A Review of Object Detection Techniques in IoT-Based Intelligent Transportation Systems
Source abstract
Abstract
The Intelligent Transportation System (ITS), as a vital means to alleviate traffic congestion and reduce traffic accidents, demonstrates immense potential in improving traffic safety and efficiency through the integration of Internet of Things (IoT) technologies. The enhancement of its performance largely depends on breakthrough advancements in object detection technology. However, current object detection technology still faces numerous challenges, such as accuracy, robustness, and data privacy issues. These challenges are particularly critical in the application of ITS and require in-depth analysis and exploration of future improvement directions. This study provides a comprehensive review of the development of object detection technology and analyzes its specific applications in ITS, aiming to thoroughly explore the use and advancement of object detection technologies in IoT-based intelligent transportation systems. To achieve this objective, we adopted the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach to search, screen, and assess the eligibility of relevant literature, ultimately including 88 studies. Through an analysis of these studies, we summarized the characteristics, advantages, and limitations of object detection technology across the traditional methods stage and the deep learning-based methods stage. Additionally, we examined its applications in ITS from three perspectives: vehicle detection, pedestrian detection, and traffic sign detection. We also identified the major challenges currently faced by these technologies and proposed future directions for addressing these issues. This review offers researchers a comprehensive perspective, identifying potential improvement directions for object detection technology in ITS, including accuracy, robustness, real-time performance, data annotation cost, and data privacy. In doing so, it provides significant guidance for the further development of IoT-based intelligent transportation systems.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Vehicle License Plate Recognition
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (100+)
Show source-provided references
1. Our World in Data. Deaths from road injuries, 2000 to 2021 [Internet]. Oxford, UK: Global Change Data Lab; 2024 [cited 2024 Oct 22]. Available from: https://ourworldindata.org/explorers/global-health?tab=table&showSelectionOnlyInTable=1&country=OWID_ASI~OWID_AFR~OWID_EUR~OWID_NAM~OWID_SAM~OWID_WRL&pickerSort=asc&pickerMetric=entityName&Health+Area=Injuries&Indicator=Road+injuries&Metric=Number+of+deaths&Source=WHO+%28GHE%29.
2. Vehicle ownership and income growth, worldwide: 1960–203010.5547/ISSN0195-6574-EJ-Vol28-No4-7 ↗
3. OICA. Total world vehicles in use 2020 [Internet]. [cited 2024 Oct 22]. Available from: https://www.oica.net/wp-content/uploads/Total-World-vehicles-in-use-2020.pdf.
4. IoT for intelligent traffic system
5. Smart transportation: an overview of technologies and applications10.3390/s23083880 ↗
6. Nassreddine · IoT edge intelligence · 2024 · 29110.1007/978-3-031-58388-9_10 ↗
7. Real-time computer vision for transportation safety using deep learning and IoT
8. A smart iot enabled end-to-end 3D object detection system for autonomous vehicles10.1109/TITS.2022.3210490 ↗
9. Object detection in traffic videos: a survey10.1109/TITS.2023.3258683 ↗
10. Robust real-time face detection
11. Histograms of oriented gradients for human detection
12. A discriminatively trained, multiscale, deformable part model
13. Region-based convolutional networks for accurate object detection and segmentation10.1109/TPAMI.2015.2437384 ↗
14. You only look once: unified, real-time object detection
