基于无人机视频流的草原放牧家畜在线检测和体重估算

Real-time detection and weight estimation of grassland livestock based on unmanned aircraft system video streams

精确实时的家畜数据对发展现代畜牧业、保障畜产品有效供给和草原生态系统平衡,促进草原可持续发展至关重要。目前这些数据主要通过地面调查和基层上报方式获取,成本高、实时性差。本文在构建家畜深度学习识别模型和体重估算模型基础上,建立了基于浏览器/服务器(B/S)架构的家畜实时监控系统(http://218.202.104.82:5806/vid),利用无人机视频流,实现了家畜的在线识别、计数和体重估算。家畜识别模型训练使用了13803张无人机影像块和视频图像帧,牛的检出率为90.51%,错检率为11.64%,漏检率为9.49%,羊的检出率为91.47%,错检率为7.04%,漏检率为8.53%。体重估算模型构建采用了在青海和内蒙等地实测的头体长和牛体重数据,对牛和羊体重的估算精度分别为90.28%和90.00%。该系统将无人机和深度学习等技术应用于家畜监控领域,对禁牧、休牧等草原放牧家畜监管,以及帮助牧民远程监控家畜有重要意义。

Accurate and real-time livestock data are crucial to developing modern animal husbandry, ensuring effective supply of animal products, and promoting ecosystem balance and sustainable development of grasslands. The acquirement of livestock data mainly relies on field surveys and grassroots’ reports. These data are laborious and non-real-time. In this study, a real-104.82:5806/vid). A deep-learning-based livestock detection model and a weight estimation model are developed. The system could detect and count livestock, and estimate their weight using unmanned aircraft system( UAS) live video streams. The livestock detection model istrained using 13803 UAS image tiles and video picture frames. The true positive rate, false posi-tive rate, and loss positive rate of the model for cattle detection are 90.51%, 11.64%, and9.49%, respectively. The true positive rate, false positive rate, and loss positive rate of the mod-el for sheep detection are 91.47%, 7. 04%, and 8. 53%, respectively. The weight estimationmodel is built based on the head-body length and weight data collected in Inner Mongolia Autono-mous Region and Qinghai Province, with accuracy of 90.28% and 90.00% for cattle and sheepweight estimation, respectively. The system utilizes UASs and deep learning technologies for live-stock monitoring, having an expected application prospect in the fields of grassland supervision(including grazing prohibition and rest grazing), and assisting herdsmen in remotely monitoringtheir livestock.