SBIR Phase II: Apple Yield Mapping using Computer Vision

该项目的更广泛的影响/商业潜力是使用拟议的产量测绘自动化系统，以便种植者能够改进他们的种植和收获过程。产量制图对水果种植者来说至关重要。准确的估计对销售操作、收获时间物流和作物管理非常有益。目前，产量测绘是手工进行的，这是一个困难、费力的过程，容易出现抽样和计数错误。拟议中的系统将使种植者能够在使用更少资源的情况下，以更高的价格出售更好的水果。通过提高收获数量和时间的确定性，该系统还将提高整个水果供应链的效率，使新鲜水果更容易以更一致的价格在商店里买到。这个小企业创新研究（SBIR）二期项目将解决水果作物自动产量测绘的问题。该公司提出，与其依赖昂贵的传感设备，如基于激光雷达扫描仪，不如使用现成的商业组件，建立一个强大而廉价的水果测绘系统。为了实现这一目标，必须克服重大的计算机视觉和系统挑战。这包括：(1)适应和开发用于水果检测和分割的精确计算机视觉算法，以及用于水果大小和跟踪以及绘制叶子的几何算法；(2)建立可用、可靠的扫描、上传、云处理和结果可视化系统。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/commercial potential of this project is to use the proposed system for automation of yield mapping so that growers will be able to improve their growing and harvesting processes. Yield mapping is critical for fruit growers. An accurate estimate is enormously beneficial to sales operations, harvest time logistics, and crop management. Currently, yield mapping is performed manually in a difficult, laborious process prone to sampling and counting error. The proposed system would enable growers to sell better fruit at higher prices while using less resources. By bringing improved certainty to harvest quantity and timing, the system will also improve the efficiency of the entire fruit supply chain, making fresh fruit more readily available in stores at more consistent prices.This Small Business Innovation Research (SBIR) Phase II project will address the problem of automated yield mapping for fruit crops. Rather than relying on expensive sensing equipment such as laser-based lidar scanners, the company proposes to build a robust, yet inexpensive, fruit mapping system using commercial, off-the-shelf components. In order to achieve this goal, significant computer vision and systems challenges must be overcome. These include: (1) Adapting and developing accurate computer vision algorithms for fruit detection and segmentation, as well as geometric algorithms for sizing and tracking fruit and mapping the foliage; and (2) Building usable and reliable systems for scanning, upload, cloud processing, and results visualization.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.