Cloud-based computer vision for consumer video monitoring application

用于消费者视频监控应用的基于云的计算机视觉

• 批准号: 452820-2013
• 负责人: Ray, Nilanjan
• 金额: $ 1.78万
• 依托单位: University of Alberta
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2013
• 资助国家: 加拿大
• 起止时间: 2013-01-01 至 2014-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=541130
• 关键词: Cloud; based; computer; vision; consumer

Availability of inexpensive video cameras, easy accessibility of internet, advancement in automated video processing/computer vision and emergence of cloud computing have contributed to the opportunity to create this research proposal. This research proposes cost efficient ways for automated consumer video monitoring. The proposed monitoring system would consist of webcams or IP cameras for home or small business monitoring. Motion events will be automatically detected and classified into different categories. Categorized motion events will be stored on a cloud platform. The user will have the capability to browse through these events, much like browsing recorded TV programs on a PVR. The uniqueness of this research provides flexibility to the user to define motion events or event categories. Real-time and accurate computer vision and automated image and video analysis algorithms will run on the cloud platform to achieve this feat.

廉价摄像机的可用性、互联网的便捷访问、自动视频处理/计算机视觉的进步以及云计算的出现促成了创建本研究提案的机会。这项研究提出了自动化消费者视频监控的经济有效的方法。拟议的监控系统将包括用于家庭或小型企业监控的网络摄像头或 IP 摄像机。运动事件将被自动检测并分类为不同的类别。分类的运动事件将存储在云平台上。用户将能够浏览这些事件，就像浏览 PVR 上录制的电视节目一样。这项研究的独特性为用户提供了定义运动事件或事件类别的灵活性。实时准确的计算机视觉以及自动化图像和视频分析算法将在云平台上运行来实现这一壮举。