ABSTRACT
With the development of traffic system, real-time monitoring and accurate identification of road traffic conditions become increasingly important. This paper proposes an intelligent traffic monitoring solution based on YOLOv5 algorithm in visible light imaging system, aiming at improving the NMS algorithm of YOLOv5 algorithm by selecting appropriate activation function, improving IoU algorithm used in screening candidate boxes, and improving the recognition accuracy of the whole algorithm for complex road conditions, so as to improve the detection effect of the whole system. Finally, two data sets are selected for model learning and testing. it is verified that the recognition accuracy of the improved algorithm is improved by about 2.5 percentage points compared with the original YOLOv5 algorithm.
- Yan Ke and R. Sukthankar, "PCA-SIFT: a more distinctive representation for local image descriptors[C]," Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004., pp. II-II.Google Scholar
- REDMON J, DIVVALA S, GIRSHICK R, You Only Look Once: Unified, Real-Time Object Detection[C] The IEEE Conference on Computer Vision and Pattern Recognition. Las- Vegas: IEEE, 2016:779-788.Google Scholar
- WEI L, DRAGOMIR A, DUMITRU E, SSD: Single Shot Multi Box Detector[C] Proceedings of the 2014 European Conference on Computer Vision, LNCS9905. Cham: Sprin ger,2016:21-37.Google Scholar
- REDMON J, FARHADI A. YOLO9000: Better, Faster, Stronger[C]. The IEEE Conference on Computer Vision and Pattern Recognition,2017:7263-7271.Google Scholar
- C. -Y. Wang, A. Bochkovskiy and H. -Y. M. Liao, "Scaled-YOLOv4: Scaling Cross Stage Partial Network," 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 13024-13033.Google Scholar
- Tao. Research on road condition recognition technology based on near infrared imaging [D]. Hunan University, 2020.Google Scholar
- Yuan Research on infrared road condition detection and recognition in complex environment [J]. Modern Computer, 2022, 28 (23):1-9Google Scholar
- Deng. Research on traffic scene recognition based on YOLOv5 and improved DeepLabv3+[D]. Tianjin Vocational and Technical Normal University, 2023.Google Scholar
- Hua Study on surface defect detection algorithm of aluminum profile based on improved YOLOV5 [D]. Tianjin Vocational and Technical Normal University, 2023.Google Scholar
- Abdullah As Sami;Saadman Sakib;Kaushik Deb;Iqbal H. Sarker Improved YOLOv5-Based Real-Time Road Pavement Damage Detection in Road Infrastructure Management Journal[J]Algorithms. Volume 16 , Issue 9. 2023.Google Scholar
- LI Haiqing. Research and implementation of underwater swarm target detection based on improved YOLOv5 [D]. Dalian Ocean University, 2023.Google Scholar
- Cardellicchio Angelo;Ruggieri Sergio;Nettis Andrea;Mosca Nicola;Uva Giuseppina;Renò Vito On the use of YOLOv5 for detecting common defects on existing RC bridges Conference[P] . Volume 12621, 2023. PP 126210G-126210G-8.Google Scholar
- Wang Bolt Positioning Detection Based on Improved YOLOv5 for Bridge Structural Health Monitoring Journal[J] Sensors.Volume 23 , Issue 1. 2022. PP 396-396.Google Scholar
- Falahat Shahrzad;Karami Azam Maize tassel detection and counting using a YOLOv5-based model Journal[J]Multimedia Tools and Applications. Volume 82 , Issue 13 . 2022. PP 19521-19538.Google Scholar
Recommendations
Research on Recognition Methods of Bus Front Road Condition Based on Video
ICIG '13: Proceedings of the 2013 Seventh International Conference on Image and GraphicsAmong safe driving systems of intelligent transportation, the road recognition for bus driving, Especially for front road driving condition, becomes more and more important. Moreover, improving the road recognition of public transport vehicles can ...
CNN‐Transformer for visual‐tactile fusion applied in road recognition of autonomous vehicles
Highlights- CNN-transformer based multimodal fusion network for accurate road recognition.
- Fusion feature extractor structure improving the performance of bottleneck fusion method.
- Vision-tactile fusion road recognition method based on ...
AbstractReliable autonomous driving requires comprehensive environment perception, among which the road recognition is critical for autonomous vehicles to achieve adaptability, reliability, and safety. Existing equipped sensors such as cameras, LiDAR, ...
Vision based approaches for driver assistance systems
ICAI'10: Proceedings of the 11th WSEAS international conference on Automation & informationThe number of cars on the roads increases with every passing year. Unfortunately, the number of accidents tends to be proportional with this increase. Nowadays, many vehicles are augmented with conventional driver assistance systems (DAS) such as: ...
Comments