摘要
Robotics&Machine Learning Daily News的新闻记者兼工作人员新闻编辑Friedl,Jona H(Kirkland,WA,US),Greschler,David(Kirkland,WA,US),Vogel,Jon(Kirkland,WA,US)于2024年1月4日申请专利，根据NewsRx C专利号12002008转让给NOMAD Go Inc.。（美国华盛顿州柯克兰）。新闻编辑从发明者提供的背景信息中获得了以下引文：“库存管理在很大程度上依赖于基于人类的计算和行动。一些已知的系统允许人类使用放置在仓库或零售空间的摄像机有效地管理他们大规模的库存。此外，”一些已知的系统要求严格的组织结构和每个物品的单一外观，而不区分共享同一库存单元(SKU)的物品的包装或物品的各种存储方式。此外，一些已知的系统使用具有计算机视觉的固定摄像机，并且可能需要限制在特定区域和具有多个盲点的视图的多个摄像机。这种已知的技术通常依赖于人工数据输入，这可能是乏味的，耗时的，并导致不准确的。因此，需要一个用于预测和实时库存管理的计算机视觉和机器学习系统。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-A patent by the inventors Friedl, Jona h (Kirkland, WA, US), Greschler, David (Kirkland, WA, US), Vogel, Jon (Kirkland, WA, US), filed on January 4, 2024, was published online on June 4, 2024, accord ing to news reporting originating from Alexandria, Virginia, by NewsRx correspon dents. Patent number 12002008 is assigned to NOMAD Go Inc. (Kirkland, Washington, Unite d States). The following quote was obtained by the news editors from the background informa tion supplied by the inventors: "The management of an inventory relies heavily o n human-based calculations and actions. Some known systems allow humans to effic iently manage their inventory of items at massive scales using cameras placed in warehouses or retail spaces. Furthermore, some known systems require strict org anization and single appearances per item without distinguishing the packaging o f items sharing the same stock keeping unit (SKU) or the various ways items are stored. Moreover, some known systems use fixed cameras with computer vision and can require multiple cameras that are limited to specific areas and views with m ultiple blind spots. Additionally, such known technologies often rely on manual data input, which can be tedious, time consuming, and lead to inaccuracies. "Accordingly, a need exists for a computer vision and machine learning system fo r predictive and real-time inventory management."
NETL
NSTL