Kentsel Trafik Analizinde 3D/2D Hibrit CNN Mimarisi ve Dikkat Mekanizmalarının İncelenmesi

Öz

Kentsel trafik yönetimi, artan nüfusa ve motorlu taşıt sayısına bağlı olarak her geçen gün daha da karmaşık hale gelmektedir. Trafik yoğunluğu veya yol durumlarının, zamanında ve doğru bir şekilde öngörülememesi durumunda; anlık değişen trafik olayları (kazalar, tıkanıklıklar vb.) tespit edilememekte ve müdahale süresinde gecikmeler yaşanmasına sebep olmaktadır. Bu durum ulaşımın güvenliği, çevre zararları ve ekonomik kayıplar açısından ciddi problemlerin doğmasına neden olmaktadır. Bu bağlamda klasik ve geleneksel yöntemler kullanılarak yapılan trafik analizleri yetersiz kalmaktadır. Ayrıca bu yöntemler kullanılarak büyük hacimli ve çok boyutlu veri setlerinin işlenmesi de mümkün olmamaktadır. Bu nedenle hem zamansal hem de mekânsal boyutlarda çıkarım yapabilen güvenilir ve güçlü yaklaşımlara ihtiyaç duyulmaktadır. Bu çalışma kapsamında, bluetooth sensörleri tarafından toplanan veriler kullanılarak 2D CNN, 3D CNN ve 3D/2D hibrit CNN mimarileri ile kentsel trafik yoğunluğu tahmini ve sınıflandırması yapıldı. 3D/2D hibrit CNN mimarisinin en yüksek açıklayıcılığa (R2=0.9286) sahip olduğu tespit edildi. Ayrıca 3D/2D hibrit CNN mimarilerinde dikkat mekanizmalarının kullanımı literatür temelli incelendi.

Anahtar Kelimeler

• 2D CNN
• 3D CNN
• 3D/2D hibrit CNN
• dikkat mekanizmaları

Investigation of 3D/2D Hybrid CNN Architecture and Attention Mechanisms in Urban Traffic Analysis

Öz

Urban traffic management is becoming increasingly complex due to the growing population and the rising number of motor vehicles. When traffic density or road conditions cannot be predicted accurately and in a timely manner, dynamic traffic events (e.g., accidents, congestion, etc.) cannot be detected, which leads to delays in response times. This situation causes serious problems in terms of transportation safety, environmental harm, and economic losses. In this context, traffic analyses based on conventional and classical methods have become inadequate. Moreover, these methods are not capable of processing large-scale and high-dimensional datasets. Therefore, there is a growing need for reliable and robust approaches that can make inferences in both temporal and spatial dimensions. In this study, urban traffic density prediction and classification were performed using data collected from Bluetooth sensors and employing 2D CNN, 3D CNN, and 3D/2D hybrid CNN architectures. Among these models, the 3D/2D hybrid CNN architecture achieved the highest explanatory power (R2 = 0.9286). In addition, the use of attention mechanisms in 3D/2D hybrid CNN architectures was examined based on a literature review.

Anahtar Kelimeler

• 2D CNN
• 3D CNN
• 3D/2D hybrid CNN
• attention mechanisms

Kaynakça

1. [1] McCulloch W. S. and Pitts W., “A logical calculus of the ideas immanent in nervous activity”, Bulletin of Mathematical Biophysics, 5(4): 115-133, (1943).
2. [2] Rosenblatt F., “The perceptron: A probabilistic model for information storage and organization in the brain”, Psychological Review, 65(6): 386–408, (1958).
3. [3] Rumelhart D. E., Hinton G. E. and Williams, R. J., “Learning representations by back-propagating errors”, Nature, 323: 533–536, (1986).
4. [4] Bengio Y., “Learning deep architectures for AI”, Foundations and Trends® in Machine Learning, 2(1): 1-127, (2009).
5. [5] Hinton G. E., Osindero S. and Teh Y. W., “A fast learning algorithm for deep belief nets”, Neural Computation, 18(7): 1527-1554, (2006).
6. [6] Krizhevsky A., Sutskever I. and Hinton G. E., “ImageNet classification with deep convolutional neural networks”, Advances in Neural Information Processing Systems (NeurIPS), 25: 1097-1105, (2012).
7. [7] LeCun Y., Bengio Y. and Hinton G., “Deep learning”, Nature, 521: 436-444, (2015).
8. [8] LeCun Y., Bottou L., Bengio Y. and Haffner, P., “Gradient-based learning applied to document recognition”, Proceedings of the IEEE, 86(11): 2278-2324, (1998).

