Fleet-Vehicle Identification and Geolocalization using Inexpensive Cameras******

使用廉价相机进行车队车辆识别和地理定位*****

• 批准号: 537057-2018
• 负责人: Nicolescu, Gabriela
• 金额: $ 1.82万
• 依托单位: École Polytechnique de Montréal
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=660928
• 关键词: Fleet; Vehicle; Identification; Geolocalization; using

The objective of the project is to develop image-recognition software that can process images and video captured with fixed, inexpensive surveillance cameras to recognize vehicles from a fleet and measure their position and speed. We will propose image processing and artificial intelligence methods to recognize suitably marked vehicles (eg patterns on roofs and sides), even in less than ideal situations: poor lighting, snow, rain, dust. We will provide also optimization and characterization of required processing power and related infrastructure.**

该项目的目标是开发图像识别软件，该软件可以处理用固定的廉价监控摄像头拍摄的图像和视频，以识别车队中的车辆，并测量它们的位置和速度。我们将提出图像处理和人工智能方法来识别适当标记的车辆(例如车顶和侧面的图案)，即使在不太理想的情况下也是如此：光线不佳、下雪、下雨、灰尘。我们还将提供所需处理能力和相关基础设施的优化和表征。**