This project addresses USDA-NIFA Program Area Priority A1261 - Inter-Disciplinary Engagement in Animal Systems, with an emphasis on "Precision Animal Management". It aligns well with USDA Strategic Goals 2 and 3. The long-term goal of this project is to enhance broiler welfare and production efficiency through adoption of advanced Precision Livestock Farming (PLF) technologies, thus improving the sustainability and competitiveness of the U.S. poultry industry. This interdisciplinary and integrative project aims to engineer, apply, and demonstrate an affordable computer vision system (CVS) for real-time monitoring of broiler behavioral animal-based measures (ABMs) at commercial farms through a series of research and Extension activities. This new CVS will be used to collect objective broiler behavioral ABM data for evaluating the effects of management practices on these ABMs. The project will also develop unique regional and national PLF Extension programs to demonstrate the resultant CVS, educate stakeholders, and disseminate research outcomes from this project. The four specific project objectives include:Develop a CVS that employs advanced AI algorithms for real-time monitoring of broiler behavioral ABMs using existing video data [Research]. The proposed system will be developed by expanding the capability of our current CVS system that monitors real-time bird activity and distribution and estimates bird gait score. Light-weight deep learning and machine learning algorithms will be developed for automatic behavioral ABM identification using data collected in our previously funded projects. Focusing on welfare-related comfort behaviors (e.g. stretching, preening, and dustbathing) and production-related behaviors (e.g. eating and drinking), we will develop a benchmark database with high-quality behavior annotations.Identify the interactions of broiler behavioral ABMs with flock management factors, and collect baseline broiler behavioral ABMs [Research]. Using the newly developed CVS (Obj 1), we will investigate the effects of stocking density and light intensity on broiler behavioral ABMs in both lab-scale and farm-scale trials. We will also collect baseline broiler behavioral ABMs under typical management practices at commercial farms.Enhance CVS using new lab and field data [Research]. We will enhance the CVS using new data collected in the lab and farm trials (Obj 2) and feedback from poultry stakeholders (Obj 4).Establish a PLF module in regional and national poultry extension programs [Extension]. We will educate broiler producers on PLF technologies in use and under research, disseminate PLF research outcomes (Obj 2), evaluate challenges and needs forPLFtechnologies based on the feedback of the extension activities, and demonstrate the function and performance of the final CVS (Obj 3) to stakeholders.