An Improved YOLOv2 for Vehicle Detection
Open accessThe DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index.
Abstract
Vehicle detection is one of the important applications of object detection in intelligent transportation systems. It aims to extract specific vehicle-type information from pictures or videos containing vehicles. To solve the problems of existing vehicle detection, such as the lack of vehicle-type recognition, low detection accuracy, and slow speed, a new vehicle detection model YOLOv2_Vehicle based on YOLOv2 is proposed in this paper. The k-means++ clustering algorithm was used to cluster the vehicle bounding boxes on the training dataset, and six anchor boxes with different sizes were selected. Considering that the different scales of the vehicles may influence the vehicle detection model, normalization was applied to improve the loss calculation method for length and width of bounding boxes. To improve the feature extraction ability of the network, the multi-layer feature fusion strategy was adopted, and the repeated convolution layers in high layers were removed. The experimental results on the Beijing Institute of Technology (BIT)-Vehicle validation dataset demonstrated that the mean Average Precision (mAP) could reach 94.78%. The proposed model also showed excellent generalization ability on the CompCars test dataset, where the “vehicle face” is quite different from the training dataset. With the comparison experiments, it was proven that the proposed method is effective for vehicle detection. In addition, with network visualization, the proposed model showed excellent feature extraction ability.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures













Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“As compared to the state-of-the-art-algorithms that were employed in previous studies [29][30][31][32], the proposed algorithm attained the highest performance, followed by [29], as it achieved a mAP of 94.78% while using the YOLOv2 object detector as shown in Figure 10.”
From A Real-Time Computer Vision Based Approach to Detection and Classification of Traffic Incidents (2023) · cites this paper as [29]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“A new network structure Darknet-19 was designed by removing the full connection layers of the network, and batch normalization [27] was applied to each layer Referring to the anchor mechanism of Faster R-CNN, k-means clustering was used to obtain the anchor boxes.”
MethodLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472 · cited 1×
“However, as a general object detection model, YOLOv2 is applicable to cases where there are a variety of classes to be detected, and the differences among the classes are large, such as persons, horses, and bicycles.”
You Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 3×
“In this paper, two vehicle datasets collected from road monitoring, the Beijing Institute of Technology (BIT)-Vehicle [28] and CompCars [29], were used.”
MethodDensely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684
Girshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 2×
“CompCars only includes two vehicle types: sedan and SUV, with more than 40,000 images.”
BackgroundYOLO9000: Better, Faster, Stronger
Redmon, J., & Farhadi, A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.690 · Cited by 19,892
Visualizing and understanding convolutional networks
Zeiler, M. D., & Fergus, R. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10590-1_53 · Cited by 15,839 · cited 3×
“This method can observe of vehicle edge information, which indicates that the convolution kernel in the first layer successfully shows”
MethodInfluential
Research works printed but not linked: 2
- Sang, 2017, Vehicle detection based on faster-RCNN, J. Chongqing Univ. (Nat. Sci. Ed.), 40, 32
- Arthur, D., and Vassilvitskii, S. (2007, January 7–9). k-means++: The advantages of careful seeding. Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms, New Orleans, LA, USA.
Cited by
Items 201–232 of 232
Vehicle Detection based on Positive Samples Selection with Low Threshold
Sang, J., Qiao, X., Wu, Z., Tian, S., & Xia, X. (2022) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc53654.2022.9945360
“In [20], the loss function and the generation method of the anchor box in YOLOv2 were improved to increase accuracy and speed of vehicle detection.”
BackgroundMethodSelf-citationAge Detection of Catfish Breeding Based on Size Using The YOLO V3
Yaddarabullah, Y., Firdaus, M. R., Permana, S. D. H., Syahputra, A., Arifitama, B., & Krishnasari, E. D. (2022) · International Seminar on Research of Information Technology and Intelligent Systems (ISRITI) · doi:10.1109/isriti56927.2022.10053004
“This is because in YOLOv2 the detection process is only carried out on 2 different scales [5].”
