SBIR Phase I: A SMART CAMERA SYSTEM FOR WATER RISK ASSESSMENT

SBIR一期项目的更广泛影响是，通过向水管理利益相关者提供改进的当前和未来信息，减少人类生命和财产的损失。到本世纪中叶，海平面上升和更强的暴雨可能会使沿海社区的洪水损失每年增加230亿美元。传统的水监测传感器有主要的缺点，例如：经常维护、昂贵的安装和特定于应用程序的专用硬件。持续监测洪水易发地区和优化淡水保护用水管理需要新技术，包括低成本和强大的水控制系统，以限制人类生命、作物、财产和牲畜的损失。该项目提出的系统将提供一个强大的解决方案来监测水，并使用深度学习算法进行天气预测。该SBIR项目建议开发一种具有成本效益的基于智能摄像头的解决方案，以对公共水道的洪水进行高度准确的风险评估。一个关键的创新是在资源有限的平台上运行复杂的计算机视觉算法。该系统将使用通常用于测量固体物体速度的低成本激光雷达来测量水的速度。此外，该项目将开发一个偏振光系统，以达到系统精度要求，包括实时校正由于现场条件导致的相机视觉障碍引起的误差。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。

The broader impact of this SBIR Phase I project is to reduce the loss of human life and property by providing improved current and future information to water management stakeholders. Rising sea levels and heavier downpours could increase flooding costs in coastal communities by $23 billion per year by mid-century. Traditional water monitoring sensors have major drawbacks, such as: constant maintenance, costly installation, and application-specific dedicated hardware. Continuous monitoring of flood prone areas and optimized water management for freshwater conservation requires new technologies, including low-cost and robust water control systems to limit the loss of human lives, crops, property, and livestock. The proposed system in this project will offer a robust solution to monitor water and make weather predictions using deep learning algorithms. This SBIR project proposes to develop a cost-effective smart camera-based solution to perform a highly accurate risk assessment of flooding in public waterways.A key innovation is running sophisticated computer vision algorithms on a resource-limited platform. The proposed system will use low-cost LIDARs, generally used for measuring solid object velocities, for water velocity measurements. Furthermore, the project will develop a polarized-light system to achieve the system accuracy requirements, including real-time corrections of errors induced by visual impairments of the camera due to field conditions.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.