Abstract
Recent advancements in computer vision, driven by artificial intelligence, have significantly enhanced monitoring systems. One notable application is traffic monitoring, which leverages computer vision alongside deep learning-based object detection and counting. We present an offline, real-time traffic monitoring system that couples a pre-trained YOLOv11 detector with BoT-SORT/ByteTrack for multi-object tracking, implemented in PyTorch/OpenCV and wrapped in a Qt-based desktop UI. The CNN pipeline enables efficient vehicle detection and counting from video streams without cloud dependencies. Across diverse scenes, the system achieves (66.67-95.83%) counting accuracy. Class-wise detection yields high precision (cars: 0.97-1.00; trucks: 1.00) with strong recall (cars: 0.82-1.00; trucks: 0.70-1.00), resulting in F1 scores of (0.90-1.00 for cars and 0.82-1.00 for trucks). While adverse weather conditions may negatively impact this performance, results remain robust in typical conditions. By integrating lightweight models with an accessible, cloud-independent interface, this paper contributes to the modernization and development of future smart cities by showing the capacity of AI-driven traffic monitoring systems.
Abstract (translated)
在人工智能的驱动下，计算机视觉技术的 recent 显著提升了监控系统性能。其中一个 notable 应用是交通监控，该技术结合计算机视觉与基于深度学习的物体检测和计数方法。本研究提出了一种离线实时交通监控系统，该系统将预训练的 YOLOv11 检测器与 BoT-SORT/ByteTrack 多目标跟踪算法相结合，采用 PyTorch/OpenCV 实现，并封装在基于 Qt 的桌面界面中。该 CNN 处理流程无需云端支持即可高效实现视频流中的车辆检测与计数。在不同场景下，系统计数准确率达到 (66.67%-95.83%)。按车辆类别细分的检测结果显示：小汽车检测精确度为 0.97-1.00，卡车精确度为 1.00；召回率方面，小汽车为 0.82-1.00，卡车为 0.70-1.00，相应的 F1 分数分别为 (小汽车 0.90-1.00，卡车 0.82-1.00)。尽管恶劣天气条件可能对性能产生负面影响，但在典型环境下系统表现依然稳健。通过将轻量化模型与无需云端的易用界面相结合，本研究展示了人工智能驱动交通监控系统的能力，为未来智慧城市的现代化与发展做出了贡献。
URL
https://arxiv.org/abs/2604.04080