BackgroundApplication of YOLOv5 in Device Detection of Hydropower Station
Zhao, S., Wen, C., Zhao, Y., Nie, L., Zhang, X., Zou, J., & Wu, Y. (2023) · Lecture notes in electrical engineering · doi:10.1007/978-981-99-0553-9_6
A Improved Prior Box Generation Method for Small Object Detection
Zhou, X., Luo, Z., & Li, S. (2023) · Communications in computer and information science · doi:10.1007/978-981-99-2385-4_35
Mostofa, M. (2023) · The Research Repository @ WVU (West Virginia University) · doi:10.33915/etd.11758
Real time car make and model detection for Indian market
Darapaneni, N., Konduru, V. R., Karthikeyan, Karthik, P., Prasanna, S., Darshini, D., Paduri, A. R., & Velavan, S. (2023) · AIP conference proceedings · doi:10.1063/5.0133483
Novel Annotation and Metrics for Mangrove Species Classification Using Bounding Box Object Detection
Lim, H. S., Lee, Y., Eu, K. S., Lin, M., Yap, K. M., & Chia, W. C. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222498 · Cited by 1
“This application is highly customisable and was used to detect various objects in past research including human faces [1, 2], pedestrians [3], vehicles [4], and forest objects [5 – 7].”
MethodUSING YUMI ROBOT AND RGB-D CAMERA WITH YOLOV5 FOR PICK-AND-PLACE APPLICATION, 1-10.
Kijdech, D., & Vongbunyong, S. (2023) · International Journal of Robotics and Automation · doi:10.2316/j.2023.206-0927
Multi-Sensor Information Fusion Algorithm for Night Driving Vehicle DetectionPreprint
Chen, Y., Shi, Z., Luo, H., Li, W., & Deng, Y. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4626471
A novel vehicle line-pressing detection framework based on 3D object detection
Zheng, G., Jie, L., Qin, Y., & Tan, B. (2023) · International Conference on Signal Processing and Computer Science (SPCS) · doi:10.1117/12.3012199
“After the development of deep learning, these tasks can be completed using targeted neural networks [6][8] .”
BackgroundMulti-sensor information fusion algorithm for night driving vehicle detectionPreprint
Chen, Y., Shi, Z., Luo, H., Li, W., & Deng, Y. (2024) · Research Square · doi:10.21203/rs.3.rs-3910068/v1
Classifying Negative Objects With Neural Networks
Hooper, K. (2024) · doi:10.32920/25190984.v1
Geospatial Object Detection in Hyperspectral Imagery Using Spectral-Spatial Networks
Prabakar, D., Pranavan, S., Kumbhkar, M., Sah, S., Patil, H., & Firos, A. (2023) · International Conference on Intelligent Computing and Control for Engineering and Business Systems (ICCEBS) · doi:10.1109/iccebs58601.2023.10449075
“The use of hyperspectral imaging, often known as HSI, has become more important in the field of remote sensing since it provides a wealth of information over a wide variety of spectral bands[1].”
DATS: Driver assistance text spotter
Dave, D., Gandhi, J., Patel, P. B., & Bagi, R. (2024) · AIP conference proceedings · doi:10.1063/5.0209773
Car Detection Using Histogram of Oriented Gradients (HOG) Features on Autonomous Vehicle
Gunawan, Setiawan, W., Sugiarto, B., Laluma, R. H., Suharsono, T. N., & Sukmana, R. N. (2024) · International Conference on Telecommunication Systems, Services, and Applications · doi:10.1109/tssa63730.2024.10864315
“Various YOLO structures and architectures are used in research to detect cars with varying accuracy [8], [9], [10].”
