RR:Instrumentation for Experimental Research in Intelligent Information Access, Environmental Monitoring, and Large-Scale Pedestrian Simulation

RR：智能信息访问、环境监测和大规模行人模拟实验研究仪器

• 批准号: 0423733
• 负责人: Ronald Metoyer
• 金额: $ 8万
• 依托单位: Oregon State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2004
• 资助国家: 美国
• 起止时间: 2004-08-15 至 2006-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0423733&HistoricalAwards=false
• 关键词: RR; Instrumentation; Experimental; Research; Intelligent

This project, purchasing networking equipment, a database server, file servers, and digital video editing, supports the following research projects: Intelligent desktop interfaces, Large-scale pedestrian evacuation simulation, and Insect recognition and population counting for environmental monitoring.In order to provide computer users with a more timely and useful application context, the 1st project, Task Tracer, develops machine learning algorithms for understanding task context and a new paradigm for desktop interface design. The 2nd project produces large volumes of motion data of people with varying abilities (able-bodied, wheel chair mobile, etc.) for distribution to the pedestrian simulation community via the web. New and accurate models for 2- and 3-dimensional representations of human motion for heterogeneous crowds are expected from this work. Image processing and computer vision techniques for segmenting and identifying various insects are employed in the last project; whereas machine learning techniques are utilized for processing this data to classify the insect into the proper genus species. This project should result in new devices for imaging insect specimens.Broader Impact: This research should improve the effectiveness of computers for: Understanding and streamlining human productivity with computer systems, Helping emergency personnel to better understand and respond to emergency egress situations, and Automating the recognition and counting of insect populations for important environmental monitoring activities.

本项目购买网络设备、数据库服务器、文件服务器和数字视频编辑，支持以下研究项目：智能桌面接口、大规模行人疏散模拟、环境监测昆虫识别和种群计数。为了给计算机用户提供更及时、更有用的应用上下文，第一个项目任务跟踪器（Task Tracer）开发了用于理解任务上下文的机器学习算法和桌面界面设计的新范式。第二个项目产生大量不同能力的人（健全人，轮椅移动等）的运动数据，通过网络分发给行人模拟社区。新的和准确的2和3维的人体运动表示异质人群的模型有望从这项工作。最后一个项目采用图像处理和计算机视觉技术对各种昆虫进行分割和识别；而机器学习技术则用于处理这些数据，以将昆虫分类到适当的属物种中。这个项目应该会产生用于昆虫标本成像的新设备。更广泛的影响：本研究将提高计算机的有效性：通过计算机系统了解和简化人类生产力，帮助应急人员更好地了解和响应紧急出口情况，以及为重要的环境监测活动自动化识别和计数昆虫种群。