Object detection from low-resolution embedded security camera lighting system

低分辨率嵌入式安全摄像头照明系统的物体检测

• 批准号: 503659-2016
• 负责人: Ward, Rabab
• 金额: $ 1.82万
• 依托单位: University of British Columbia
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2016
• 资助国家: 加拿大
• 起止时间: 2016-01-01 至 2017-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=609537
• 关键词: Object; detection; low; resolution; embedded

The video surveillance system of interest here is a smart light bulb with the embedded sound and security camera installed and wirelessly communicated. The collected video data can be used for smart home control and monitoring. The efficacy of such a system is highly dependent on our capabilities to automatically analyze and interpret the video data. Furthermore, a major concern in video surveillance systems is privacy. Therefore, the collected video needs to be blurred at the server side to prevent any identity recognition. The goal of this project is to develop advanced machine learning and video processing methods for automated object recognition from low-resolution blurred video surveillance data. A reliable solution to this problem will provide the partner company a unique tool in the field of smart video surveillance that ensures privacy while providing smart home management.

这里感兴趣的视频监控系统是一个智能灯泡，安装了嵌入式声音和安全摄像头并进行无线通信。采集的视频数据可用于智能家居控制和监控。这样一个系统的有效性高度依赖于我们自动分析和解释视频数据的能力。此外，视频监控系统中的一个主要问题是隐私。因此，收集的视频需要在服务器端进行模糊处理，以防止任何身份识别。该项目的目标是开发先进的机器学习和视频处理方法，用于从低分辨率模糊视频监控数据中自动识别对象。这个问题的可靠解决方案将为合作伙伴公司提供智能视频监控领域的独特工具，在提供智能家居管理的同时确保隐私。