论文部分内容阅读
针对目前山羊行为分类的不准确以及定位困难的问题,提出了一种基于ArcGIS和SOM神经网络算法的山羊定位和行为分类系统,系统由数据层、算法服务层、图形化展示层组成。设计了对山羊三轴加速度数据的SOM神经网络算法分类器,实现了对山羊静止、行走、采食、跳跃等主要日常行为的分类,并结合了ArcGIS二次开发平台在地图上显示山羊地理位置。实验结果表明:本文设计的系统运行安全、数据准确,对山羊科学饲养提供了依据。
In order to solve the problem of inaccurate classification and difficult localization of goats, a system of goat location and behavior classification based on ArcGIS and SOM neural network algorithm is proposed. The system consists of data layer, algorithm service layer and graphical display layer. The SOM neural network algorithm classifier for goats triaxial acceleration data was designed, and the classification of main daily behaviors such as still, walking, feeding and jumping of goats was realized. Combined with the ArcGIS secondary development platform, the geographical location of goats . The experimental results show that the system designed in this paper is safe and data-accurate, which provides a scientific basis for goat breeding.