9. [9] Goodfellow I., Bengio Y. and Courville A., “Deep learning”, MIT Press, 9780262035613, (2016).
10. [10] Hubel D. H. and Wiesel T. N., “Receptive fields and functional architecture of monkey striate cortex”, The Journal of Physiology, 195(1): 215-243, (1968).
11. [11] Fukushima K., “Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position”, Biological Cybernetics, 36(4): 193-202, (1980).
12. [12] Waibel A., Hanazawa T., Hinton G., Shikano K. and Lang K. J., “Phoneme recognition using time-delay neural networks”, IEEE Transactions on Acoustics, Speech, and Signal Processing, 37(3): 328-339, (1989).
13. [13] Simonyan K. and Zisserman A., “Very deep convolutional networks for large-scale image recognition”, International Conference on Learning Representations (ICLR), USA, (2015).
14. [14] Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Erhan D., Vanhoucke V. and Rabinovich A., “Going deeper with convolutions”, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1-9, (2015).
15. [15] He K., Zhang X., Ren S. and Sun J., “Deep residual learning for image recognition”, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770-778, (2016).
16. [16] Dosovitskiy A., Beyer L., Kolesnikov A., Weissenborn D., Zhai X., Unterthiner T., Dehghani M., Minderer M., Heigold G., Gelly S., Uszkoreit J. and Houlsby N., “An image is worth 16x16 words: Transformers for image recognition at scale”, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Online, (2021).
17. [17] 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), ABD, 779-788, (2016).
18. [18] Long J., Shelhamer E. and Darrell, T., “Fully convolutional networks for semantic segmentation”, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), ABD, 3431-3440, (2015).
19. [19] Ari B., Sobahi N., Alçin Ö.F., Sengur A. and Acharya U.R., “Accurate detection of autism using Douglas-Peucker algorithm, sparse coding based feature mapping and convolutional neural network techniques with EEG signals”, Computers in Biology and Medicine, 143, 105311, (2022).
20. [20] Tran D., Bourdev L., Fergus R., Torresani L. and Paluri M., “Learning spatiotemporal features with 3D convolutional networks”, In Proceedings of the IEEE International Conference on Computer Vision (ICCV), Chile, 4489-4497, (2015).
21. [21] Ji S., Xu W., Yang M. and Yu K., “3D convolutional neural networks for human action recognition”, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 35(1): 221–231, (2013).
22. [22] Weng J., Ahuja N. and Huang T. S., “Learning recognition and segmentation using the Cresceptron”, International Journal of Computer Vision (IJCV), 25(2): 105–139.
23. [23] Qiu Z., Yao T. and Mei T., “Learning spatio-temporal representation with pseudo-3D residual networks”, In Proceedings of the IEEE International Conference on Computer Vision (ICCV), Italy, 5533-5541, (2017).
24. [24] Carreira J. and Zisserman A., “Quo vadis, action recognition? A new model and the kinetics dataset”, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), ABD, 6299-6308, (2017).
25. [25] Singh S. P., Wang L., Gupta S., Goli H., Padmanabhan P. and Gulyás B., “3D deep learning on medical images: A review”, Sensors, 20(18): 5097, (2020).
26. [26] Kamnitsas K., Ledig C., Newcombe V. F. J., Simpson J. P., Kane A. D., Menon D. K., Rueckert D. and Glocker B., “Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation”, Medical Image Analysis (MEDIA), 36: 61-78, (2017).
27. [27] Ari A., “Multipath feature fusion for hyperspectral image classification based on hybrid 3D/2D CNN and squeeze-excitation network”, Earth Science Informatics (ESIN), 16(1): 175-191, (2023).
28. [28] Zhang J., Zheng Y. and Qi D., “Deep spatio-temporal residual networks for citywide crowd flows prediction”, arXiv preprint, arXiv:1610.00081, (2016).
29. [29] Zhao Z., Chen W., Wu X., Chen P. C. and Liu J., “LSTM network: A deep learning approach for short-term traffic forecast”, IET Intelligent Transport Systems, 11(2): 68-75, (2017).
30. [30] Li Y., Yu R., Shahabi C. and Liu Y., “Diffusion convolutional recurrent neural network: Data-driven traffic forecasting”, International Conference on Learning Representations (ICLR), Kanada, (2018).
