Çift Sıra Parklanma Durumunun Nesne Tespit Algoritması YOLOv8 ile Tespit Edilmesi

Öz

Çift sıra parklanma durumunun trafik sıkışıklığı, trafik akış koşulları, trafik güvenliği gibi trafik göstergeleri üzerinde birçok olumsuz etkisi vardır. Çift sıra parklanma sürücülerin davranışsal ve trafik alışkanlıklarını etkileyen parametreleri içermektedir. Park ihlalinin önüne geçmek için çeşitli denetim faaliyetleri ve cezai yaptırımlar uygulanmaktadır. Bu çalışma kapsamında çift sıra parklanmanın derin öğrenme algoritmalarından olan YOLOv8 modeliyle tespit edilmesi amaçlanmıştır. Bu doğrultuda, İzmit ve Erzurum'da bulunan ve trafik yoğunluğu yüksek caddeler dikkate alınarak toplam 891 görüntüden oluşan bir veri seti oluşturulmuştur. YOLO modeli sonucunda ölçüm parametresi F1 skor değeri 0.83 olarak elde edilmiştir. Modelin çift sıra parklanma, normal parklanma ve tüm veri setine ait mAP@0.5 değerleri sırasıyla 0.851, 0.922 ve 0.886 olarak elde edilmiştir. Diğer performans parametreleri de incelendiğinde modelin çift sıra parklanma durumunu başarılı bir şekilde tespit ettiği sonucuna varılmıştır. Model performans sonuçlarına göre çift sıra ve normal parklanma durumlarının %89'u doğru bir şekilde tespit edilmiştir. Çift sıra parklanma tespitine yönelik yapılacak çalışmalar için bir veri seti altyapısı oluşturulmuştur. Çalışma ile park ihlallerinin otomatik tespit edilmesi ve sürücülerin anlık uyarılması sistemlerinin ilk etap çalışması gerçekleştirilmiştir.

Anahtar Kelimeler

• Çift sıra parklanma
• YOLOv8
• Trafik güvenliği
• Akıllı ulaşım sistemleri
• Trafik sıkışıklığı

Detection of Double Parking Situation with Object Detection Algorithm YOLOv8

Öz

Double parking has many negative effects on traffic indicators such as traffic congestion, traffic flow conditions, and traffic safety. Double parking includes parameters that affect drivers' behavioral and traffic habits. Various inspection activities and penal sanctions are implemented to prevent parking violations. Within the scope of this study, it is aimed to detect double parking with the YOLOv8 model, one of the deep learning algorithms. In this direction, a data set consisting of a total of 891 images was created, taking into account the streets with high traffic density in İzmit and Erzurum. As a result of the YOLO model, the measurement parameter F1 score value was obtained as 0.83. The mAP@0.5 values of the model for double parking, normal parking and the entire data set were obtained as 0.851, 0.922 and 0.886, respectively. When other performance parameters were examined, it was concluded that the model successfully detected the double parking situation. According to the model performance results, 89% of double and normal parking situations were detected correctly. A data set infrastructure has been created for studies on the detection of double parking. With this study, the initial work of the systems for automatic detection of parking violations and instant warning of drivers was carried out.

Anahtar Kelimeler

• Double parking
• YOLOv8
• Traffic safety
• Intelligent transportation systems
• Traffic congestion

Kaynakça

1. Alemdar, K. D. (2023). Sürücü dikkat dağınıklığının çevresel etkilerinin incelenmesi ve nesne tespit algoritmaları ile tespit edilmesi. Doktora tezi. Erzurum Teknik Üniversitesi Fen Bilimleri Enstitüsü, Erzurum.
2. Alho, A. R., de Abreu e Silva, J., de Sousa, J. P. ve Blanco, E. (2018). Improving mobility by optimizing the number, location and usage of loading/unloading bays for urban freight vehicles. Transportation Research Part D: Transport and Environment, 61, 3–18. https://doi.org/10.1016/j.trd.2017.05.014
3. Arnott, R., Inci, E. ve Rowse, J. (2015). Downtown curbside parking capacity. Journal of Urban Economics, 86, 83–97. https://doi.org/10.1016/j.jue.2014.12.005
4. Bayram, A. F. ve Nabiyev, V. (2023). Derin öğrenme tabanlı saklanan kamufle tankların tespiti: son teknoloji YOLO ağlarının karşılaştırmalı analizi. Gümüşhane Üniversitesi Fen Bilimleri Dergisi, 13(4), 1082-1093. https://doi.org/10.17714/gumusfenbil.1271208
5. Buhl, N. (2023). F1 Score in Machine Learning. Erişim adresi: https://encord.com/blog/f1-score-in-machine-learning/#h1 (Erişim tarihi: 10.04.2024)
6. Çavdar, I. H. ve Faryad, V. (2019). New design of a supervised energy disaggregation model based on the deep neural network for a smart grid. Energies, 12(7). https://doi.org/10.3390/en12071217
7. Chen, Y., Xu, H., Zhang, X., Gao, P., Xu, Z. ve Huang, X. (2023). An object detection method for bayberry trees based on an improved YOLO algorithm. International Journal of Digital Earth, 16(1), 781–805. https://doi.org/10.1080/17538947.2023.2173318
8. Chen, Z., Zhu, Q., Zhou, X., Deng, J. ve Song, W. (2024). Experimental Study on YOLO-Based Leather Surface Defect Detection. IEEE Access, 12, 32830–32848. https://doi.org/10.1109/ACCESS.2024.3369705