Real-World Security Applications Through Computer Vision
Dalai, A. K., & Mohapatra, H. (2025) · IGI Global eBooks · doi:10.4018/979-8-3693-9405-2.ch013
Vehicle Detection Using Image Conversion Percentage to Binary Method Based on K-Means
Irawan, C., Ningrum, A. P., & Nohan, R. (2024) · Journal of Applied Intelligent System · doi:10.62411/jais.v9i2.11178
Multi-Input All-Weather Streetlight to Reduce Carbon Footprint of Illuminating Road Infrastructure
Chen, C., & Shen, W. (2024) · Journal of Student Research · doi:10.47611/jsrhs.v13i4.8337
Dong, L., Zhu, H., Chen, Y., & Wang, D. (2025) · Review of Scientific Instruments · doi:10.1063/5.0276882
Sherin, B. C., & Poovammal, E. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-6435-1_19
Sung, Y.-T., Kim, S., Park, N., Park, G., & Park, J. (2025) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981251378132
Yuan, Q., Raza, M., Salem, S., Kwon, H., Hussain, J., Gu, Y. H., & Al–antari, M. A. (2025) · International Conference on Intelligent Perception and Computer Vision (CIPCV) · doi:10.1109/cipcv65863.2025.00017
“In [18]”
Li, Y. E., Yang, W., Tian, H., Wang, Z., & Wang, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-9849-3_8
Digital image processing methods in industry environment
Svobodová, P. (2024) · DSpace VŠB-TUO (VŠB-TUO)
Application of machine learning models for tracking and counting heavy vehicles on highways
Araújo, B. D. S., Achy, A. R. A., Brito, D. J. C., Pagano, T. P., Costa, W. G. S., & Almeida, M. S. S. (2025) · Congresso Brasileiro de Inteligência Computacional (CBIC) · doi:10.21528/cbic2025-1184872
Research on Flame and Smoke Target Detection Algorithm Based on Improved YOLOv8n
Geng, T., & Yang, W. (2025) · IEEE International Conference on Control, Electronics and Computer Technology (ICCECT) · doi:10.1109/iccect64621.2025.11339604
“For example, Xue et al. improved YOLOv2 [6] for firework detection in complex backgrounds, significantly enhancing the model’s environmental adaptability.”
EFI-YOLO: An enhanced framework for industrial object detection
Zhao, K., Miao, X., Huang, C., & Li, J. (2026) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2026.104796
Lightweight YOLOv8-Based Vehicle Detector for Intelligent Transportation Systems in South Asia
Haq, S., Mishra, S. R., Behera, B., & Dash, S. (2026) · International Conference on Emerging Systems and Intelligent Computing (ESIC) · doi:10.1109/esic68176.2026.11495758
“Their method enabled the object detection system to run at 5-17 fps. Sang et al. [11] in 2018, in order to detect vehicles and to improve the YOLOv2 model, proposed a YOLOv2 Vehicle model that was designed with a multi-layer feature fusion strategy along with the removal of the repeated convolution…”
Leleszi, A., Vass, D., Polonkai, D., Cseh, B., & Dózsa, C. (2026) · IME - Az egészségügyi vezetők szaklapja · doi:10.53020/ime-2026-102
Yang, R., Xu, Y., Ge, L., & Yu, L. (2025) · doi:10.1109/ccat68690.2025.00010
“Subsequent versions YOLOv2 [7] and YOLOv3 [8] further enhanced detection performance by introducing anchor box mechanisms and multi-scale detection techniques.”
Li, Z., Huang, M., Zhu, H., Tang, Y., Zhao, J., & Zhang, Q. (2026) · International Journal of Rock Mechanics and Mining Sciences · doi:10.1016/j.ijrmms.2026.106655
Self-citationRagoubi, M., Tebbaa, B., Hajjami, L. E., Arsalane, A., Klilou, A., & Žuraulis, V. (2026) · Sustainability · doi:10.3390/su18179108
Metadata
- Authors
- Sang, J., Wu, Z., Guo, P., Hu, H., Xiang, H., Zhang, Q., & Cai, B.
- Affiliations
- Chongqing University (CN)
- Corresponding author
- Sang, Jun
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2018-12-04
- Volume
- 18
- Issue
- 12
- Pages
- 4272
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC6308705
- OpenAlex
- W2902306794
- Cited by
- 223
- References
- 28
- Citations by year
- 2026: 9 · 2025: 28 · 2024: 36 · 2023: 43 · 2022: 35 · 2021: 31 · 2020: 29 · 2019: 11 · 2018: 1
- FWCI
- 9.00
- Type
- journal article
- Language
- en
- Open access
- gold
- dx.doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, other-oa)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Chongqing Research Program of Basic Science and Frontier Technology (cstc2017jcyjB0305)
- National Key R&D Program of China (2017YFB0802400)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- vehicle detection, YOLOv2, K-means clustering, multi-scale feature fusion, vehicle type recognition, intelligent transportation systems, object detection, network visualization
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.