31. [31] Jiang J. and Feofilova A., “Research on urban road traffic flow prediction based on hybrid CNN-LSTM model”, Applied and Computational Engineering (ACE), 129: 77-83, (2025).
32. [32] Bahdanau D., Cho K. and Bengio Y., “Neural machine translation by jointly learning to align and translate”, arXiv preprint arXiv:1409.0473, (2014).
33. [33] Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., Gomez A. N., Kaiser Ł. and Polosukhin I., “Attention is all you need”, Advances in Neural Information Processing Systems (NeurIPS), ABD, 5998-6008, (2017).
34. [34] Luong M. T., Pham H. and Manning C. D., “Effective approaches to attention-based neural machine translation”, arXiv preprint arXiv:1508.04025, (2015).
35. [35] Chorowski J., Bahdanau D., Serdyuk D., Cho K. and Bengio Y., “Attention-based models for speech recognition”, Advances in Neural Information Processing Systems (NeurIPS), Kanada, 577-585, (2015).
36. [36] Woo S., Park J., Lee J. Y. and Kweon I. S., “CBAM: Convolutional block attention module”, In Proceedings of the European Conference on Computer Vision, Almanya, 3-19, (2018).
37. [37] Wang X., Girshick R., Gupta A. and He K., “Non-local neural networks”, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), ABD, 7794-7803, (2018).
38. [38] Xu K., Ba J. L., Kiros R., Cho K., Courville A., Salakhutdinov R., Zemel R. S. and Bengio Y., “Show, attend and tell: Neural image caption generation with visual attention”, In Proceedings of the 32nd International Conference on Machine Learning (ICML), Fransa, 2048-2057, (2015).
39. [39] Zhang H., Wu C., Zhang Z., Zhu Y., Lin H., Zhang Z. and Sun J., “ResNeSt: Split-attention networks”, arXiv preprint arXiv:2004.08955, (2020).
40. [40] Choromanski K., Likhosherstov V., Dohan D., Song X., Gane A., Sarlos T., Hawkins P., Davis J., Mohiuddin A., Kaiser Ł., Belanger D., Colwell L. and Weller A., “Rethinking attention with performers”, In Proceedings of the International Conference on Learning Representations (ICLR), Avusturya, Online, (2021).
41. [41] Touvron H., Cord M., Douze M., Massa F., Sablayrolles A. and Jégou H., “Training data-efficient image transformers & distillation through attention”, In Proceedings of the 38th International Conference on Machine Learning (ICML), Online, 10347-10357, (2021).
42. [42] Guo M. H., Lu C. Z., Liu Z.-N., Cheng M. M. and Hu S. M., “Attention mechanisms in computer vision: A survey”, Computational Visual Media (CVM), 8(3): 331-368, (2022).
43. [43] Vlahogianni E. I., Karlaftis M. G. and Golias J. C., “Short-term traffic forecasting: Where we are and where we’re going”, Transportation Research Part C, 43: 3-19, (2014).
44. [44] Williams B. M. and Hoel L. A., “Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process: Theoretical basis and empirical results”, Journal of Transportation Engineering, 129(6): 664-672, (2003).
45. [45] Duan Y., Lv Y. and Wang F. Y., “Travel time prediction with LSTM neural network”, In Proceedings of the 19th IEEE International Conference on Intelligent Transportation Systems (2016), Brezilya, 1053-1058, (2016).
46. [46] Yu B., Yin H. and Zhu Z., “Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting”, In Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI), Sweden, 3634-3640, (2018).
47. [47] Gu Y., Qian S. and Chen F., “From Twitter to detector: Real-time traffic incident detection using social media data”, Transportation Research Part C: Emerging Technologies, 73: 336-347, (2016).
48. [48] Yao Y., Xu M., Wang Y., Crandall D. J. and Atkins, E. M., “Unsupervised traffic accident detection in first-person videos”, In Proceedings of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), China, 273–280, (2019).
49. [49] Ghahremannezhad H., Shi H. and Liu C., “Real-time accident detection in traffic surveillance using deep learning”, In Proceedings of the 2022 IEEE International Conference on Imaging Systems and Techniques (IST), China, 1-6, (2022).
50. [50] Noomwongs N., Thitipatanapong R., Chantranuwathana S. and Klongnaivai S., “Driver behavior detection based on Multi-GNSS precise point positioning technology”, Applied Mechanics and Materials, 619: 327-331, (2014).
