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

这一小型企业创新研究(SBIR)第二阶段项目的更广泛影响包括一项旨在检测、识别和确定田间虫害水平的人工智能技术，提供实时综合决策支持系统。通过提高投入的特异性和及时性，更有效和更精确的昆虫监测将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的经济竞争力，通过减少杀虫剂的使用对美国公众的健康和福利产生积极影响，并向农村人口介绍突出投资于科学、技术、工程和数学(STEM)教育的好处的技术。这项技术可以通过减少杀虫剂支出和减少虫害造成的损害，大大节省每英亩土地的成本。考虑到害虫每年造成450亿美元的农作物损害，而美国农场每年在杀虫剂上的支出超过250亿美元，这可能会为该行业节省大量资金。该项目提供了一个人工智能(AI)驱动的昆虫诱捕系统，可以实时识别各种昆虫。该项目的主要目标是完成害虫综合管理工具的开发，该工具能够吸引、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访问互联网的设备提供数据和管理决策。为了实现这一目标，这项技术将部署在几个农业系统中，在那里他们将不断获得昆虫图像形式的数据。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。在该项目期间，该系统将学习识别农地和果园中各种重要和常见的昆虫，并将开发一个用户友好的界面，将结果传递给用户。最终目标是一种一体化的虫害管理工具，可以在美国的任何农业系统中部署，在帮助农民管理虫害问题的同时，将杀虫剂的使用降至最低并增加产量。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势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.