9. Cherrett, T., Allen, J., McLeod, F., Maynard, S., Hickford, A. ve Browne, M. (2012). Understanding urban freight activity - key issues for freight planning. Journal of Transport Geography, 24, 22–32. https://doi.org/10.1016/j.jtrangeo.2012.05.008
10. Chiara, G. D. ve Goodchild, A. (2020). Do commercial vehicles cruise for parking? Empirical evidence from Seattle. Transport Policy, 97, 26–36. https://doi.org/10.1016/j.tranpol.2020.06.013
11. Choo, H., Kim, M., Choi, J., Shin, J., & Shin, S. Y. (2020). Influenza Screening via Deep Learning Using a Combination of Epidemiological and Patient-Generated Health Data: Development and Validation Study. Journal of medical Internet research, 22(10), e21369. https://doi.org/10.2196/21369
12. Chrysostomou, K., Petrou, A., Aifadopoulou, G. ve Morfoulaki, M. (2019). Microsimulation Modelling of the Impacts of Double-Parking Along an Urban Axis. Nathanail, E.G. ve Karakikes, I. D. (Ed.), Data Analytics: Paving the Way to Sustainable Urban Mobility (s. 164–171). Yer: Springer International Publishing.
13. Dezi, G., Dondi, G. ve Sangiorgi, C. (2010). Urban freight transport in Bologna: Planning commercial vehicle loading/unloading zones. Procedia - Social and Behavioral Sciences, 2(3), 5990–6001. https://doi.org/10.1016/j.sbspro.2010.04.013
14. Estepa, R., Estepa, A., Wideberg, J., Jonasson, M. ve Stensson-Trigell, A. (2017). More Effective Use of Urban Space by Autonomous Double Parking. Journal of Advanced Transportation, 2017, 8426946. https://doi.org/10.1155/2017/8426946
15. Gao, J., Xie, K. ve Ozbay, K. (2018). Exploring the Spatial Dependence and Selection Bias of Double Parking Citations Data. Transportation Research Record, 2672(42), 159–169. https://doi.org/10.1177/0361198118792323
16. Goodfellow, I., Bengio, Y. ve Courville, A. (2016). Deep learning. MIT press. www.deeplearningbook.org
17. Hasnine, M. S. ve Habib, K. N. (2020). Transportation demand management (TDM) and social justice: A case study of differential impacts of TDM strategies on various income groups. Transport Policy, 94, 1–10. https://doi.org/10.1016/j.tranpol.2020.05.002
18. Hendry ve Chen, R. C. (2019). Automatic License Plate Recognition via sliding-window darknet-YOLO deep learning. Image and Vision Computing, 87, 47–56. https://doi.org/10.1016/j.imavis.2019.04.007
19. Ho, G. T. S., Tsang, Y. P., Wu, C. H., Wong, W. H. ve Choy, K. L. (2019). A computer vision-based roadside occupation surveillance system for intelligent transport in smart cities. Sensors (Switzerland), 19(8). https://doi.org/10.3390/s19081796
20. Kadkhodaei, M., Shad, R. ve Ziaee, S. A. (2022). Affecting factors of double parking violations on urban trips. Transport Policy, 120, 80–88. https://doi.org/10.1016/j.tranpol.2022.02.015
21. Khaliq, A., Der Waerden, P. Van, Janssens, D. ve Wets, G. (2019). A Conceptual Framework for Forecasting Car Driver’s On-Street Parking Decisions. Transportation Research Procedia, 37, 131–138. Elsevier B.V. https://doi.org/10.1016/j.trpro.2018.12.175
22. Kim, Y.J., Yoo, E.Y. ve Kim K.G. (2021) Deep learning based pectoral muscle segmentation on Mammographic Image Analysis Society (MIAS) mammograms. Precision and Future Medicine, 5(2), 77-82. https://doi.org/10.23838/pfm.2020.00170
23. Kladeftiras, M. ve Antoniou, C. (2013). Simulation-Based Assessment of Double-Parking Impacts on Traffic and Environmental Conditions. Transportation Research Record, 2390(1), 121–130. https://doi.org/10.3141/2390-13