15. Wang A, Chen H, Liu L, Chen K, Lin Z, Han J, et al. YOLOv10: real-time end-to-end object detection. arXiv: 2405.14458. 2024.10.52202/079017-3429 ↗
16. A comparison of YOLO based vehicle detection algorithms
17. Real time road lane detection and vehicle detection on YOLOv8 with interactive deployment
18. Challenges and advances in UAV-based vehicle detection using YOLOv9 and YOLOv10
19. Analysis of performance of YOLOv8 algorithm for pedestrian detection
20. An efficient approach for pedestrian detection
21. End-to-end object detection with transformers
22. DETRs Beat YOLOs on real-time object detection
23. SSD: single shot multibox detector
24. Rapid object detection using a boosted cascade of simple features
25. An enhanced Viola-Jones vehicle detection method from unmanned aerial vehicles imagery10.1109/TITS.2016.2617202 ↗
26. An efficient hardware implementation of HOG feature extraction for human detection10.1109/TITS.2013.2284666 ↗
27. Pedestrian detection based on HOG features and svm realizes vehicle-human-environment interaction
28. Intelligent system utilizing HOG and CNN for thermal image-based detection of wild animals in nocturnal periods for vehicle safety10.1080/08839514.2022.2031825 ↗
29. The study on vehicle detection based on DPM in traffic scenes
30. Effective pedestrian detection using deformable part model based on human model10.1007/s12555-016-0322-1 ↗
31. Classify vehicles in traffic scene images with deformable part-based models10.1007/s00138-017-0890-y ↗
32. ImageNet classification with deep convolutional neural networks10.1145/3065386 ↗
33. Going deeper with convolutions
34. Rich feature hierarchies for accurate object detection and semantic segmentation
35. Fast R-CNN
36. Faster R-CNN: towards real-time object detection with region proposal networks10.1109/TPAMI.2016.2577031 ↗
37. Focal loss for dense object detection
38. CornerNet: detecting objects as paired keypoints
39. Zhou X, Wang D, Krähenbühl P. Objects as points. arXiv:1904.07850. 2019.10.1007/978-3-030-58548-8_28 ↗
40. EfficientDet: scalable and efficient object detection
41. Spatial pyramid pooling in deep convolutional networks for visual recognition10.1109/TPAMI.2015.2389824 ↗
42. Mask R-CNN
43. Cascade R-CNN: high quality object detection and instance segmentation10.1109/TPAMI.2019.2956516 ↗
44. YOLO9000: better, faster, stronger
45. Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv:1502.03167. 2015.
46. Redmon J, Farhadi A. YOLOv3: an incremental improvement. arXiv:1804.02767. 2018.
47. Bochkovskiy A, Wang CY, Liao HYM. YOLOv4: optimal speed and accuracy of object detection. arXiv:2004.10934. 2020.
48. CSPNet: a new backbone that can enhance learning capability of CNN
49. Khanam R, Hussain M. What is YOLOv5: a deep look into the internal features of the popular object detector. arXiv:2407.20892. 2024.
50. Enhancing geometric factors in model learning and inference for object detection and instance segmentation10.1109/TCYB.2021.3095305 ↗
51. Li C, Li L, Jiang H, Weng K, Geng Y, Li L, et al. YOLOv6: a single-stage object detection framework for industrial applications. arXiv:2209.02976. 2022.
52. RepVGG: making VGG-style ConvNets great again
53. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
54. Wang CY, Liao HYM, Yeh IH. Designing network design strategies through gradient path analysis. arXiv:2211.04800. 2022.
55. Yaseen M. What is YOLOv8: an in-depth exploration of the internal features of the next-generation object detector. arXiv:2408.15857. 2024.
56. YOLOv9: learning what you want to learn using programmable gradient information
57. Ge Z, Liu S, Wang F, Li Z, Sun J. YOLOX: exceeding YOLO series in 2021. arXiv:2107.08430. 2021.
58. Wang CY, Yeh IH, Liao HYM. You only learn one representation: unified network for multiple tasks. arXiv:2105.04206. 2021.
