SBIR Phase I: CAS: Tree Root Quality Inspection System with Noninvasive Evaluation

这个小企业创新研究（SBIR）一期项目提高了苗木的质量、价值、效益和寿命，使它们能够在人们生活的环境中茁壮成长。树木对人类健康、生态系统服务和减缓气候变化有无数好处，当它们长期茁壮成长时，就会充分发挥其潜力。目前的方法不能充分解决树根质量问题。通过现代化的根检查，这项技术不仅可以提高行业标准，提高经济竞争力，还可以在全球范围内促进环境管理。有了这项技术，树艺师和种植者可能能够识别地上的树根缺陷，并采取纠正措施，以促进这些重要的树木资产生长到成熟所需的优质根系。SBIR一期项目的重点是开发一种三维、非破坏性、探地雷达（GPR）计算机断层扫描（CT）系统，该系统具有尖端的软件分析，用于检查和评估苗木容器生长根系的质量。这项创新是基于这样一种认识，即GPR信号是由活组织和周围土壤之间的巨大差异产生的。该技术可以检测到严重的根系缺陷，如果在植树前不加以纠正，可能会导致树木早期死亡。数据将在一种创新设备的帮助下收集，该设备设计用于无缝捕获集装箱种植树木的3D根数据，使用商用GPR系统，无线天线作为集装箱周围的第二层，模拟x射线CT扫描仪的精度。一种新的根系质量分类模型将为根系分析软件程序的开发提供信息，为机器学习模型构建初始GPR数据集，并随后采用主动学习方法。最初的实验将集中在数量足够大的一种或两种树木上。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。

This Small Business Innovation Research (SBIR) Phase I project improves the quality, value, benefits, and life span of nursery stock trees, so they will grow and thrive in the landscape where people live. Trees, with their myriad benefits for human health, ecosystem services, and climate mitigation achieve their full potential when they thrive long term. Current methods fail to adequately address tree root quality. By modernizing root inspection, this technology will not only improve industry standards and boost economic competitiveness, but also promote environmental stewardship on a global scale. With this technology, arborists and growers may be able to identify above-ground tree root defects and take corrective action to promote good quality root systems that are needed for these important tree assets to grow to maturity. This SBIR Phase I project focuses on the development of a 3-dimensional, non-destructive, ground penetrating radar (GPR) computed tomography (CT) system with cutting-edge software analytics to inspect and assess the quality of container-grown root systems in nursery stock trees. This innovation is based on the understanding that the GPR signals are generated by the large differentials between live tissues and the surrounding soil. The technology detects serious root system defects that could cause early tree mortality if not corrected before the tree is planted. The data will be collected with the help of an innovative apparatus designed to seamlessly capture 3D root data from container-grown trees using a commercial GPR system with a wireless antenna that works as a secondary layer around the container, emulating the precision of an X-ray CT scanner. A novel root quality classification model will inform the development of a root analysis software program to build an initial GPR dataset for the machine learning model and subsequently to adopt an active learning approach. Initial experiments will focus on a sufficiently large number of one or two species of trees.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.