ADER (Automated Detection, Ejection & Recovery)

ADER（自动检测、弹出

• 批准号: 10020988
• 金额: $ 94.47万
• 依托单位: RECYCLEYE LTD
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=10020988
• 关键词: ADER; Automated; Detection; Ejection; Recovery

Recycleye presents Project ADER (Automated Detection, Ejection & Recovery), an AI-driven waste sorting solution that will put the country on track to meet the Plastics Pact target of 70% of plastics packaging effectively recycled, and spearhead the mission to increase recycled content across all plastic packaging by increasing the availability of high-quality and consistent recyclate.Recycleye has already developed a low-cost, AI-powered system replicating the power of human vision. It uses advanced machine learning algorithms to provide automatic, image-based detection of individual items in co-mingled waste streams, at a material and object level. It leverages a cutting-edge synthetic data generation pipeline, and Recycleye's own WasteNet - the world's largest visual database of labelled waste items, with over 2.5 million images.ADER will leverage AI in order to sort materials to a higher granularity, speed, and affordability than ever possible before. It will sort post-consumer waste into more categories (10+ vs 2-3) than existing optical sorting machines which rely on NIR alone, and so have limited scaling potential due to the sensor's high cost. Moreover, the integration of Recycleye's AI vision module enables the sorting of waste to currently impossible levels of purity and granularity, for example by distinguishing between food and non-food grade PP, PET trays and PET bottles, and can even detect less commonly collected classes, such as coloured plastics (jazz).

Ader项目介绍（自动检测、排出和回收），这是一个人工智能驱动的废物分类解决方案，将使该国走上正轨，以实现塑料公约的目标，即70%的塑料包装得到有效回收，并率先完成使命，通过增加高质量和一致的回收物的可用性，增加所有塑料包装的回收含量。人工智能驱动的系统复制人类视觉的力量。它使用先进的机器学习算法，在材料和物体层面上对混合废物流中的单个物品进行自动的、基于图像的检测。它利用了尖端的合成数据生成管道，以及Adecleye自己的Adecleye Net，这是世界上最大的带有标签的废弃物可视化数据库，拥有超过250万张图像。ADER将利用人工智能以比以往任何时候都更高的粒度，速度和可负担性来分类材料。它将把消费后的废物分类到更多的类别（10+ vs 2-3），而现有的光学分类机仅依赖于NIR，因此由于传感器的高成本，其扩展潜力有限。此外，集成了Spencleye的人工智能视觉模块，可以将废物分类到目前不可能达到的纯度和粒度水平，例如通过区分食品和非食品级PP，PET托盘和PET瓶，甚至可以检测不太常见的收集类别，例如彩色塑料（爵士乐）。