51. [51] Klein L. A., Mills M. K. and Gibson D. R. P., “Traffic Detector Handbook: Volume I”, Federal Highway Administration (FHWA), USA, (2006).
52. [52] Zhao Z., Guo H. and Wang, Y., “A multi-information fusion anomaly detection model based on convolutional neural networks and AutoEncoder”, Scientific Reports, 14, 16147, (2024).
53. [53] Chandola V., Banerjee A. and Kumar V., “Anomaly detection: A survey”, ACM Computing Surveys, 41(3): 1-58, (2009).
54. [54] Barceló J., Montero L., Marques L. and Carmona C., “Travel time forecasting and dynamic origin-destination estimation for freeways based on Bluetooth traffic monitoring”, Transportation Research Record, 2175(1): 19-27, (2010).
55. [55] Eriksson J., Girod L., Hull B., Newton R., Madden S. and Balakrishnan H., “The pothole patrol: Using a mobile sensor network for road surface monitoring”, In Proceedings of the 6th International Conference on Mobile Systems, Applications, and Services (MobiSys), USA, 29-39, (2008).
56. [56] Herrera J. C., Work D. B., Herring R., Ban X. J., Jacobson Q. and Bayen A. M., “Evaluation of traffic data obtained via GPS-enabled mobile phones: The Mobile Century field experiment”, Transportation Research Part C: Emerging Technologies, 18(4): 568-583, (2010).
57. [57] Sun Z., Bebis G. and Miller R., “On-road vehicle detection: A review”, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 28(5): 694-711, (2006).
58. [58] Lv Y., Duan Y., Kang W., Li Z. and Wang F. Y., “Traffic flow prediction with big data: a deep learning approach”, IEEE Transactions on Intelligent Transportation Systems(T-ITS), 16(2): 865-873, (2015).
59. [59] Zhu L., Yu F. R., Wang Y., Ning B. and Tang T., “Big data analytics in intelligent transportation systems: A survey”, IEEE Transactions on Intelligent Transportation Systems (T-ITS), 20(1): 383-398.
60. [60] Bayoudh K., Hamdaoui F. and Mtibaa A., “Transfer learning based hybrid 2D-3D CNN for traffic sign recognition and semantic road detection applied in advanced driver assistance systems”, Applied Intelligence, 51(1): 124–142, (2021).
61. [61] Feng R., Chen M. and Song Y., “Learning traffic as videos: Short-term traffic flow prediction using mixed-pointwise convolution and channel attention mechanism”, Expert Systems with Applications, 240: 122468, (2024).
62. [62] Zheng G., Chai W. K., Duanmu J.-L. and Katos, V., “Hybrid deep learning models for traffic prediction in large-scale road networks”, Information Fusion, 91: 685–707, (2022).
63. [63] Duman Z. N., Murat Y. Ş. ve Yılmaz M., “Koordine kavşaklarda sinyalizasyon sistemine etki eden parametrelerin incelenmesi vaka durum çalışması”, Politeknik Dergisi, 27(4): 1515-1539, (2024).
64. [64] Koçak A., Alkan M. and Arıkan M. S., “Deepfake video detection using convolutional neural network based hybrid approach”, Journal of Polytechnic, 28(3): 957-968, (2025).
65. [65] Bicer M. B., Eliiyi U. and Tursel-Eliiyi D., “Deep learning-based classification of breast tumors using raw microwave imaging data”, Journal of Polytechnic, 27(4): 1327-1334, (2024).
66. [66] Karadağ B., Arı A. ve Karadağ M., “Derin öğrenme modellerinin sinirsel stil aktarımı performanslarının karşılaştırılması”, Politeknik Dergisi, 24(4): 1611-1622, (2021).
67. [67] Karadağ B. ve Arı A., “Akıllı mobil cihazlarda YOLOv7 modeli ile nesne tespiti”, Politeknik Dergisi, 26(3): 1207-1214, (2023).
68. [68] Öztürk S., Ağapınar Şahin, S., Aksoy A. N., Arı B. and Akinbi A. “A novel approach for cardiotocography paper digitization and classification for abnormality detection”, IEEE Access, 11: 42521–42533, (2023).
69. [69] Arı B., Arı A., Şengür A. and Arslan Tuncer S., “Derin öznitelikler kullanılarak aşırı öğrenme makineleri ile kayısı yapraklarının sınıflandırılması”, 2019 International Conference on Artificial Intelligence and Data Processing (IDAP’19), 1–6, (2019).
70. [70] Arı B., Şengür A. ve Arı A., “Yerel alıcı alanlar aşırı öğrenme makinesi ile kayısı yaprağı tanıma”, International Artificial Intelligence and Data Processing Symposium (IDAP’16), 532–535. (2016).
71. [71] Karasar N., “Bilimsel araştırma yöntemi”, 23. Baskı, Nobel Akademik Yayıncılık, 978-605-5426-58-3 Ankara, (2012).