeexplore.ieee.org/document/6740844” [40] “Zeng, C., Ma, C., Wang, K., and Cui, Z., “Parking Occupancy Prediction Method Based on Multi-Factors and Stacked GRU-LSTM,” IEEE Access, 2022. DOI: 10.1109/ACCESS.2022.3169823 https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9765513”

Copyright

Copyright © 2026 Showkat Ahmad Mehraj, Upasna Setia. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.