RAPID: A Smart and Mobile Sensor Fusion Framework for Earthquake Hazard Reduction, Situational Assessment, and Relief Efforts

RAPID：用于减少地震灾害、态势评估和救灾工作的智能移动传感器融合框架

• 批准号: 1942053
• 负责人: Karen Panetta
• 金额: $ 9.85万
• 依托单位: Tufts University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2021-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1942053&HistoricalAwards=false
• 关键词: RAPID; Smart; Mobile; Sensor; Fusion

Natural disasters including earthquakes, tsunamis, hurricanes and anthropogenic disasters, such as wildfires, are dynamic situations requiring constant monitoring as numerous hazards are constantly emerging that hinder humanitarian efforts and create deadly conditions for rescue workers and victims. Most disaster area assessment, and search and rescue efforts, rely heavily on visual imagery captured from cell phones, aerial video and other forms of media. Currently, real-time assessment for disasters depends entirely on the human operator's ability to visually identify the subject of interest in the video, or images captured. Further, damaged buildings and roadways, debris, smoke, fire, and environmental conditions such as rain complicate the human observer's ability to monitor situations for detecting victims and hazards. It is essential that rescuers can assess hazardous conditions before risking their lives and be armed with smart technologies that enable them to respond to dynamic situations. The recent events in Southern California (earthquakes) and Louisiana (Hurricane Barry) provide perishable data that will enable maturation of the proposed research activities. This plus the integration within the Smart Mobile Disaster Data Collection unit for response provides further justification of the RAPID award mechanism. The researchers will create a Smart Mobile Disaster Data Collection and Assessment tool for detecting and mapping out situational hazards using intelligent data fused information collected from multiple sensor technologies. Furthermore, this project aims to create a first-of-its-kind public database of collected imagery and media from disasters to help researchers and agencies share information or to assist in developing best practices and collaborative strategies. This project will also enable better visualization of disparate sources of data, thus making it conducive for human operator to detect hazards and victims. This research will also enhance safety and security applications across a wide range of areas including firefighting, disaster relief, and search-and-rescue. The approach of seamlessly utilizing and visualizing sensor information in a situational map to guide responders' actions and enhance their efficiency reduces the risk factor for responders.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

地震、海啸、飓风等自然灾害和野火等人为灾害是需要持续监测的动态情况，因为不断出现的各种灾害阻碍了人道主义工作，并为救援人员和受害者创造了致命的条件。大多数灾区评估、搜索和救援工作严重依赖从手机、航拍视频和其他形式的媒体拍摄的视觉图像。目前，灾害的实时评估完全取决于人类操作员在视频或图像中视觉识别感兴趣主题的能力。此外，受损的建筑物和道路、碎片、烟雾、火灾以及下雨等环境条件使人类观察者监测情况以发现受害者和危险的能力复杂化。救援人员在冒生命危险之前能够评估危险状况，并配备智能技术，使他们能够对动态情况做出反应，这一点至关重要。最近发生在南加州（地震）和路易斯安那州（巴里飓风）的事件提供了易腐烂的数据，将使拟议的研究活动更加成熟。这一点加上智能移动灾害数据收集单位的响应整合，进一步证明了RAPID奖励机制的合理性。研究人员将创建一种智能移动灾害数据收集和评估工具，用于使用从多种传感器技术收集的智能数据融合信息来检测和绘制情景危害。此外，该项目旨在创建首个收集灾害图像和媒体的公共数据库，以帮助研究人员和机构共享信息或协助制定最佳做法和合作战略。该项目还将更好地实现不同数据源的可视化，从而有助于操作人员发现危险和受害者。这项研究还将加强在消防、救灾和搜救等广泛领域的安全和安保应用。在情景地图中无缝利用和可视化传感器信息来指导响应者的行动并提高其效率的方法降低了响应者的风险因素。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Comprehensive Underwater Object Tracking Benchmark Dataset and Underwater Image Enhancement With GAN

• DOI: 10.1109/joe.2021.3086907
• 发表时间: 2021-07-27
• 期刊: IEEE JOURNAL OF OCEANIC ENGINEERING
• 影响因子: 4.1
• 作者: Panetta，Karen;Kezebou，Landry;Agaian，Sos
• 通讯作者: Agaian，Sos

Face description using anisotropic gradient: thermal infrared to visible face recognition

使用各向异性梯度的面部描述：热红外到可见光面部识别

• DOI: 10.1117/12.2304898
• 发表时间: 2018
• 期刊: and Applications 2018
• 作者: Wan, Qianwen;Rao, Shishir Paramathma;Panetta, Karen;Agaian, Sos S.;Kaszowska, Aleksandra;Taylor, Holly;Voronina, V.
• 通讯作者: Voronina, V.

Quaternion based neural network for hyperspectral image classification

基于四元数的神经网络用于高光谱图像分类

• DOI: 10.1117/12.2558808
• 发表时间: 2020
• 期刊: and Applications 2020
• 作者: Rao, Shishir Paramathma;Panetta, Karen;Agaian, Sos S.
• 通讯作者: Agaian, Sos S.

Augmented reality-based vision-aid indoor navigation system in GPS denied environment

GPS 缺失环境下基于增强现实的视觉辅助室内导航系统

• DOI: 10.1117/12.2519224
• 发表时间: 2019
• 期刊: and Applications
• 作者: Rajeev, Srijith;Wan, Qianwen;Yau, Kenny;Panetta, Karen;Agaian, Sos S.
• 通讯作者: Agaian, Sos S.

Aerial Border Surveillance for Search and Rescue Missions Using Eye Tracking Techniques

• DOI: 10.1109/ths.2018.8574165
• 发表时间: 2018-10
• 期刊: 2018 IEEE International Symposium on Technologies for Homeland Security (HST)
• 作者: Qianwen Wan;Aleksandra Kaszowska;A. Samani;K. Panetta;H. Taylor;S. Agaian