24. Kobus, M. B. W., Gutiérrez-i-Puigarnau, E., Rietveld, P. ve Van Ommeren, J. N. (2013). The on-street parking premium and car drivers’ choice between street and garage parking. Regional Science and Urban Economics, 43(2), 395–403. https://doi.org/10.1016/j.regsciurbeco.2012.10.001
25. Mannini, L., Cipriani, E., Crisalli, U., Gemma, A. ve Vaccaro, G. (2017). On-Street Parking Search Time Estimation Using FCD Data. Transportation Research Procedia, 27, 929–936. Elsevier B.V. https://doi.org/10.1016/j.trpro.2017.12.149
26. Mu, L., Xian, L., Li, L., Liu, G., Chen, M. ve Zhang, W. (2023). YOLO-Crater Model for Small Crater Detection. Remote Sensing, 15(20). https://doi.org/10.3390/rs15205040
27. Nicancı Sinanoğlu, M. ve Kaya, Ş. (2024). Local Climate Zone Classification Using YOLOV8 Modeling in Instance Segmentation Method. International Journal of Environment and Geoinformatics, 11(2), 1-9. https://doi.org/10.30897/ijegeo.1456352
28. Nourinejad, M., Gandomi, A. ve Roorda, M. J. (2020). Illegal parking and optimal enforcement policies with search friction. Transportation Research Part E: Logistics and Transportation Review, 141. https://doi.org/10.1016/j.tre.2020.102026
29. Ouyang, L. ve Wang, H. (2019). Vehicle target detection in complex scenes based on YOLOv3 algorithm. IOP Conference Series: Materials Science and Engineering, 569, 052018. https://doi.org/10.1088/1757-899X/569/5/052018
30. Padalko, H., Chomko, V. ve Chumachenko, D. (2024). A novel approach to fake news classification using LSTM-based deep learning models. Frontiers in big data, 6, 1320800. https://doi.org/10.3389/fdata.2023.1320800
31. Redmon, J., Divvala, S., Girshick, R. ve Farhadi, A. (2016). You Only Look Once: Unified, Real-Time Object Detection. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 779–788.
32. Redmon, J. ve Farhadi, A. (2017). YOLO9000: Better, faster, stronger. 30th IEEE Conference on Computer Vision and Pattern Recognition, 6517–6525.
33. Sevi, M. ve Aydın, İ. (2023). Detection of Foreign Objects Around the Railway Line with YOLOv8. Computer Science, IDAP-2023 : International Artificial Intelligence and Data Processing Symposium(IDAP-2023), 19-23. https://doi.org/10.53070/bbd.1346317
34. Simićević, J., Milosavljević, N., Maletić, G. ve Kaplanović, S. (2012). Defining parking price based on users’ attitudes. Transport Policy, 23, 70–78. https://doi.org/10.1016/j.tranpol.2012.06.009
35. Spiliopoulou, C. ve Antoniou, C. (2012). Analysis of Illegal Parking Behavior in Greece. Procedia - Social and Behavioral Sciences, 48, 1622–1631. https://doi.org/10.1016/j.sbspro.2012.06.1137
36. Tzouras, P. G. ve Lázaro, C. P. (2020). Illegal parking in urban streets: connection with the geometric characteristics and its mitigation through traffic calming measures. Aeihoros, 30.
37. Ultralytics. (2023). YOLOv8. Erişim adresi: https://github.com/ultralytics/ultralytics
38. Uysal, M. ve Alver, Y. (2022). Factors Affecting Parking Choice Behaviors: The Case of Izmir. Teknik Dergi/Technical Journal of Turkish Chamber of Civil Engineers, 33(3), 11887–11901. https://doi.org/10.18400/tekderg.766468
39. Yang, M. D., Tseng, H. H., Hsu, Y. C., Yang, C. Y., Lai, M. H., & Wu, D. H. (2021). A UAV open dataset of rice paddies for deep learning practice. Remote Sensing, 13(7), 1358.
40. Xiong, J., Wu, J., Tang, M., Xiong, P., Huang, Y. ve Guo, H. (2024). Combining YOLO and background subtraction for small dynamic target detection. Visual Computer. https://doi.org/10.1007/s00371-024-03342-1