59. Real-time detection method for small traffic signs based on Yolov310.1109/ACCESS.2020.2984554 ↗
60. An improved YOLO-based road traffic monitoring system10.1007/s00607-020-00869-8 ↗
61. Real-time pedestrian detection using YOLO
62. Real-time pothole detection using YOLOv5 algorithm: a feasible approach for intelligent transportation systems
63. Real-time occupancy detection of on-street parking spaces based on an edge device
64. Investigation of YOLO models in the detection and classification of multiple negative road anomalies
65. Real-time defect detection systems for steel and wood inspection
66. Research on small target detection in optical remote sensing based on YOLOv7
67. Object detection in dense and mixed traffic for autonomous vehicles with modified YOLO10.1109/ACCESS.2023.3335826 ↗
68. Enhancing vehicle detection in intelligent transportation systems via autonomous UAV platform and YOLOv8 integration10.1016/j.asoc.2024.112015 ↗
69. Utilizing YOLOv8 for enhanced traffic monitoring in intelligent transportation systems (ITS) applications10.1016/j.dsp.2024.104594 ↗
70. YOLOv9 in action: a futuristic approach to real-time license plate recognition and analysis
71. Real-time pedestrian detection algorithm of YOLOv10 under complex lighting and occlusion conditions
72. Traffic flow detection method based on improved SSD algorithm for intelligent transportation system10.1371/journal.pone.0300214 ↗
73. Inception-SSD: an improved single shot detector for vehicle detection10.1007/s12652-020-02085-w ↗
74. Deep residual learning for image recognition
75. Feature pyramid networks for object detection
76. Stacked hourglass networks for human pose estimation
77. Lin TY, Maire M, Belongie S, Bourdev L, Girshick R, Hays J, et al. Microsoft COCO: common objects in context. arXiv:1405.0312. 2015.10.1007/978-3-319-10602-1_48 ↗
78. Beyond bags of features: spatial pyramid matching for recognizing natural scene categories
79. Selective search for object recognition10.1007/s11263-013-0620-5 ↗
80. Dynamic R-CNN: towards high quality object detection via dynamic training
81. Sparse R-CNN: end-to-end object detection with learnable proposals
82. A comparative analysis of multi-label deep learning classifiers for real-time vehicle detection to support intelligent transportation systems10.3390/smartcities6050134 ↗
83. Real-time vehicle detection based on improved YOLO v510.3390/su141912274 ↗
84. YOLO-FA: type-1 fuzzy attention based YOLO detector for vehicle detection10.1016/j.eswa.2023.121209 ↗
85. Kenk MA, Hassaballah M. DAWN: vehicle detection in adverse weather nature dataset. arXiv:2008.05402. 2020.
86. UA-DETRAC: a new benchmark and protocol for multi-object detection and tracking10.1016/j.cviu.2020.102907 ↗
87. Research on vehicle detection based on improved YOLOX_S10.1038/s41598-023-50306-x ↗
88. Coordinate attention for efficient mobile network design
89. Mungoli N. Adaptive feature fusion: enhancing generalization in deep learning models. arXiv:2304.03290. 2023.
90. A traffic surveillance multi-scale vehicle detection object method base on encoder-decoder10.1109/ACCESS.2020.2979260 ↗
91. Revolutionizing target detection in intelligent traffic systems: YOLOv8-snakevision10.3390/electronics12244970 ↗
92. Vehicle detection in urban traffic surveillance images based on convolutional neural networks with feature concatenation10.3390/s19030594 ↗
93. Multi-object detection in traffic scenes based on improved SSD10.3390/electronics7110302 ↗
94. Gabor convolutional networks10.1109/TIP.2018.2835143 ↗
95. Sudhakaran S, Lanz O. Convolutional long short-term memory networks for recognizing first person interactions. arXiv:1709.06495. 2017.10.1109/ICCVW.2017.276 ↗
96. Vehicle object detection based on improved RetinaNet10.1088/1742-6596/1757/1/012070 ↗
97. Weighted feature pyramid networks for object detection
98. Vehicle recognition algorithm based on Haar-like features and improved AdaBoost classifier10.1007/s12652-021-03332-4 ↗
99. Occlusion-aware detection for internet of vehicles in urban traffic sensing systems10.1007/s11036-020-01668-3 ↗
100. An edge traffic flow detection scheme based on deep learning in an intelligent transportation system10.1109/TITS.2020.3025687 ↗