Title: Vehicle traffic information management based on big data technology and DETR algorithm

Authors: Qizheng Yang; Meng Gao; Kai Sun

Addresses: Shandong Zaofa Property Co., Ltd., Zaozhuang, Shandong Province, China ' Xuzhou Xuzhuang Experimental Primary School, Xuzhou, Jiangsu, China ' Zaozhuang Big Data Centre, Zaozhuang, Shandong Province, China

Abstract: A vehicle tracking algorithm using Detection Transformer (DETR) for small data annotation is proposed to enhance vehicle recognition and tracking accuracy in traffic management. Principal component analysis is introduced to handle complex vehicle feature data. Combining these methods, a vehicle traffic information management system based on big data technology and DETR is constructed. Experiments show that the DETR-based algorithm achieves an average accuracy of 0.96 and a loss value of 0.93, outperforming other algorithms. The system has a recognition accuracy of 95.2% and processes keyframe information at 65 frames per second, significantly better than other models. Results indicate that the proposed algorithm and system effectively improve vehicle information management accuracy.

Keywords: DETR; PCA; big data technology; vehicle traffic; information management.

DOI: 10.1504/IJVICS.2026.152938

International Journal of Vehicle Information and Communication Systems, 2026 Vol.11 No.2, pp.99 - 115

Received: 31 Oct 2024
Accepted: 30 Dec 2024
Published online: 15 Apr 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article