CISE-MSI:DP:Real-Time Aerial Imaging with Edge AI

CISE-MSI:DP：利用边缘 AI 进行实时航空成像

• 批准号: 2318546
• 负责人: Renny Fernandez
• 金额: $ 60万
• 依托单位: Norfolk State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-10-01 至 2026-09-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2318546&HistoricalAwards=false
• 关键词: CISE; MSI; DP; Real; Time

This project aims to develop an innovative aerial imaging system that incorporates state-of-the-art artificial intelligence (AI) for real-time data processing and analysis. This venture is a collaboration between students and faculty of the Computer Science and Engineering departments at Norfolk State University who will design, develop, and test a fully functional system capable of executing tasks such as autonomous and remote-controlled navigation, imagery, and autopilot. The significance of the project is embedded in its potential to revolutionize the field of aerial imaging and data analysis, with applications extending to agriculture, environmental monitoring, disaster response, and more. By integrating edge AI algorithms and software modules into the system, the team is set to achieve precise data processing and enhanced decision-making capabilities. Additionally, the project aspires to enhance diversity and representation in the field of engineering by providing opportunities for underrepresented minority communities and women.This project is designed to develop aerial machine vision by integrating dual-camera vision and time-of-flight technology into unmanned aerial vehicles. The amalgamation of these technologies will generate data-rich multispectral models, enabling the creation of large-scale maps for applications such as crop monitoring, yield assessment, and weed identification. The project also aims to construct collaborative systems of unmanned aerial vehicles, optimizing flight parameters and camera resolution for accurate three-dimensional reconstruction from aerial images. Furthermore, the project will facilitate edge intelligence for image processing and decision support, employing deep learning to extract information from specific segments of a hyperspectral model. This will allow for rapid and precise decision making, with applications including tree-structure and leaf-feature recognition from aerial videos. The project will also explore the implementation of complex algorithms on multicore central processing units for enhanced performance. Through this project, the team will provide students with a real-world understanding of the challenges and opportunities of unmanned aerial vehicles and how they integrate with technologies such as computer vision, machine learning, and communication protocols.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.

该项目旨在开发一种创新的航空成像系统，该系统结合了最先进的人工智能（AI）来进行实时数据处理和分析。该项目是诺福克州立大学计算机科学与工程系的学生和教师之间的合作，他们将设计、开发和测试一个功能齐全的系统，该系统能够执行自主和远程控制导航、图像和自动驾驶仪等任务。该项目的重要性在于其彻底改变航空成像和数据分析领域的潜力，其应用范围扩展到农业、环境监测、灾害应对等领域。通过将边缘人工智能算法和软件模块集成到系统中，团队将实现精确的数据处理和增强的决策能力。此外，该项目致力于通过为代表性不足的少数群体和女性提供机会来增强工程领域的多样性和代表性。该项目旨在通过将双摄像头视觉和飞行时间技术集成到无人机中来开发航空机器视觉车辆。这些技术的融合将生成数据丰富的多光谱模型，从而能够为作物监测、产量评估和杂草识别等应用创建大规模地图。该项目还旨在构建无人机协作系统，优化飞行参数和相机分辨率，以实现航空图像的精确三维重建。此外，该项目将促进图像处理和决策支持的边缘智能，利用深度学习从高光谱模型的特定部分提取信息。这将允许快速、精确的决策，其应用包括从航空视频中识别树结构和叶子特征。该项目还将探索在多核中央处理单元上实施复杂算法以增强性能。通过这个项目，该团队将帮助学生真实地了解无人机的挑战和机遇，以及它们如何与计算机视觉、机器学习和通信协议等技术相结合。该奖项反映了 NSF 的法定使命，并具有通过使用基金会的智力优点和更广泛的影响审查标准进行评估，被认为值得支持。