I-Corps: Artificial Intelligence-Driven Disaster Risk Prediction and Assessment

这个I-Corps项目更广泛的影响/商业潜力是开发一种快速和自动化的场景理解技术，以智能地评估现有的脆弱性条件，并更积极有效地评估潜在的灾害风险。 这项技术可能有助于就减少灾害影响的步骤作出明智的决定。开发和实施减少灾害风险的技术需要有效地识别和评估潜在风险的来源，但缺乏快速和自动化的工具使从业人员依赖于人工检查。拟议的创新是通过减少耗时耗力的人工检查，使目前的灾害风险评估过程更加智能和高效。如果该项目成功完成，拟议的创新将提高科学和技术对灾害风险预测和评估的快速和自动视觉传感和分析的理解。该I-Corps项目基于开发低级别图像特征与高级语义模型的综合分析，从而实现复杂环境中的强大场景理解和风险预测。通过利用来自多模态视觉传感器的大规模数据（例如，无人机、安全摄像头等），该技术将潜在灾害风险的背景自动编码到机器视觉算法中，以确定视频记录中的风险因素，并评估复杂环境中的脆弱程度和风险因素。该系统以自动化的方式生成特定于地点的管理信息，这可以增强减灾和备灾的风险决策，从而有效降低风险水平和灾害影响。该奖项反映了NSF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。

The broader impact/commercial potential of this I-Corps project is the development of a rapid and automated scene understanding technology to intelligently evaluate existing conditions of vulnerability and assess potential disaster risks more proactively and effectively. This technology may help make informed decisions regarding steps to reduce the impacts of disasters. Developing and implementing the technology to reduce disaster risk requires identifying and assessing the sources of potential risk effectively, but a lack of a rapid and automated tools makes practitioners rely on manual inspection. The proposed innovation is to make the current disaster risk assessment process more intelligent and efficient by reducing time-consuming and labor-intensive manual inspection. If the project is completed successfully, the proposed innovation will enhance scientific and technological understanding of rapid and automated visual sensing and analytics for disaster risk prediction and assessment.This I-Corps project is based on the development of the integrated analysis of low-level image features together with high-level semantic models, which enables robust scene understanding and risk prediction in complex environments. By leveraging large-scale data from multimodal visual sensors (e.g., drones, security cameras, etc.), the technology automatically encodes the context of potential disaster risk into machine vision algorithms to identify the elements at risk on video recordings and assess the degree of vulnerability and the elements at risk in complex environments. The system generates site-specific managerial information in an automated manner, which may enhance risk-informed decision-making for disaster mitigation and preparedness, thereby reducing both the level of risk and the impacts of disasters effectively. The proposed innovation may also provide the foundation for further vision-based reasoning for disaster management.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.