The quest for accurate traffic density estimation is gaining momentum globally, with Vietnam distinguished by its ranking among the top ten nations for private vehicle usage. Rapid advancements in computer vision, particularly through the development of convolutional neural network (CNN) methodologies, underscore the pressing need to incorporate these techniques into traffic density estimation efforts. In this study, three convolutional neural network (CNN) mod- els—W-Net, UASD-Net (a fusion of U-Net with Adaptive Scenario Discovery), and CSR-Net (Congested Scene Recognition Network)—are employed to quan- tify and assess traffic density based on images captured in Vietnam. Furthermore, a novel approach for reallocating label points to generate more accurate density maps is proposed. Experimental results on a composite dataset, integrating the TRANCOS, TayDo, and KienGiang datasets, demonstrate promising mean ab- solute error rates of 3.67, 4.42, and 3.82 for W-Net, UASD-Net, and CSR-Net, respectively.
Tạp chí: 11th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI 2025), Yogyakarta, Indonesia on September 26-27, 2024
Tạp chí: The 4th International Conference on Innovations in Social Sciences Education and Engineering (ICoISSEE-4) Bandung, Indonesia, July, 20th, 2024
Tạp chí khoa học Trường Đại học Cần Thơ
Lầu 4, Nhà Điều Hành, Khu II, đường 3/2, P. Xuân Khánh, Q. Ninh Kiều, TP. Cần Thơ
Điện thoại: (0292) 3 872 157; Email: tapchidhct@ctu.edu.vn
Chương trình chạy tốt nhất trên trình duyệt IE 9+ & FF 16+, độ phân giải màn hình 1024x768 trở lên