基于时空特征的奶牛视频行为识别  

Video Behavior Recognition of Dairy Cows Based onSpatio-temporal Features

作　　者：王克俭[1,2] 孙奕飞 司永胜[1,2] 韩宪忠[1,2] 何振学 WANG Kejian;SUN Yifei;SI Yongsheng;HAN Xianzhong;HE Zhenxue(College of Information Science and Technology,Hebei Agricultural University,Baoding 071001,China;Hebei Key Laboratory of Agricultural Big Data,Hebei Agricultural University,Baoding 071001,China)

机构地区：[1]河北农业大学信息科学与技术学院,保定071001 [2]河北农业大学河北省农业大数据重点实验室,保定071001

出　　处：《农业机械学报》2023年第5期261-267,358,共8页Transactions of the Chinese Society for Agricultural Machinery

基　　金：河北省重点研发计划项目(22327404D);河北农业大学精准畜牧学科群建设项目(1090064);河北省自然科学基金项目(F2020204003);国家自然科学基金项目(62102130)。

摘　　要：准确、高效的奶牛行为识别有助于疾病检测、发现异常,是感知奶牛健康的关键。通过分析奶牛在牛场中各时段的行为,提出一种基于时空特征的奶牛行为识别模型,该模型在时域段网络(TSN)的基础上融合了时态移位模块(TSM)、特征注意单元(FAU)和长短期记忆(LSTM)网络。首先,利用TSM融合时间信息以提高时序建模能力,并将时序建模后的视频帧输入TSN。其次,利用FAU融合高分辨率空间信息和低分辨率语义信息,增强模型空间特征的学习能力。最后,由LSTM聚合过去和当前信息进行奶牛行为分类。实验表明,该方法对进食、行走、躺卧、站立行为识别准确率分别为76.7%、90.0%、68.0%、96.0%,平均行为识别准确率为82.6%,和C3D、I3D、CNN-LSTM网络相比,本文模型平均行为识别准确率分别提升7.9、9.2、9.6个百分点。光照变化会对奶牛行为识别准确率产生一定影响,但本文模型受光照影响相对较小。研究成果可为感知奶牛健康和疾病预防提供技术支持。Accurate and efficient cow behavior recognition is helpful for timely disease detection and detection of abnormalities.It is the key to perceive cow health.By analyzing the behavior of cows at different periods in the cattle farm,a cow behavior recognition algorithm based on spatiotemporal features was proposed.The algorithm combined temporal shift module(TSM),feature attention unit(FAU)and long short-term memory(LSTM)networks on the basis of time-domain segment network(TSN).Firstly,TSM was used to fuse time information to improve timing modeling ability.The video frame after time sequence modeling was input to TSN.Secondly,FAU was used to integrate high resolution spatial information and low resolution semantic information to enhance the learning ability of spatial features of the algorithm.Finally,the past and current information were fused by LSTM to classify cow behavior.The results showed that the recognition accuracy of this algorithm for eating,walking,lying,and standing was 76.7%,90.0%,68.0%and 96.0%,respectively.And the average recognition accuracy was 82.6%.Compared with C3D,I3D and CNN-LSTM networks,the average recognition accuracy of this algorithm was 7.9 percentage points,9.2 percentage points and 9.6 percentage points higher,respectively.The illumination variation had a certain impact on the recognition accuracy,but the proposed algorithm was relatively less affected by light.The results can provide technical support for cow health perception and disease prevention.

关 键 词：奶牛 行为识别 时域段网络 时空特征 时态移位 长短期记忆 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]