SBIR Phase II: Real-time computer automated identification and quantification of insects entering the SolaRid insect control device (ICD)

这项小企业创新研究（SBIR）二期项目的更广泛影响包括一项人工智能技术，旨在检测、识别和确定田间虫害水平，实时提供全面的决策支持系统。更有效和精确的昆虫监测将通过提高所施用投入物的专一性和及时性，减少化学杀虫剂的使用。该项目的成功完成将有助于提高美国在世界农业市场上的经济竞争力，通过减少农药的使用对美国公众的健康和福利产生积极影响，并向农村人口介绍技术，突出科学、技术、工程和数学（STEM）教育投资的好处。这项技术可以通过减少杀虫剂的支出和减少害虫造成的损害而大大节省每亩土地。考虑到害虫每年造成450亿美元的作物损失，美国农场每年在农药上的花费超过250亿美元，这对该行业来说可能是巨大的节省。该项目提供了一个人工智能（AI）驱动的昆虫捕获系统，可以实时识别多种昆虫。该项目的主要目标是完成一种综合虫害管理工具的开发，该工具可以吸引、捕获和成像害虫，实时识别和计数害虫，并以用户友好的格式向可上网的设备提供数据和管理决策。为了实现这一目标，该技术将被部署在几个农业系统中，它们将以昆虫图像的形式持续获取数据。这些图像中的昆虫将由专家识别，数据将用于训练人工智能昆虫识别系统。在项目期间，该系统将学习识别农田和果园中重要和常见昆虫的多样性，并将开发一个用户友好的界面，以便向用户提供结果。最终目标是一个一体化的有害生物管理工具，可以在美国的任何农业系统中部署，它可以帮助农民管理有害生物问题，同时最大限度地减少农药的使用和提高产量。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 impacts of this Small Business Innovation Research (SBIR) Phase II project include an artificial intelligence technology designed to detect, identify, and determine levels of insect infestations in fields, providing a comprehensive decision support system in real-time. More efficient and precise insect monitoring would result in reduced chemical insecticide use by increasing the specificity and timeliness of the applied input. Successful completion of the project could serve to increase the economic competitiveness of the U.S. in the world agricultural market, positively impact the health and welfare of the American public through reduced pesticide use, and introduce rural populations to technology highlighting the benefits of investment in science, technology, engineering and math (STEM) education. This technology could result in significant savings per acre through decreased expenditures on pesticides and decreased damage done by pests. Considering pests cause $45 billion per year in crop damage annually, and US farms spend more than $25 billion per year on pesticides, the savings to the industry could be substantial.The project provides an artificial intelligence (AI)-driven insect trapping system that can identify a broad diversity of insects in real-time. The primary objective of the project is to finish development of an integrated pest management tool that attracts, captures, and images pest insects, identifies and counts them in real-time, and delivers data and management decisions in a user-friendly format to internet-accessible devices. In order to achieve this objective, the technology will be deployed in several agricultural systems where they will continuously obtain data in the form of insect images. The insects in these images will be identified by experts and the data will be used to train the AI insect identification system. During the project, the system will learn to identify a diversity of important and commonly encountered insects in agricultural fields and orchards and a user-friendly interface for delivering results to users will be developed. The end goal is an all-in-one pest management tool that can be deployed in any agricultural system in the United States where it can aid farmers in the management of pest problems while minimizing pesticide use and increasing yields.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.