Camera-based vehicle detection and classification

大城市中心的交通拥堵每年给通勤者造成数十亿美元的损失。它还对通勤者的健康和生活质量产生不利影响。考虑到目前的人口趋势，除非采取有力措施遏制交通增长，否则交通拥堵似乎只会随着时间的推移而恶化。城市和市政当局正在探索的一种低成本策略是使用智能交通分阶段系统来处理迅速增长的交通。这些系统通过道路网络引导交通，同时最大限度地减少旅行时间，减少停车次数并节省燃料成本。标准电力公司是一家加拿大智能交通检测硬件和系统的供应商/顾问/OEM。他们为加拿大各地的许多市政当局提供道路交通数据收集设备。在这个项目中，库雷希博士在安大略理工大学的视觉计算实验室将与标准电力公司合作开发一个基于摄像头的车辆检测和分类系统。该系统将使用安装在十字路口的交通摄像头对现场的车辆进行分类。开发的系统将为后续的交通监测和分析提供丰富的数据来源，包括改进控制交通路口的方法，以缓解交通拥堵。

Traffic congestion in big urban centers cost billions of dollars yearly to commuters. It also adversely affects the health and quality-of-life of commuters. Given current population trends, it seems that traffic congestion will only worsen over time unless drastic measures are taken to curb traffic growth. One low-cost strategy that cities and municipalities are exploring for dealing with burgeoning traffic is the use of intelligent traffic phasing systems. These systems route traffic through road networks while minimizing travel times, reducing stops and saving fuel costs. Standard Power Corporation is a Canadian supplier/consultant/OEM of intelligent traffic detection hardware and systems. They provide road traffic data gathering equipment to a number of municipalities across Canada. In this project, Dr. Qureshi's Visual Computing Lab at the University of Ontario Institute of Technology will work with Standard Power Corporation to develop a camera-based vehicle detection and classification system. The system will use traffic cameras mounted on road intersections to classify vehicles present at the scene. The developed system will provide a rich source for data for subsequent traffic monitoring and analysis, including improved ways to control traffic intersections to ease traffic congestion.