群养猪侵略性行为的深度学习识别方法  

Recognition method for aggressive behavior of group pigs based on deep learning

作　　者：高云[1,2] 陈斌 廖慧敏 雷明刚 黎煊 李静[1] 罗俊杰[1] Gao Yun;Chen Bin;Liao Huimin;Lei Minggang;Li Xuan;Li Jing;Luo Junjie(College of Engineering,Huazhong Agricultural University,Wuhan 430070,China;Cooperative Innovation Center for Sustainable Pig Production,Wuhan 430070,China;College of Animal Science and Technology,College of Animal Medicine,Huazhong Agricultural University,Wuhan 430070,China)

机构地区：[1]华中农业大学工学院,武汉430070 [2]生猪健康养殖协同创新中心,武汉430070 [3]华中农业大学动物科技学院动物医学院,武汉430070

出　　处：《农业工程学报》2019年第23期192-200,共9页Transactions of the Chinese Society of Agricultural Engineering

基　　金：“十三五”国家重点研发计划项目(2016YFD0500506);中央高校自主创新基金(2662018JC003,2662018JC010,2662017JC028);现代农业技术体系(CARS-35)

摘　　要：为了解决因传统机器视觉和图像处理方法的局限性以及复杂的猪体姿态和猪舍环境导致对群养猪侵略性行为识别的有效性、准确率较低的问题,该文基于深度学习的方法,提出使用3D CONV的群养猪侵略性行为识别算法-3DConvNet。分3个批次采集18头9.6 kg左右的大白仔猪视频图像,选用第一批次中包含28 d内各个时段的撕咬、撞击、追逐、踩踏4大类,咬耳、咬尾、咬身、头撞头、头撞身、追逐以及踩踏7小类侵略性行为以及吃食、饮水、休息等非侵略性行为共计740段(27114帧)视频作为训练集和验证集,训练集和验证集比例为3:1。结果表明,3D ConvNet网络模型在训练集上的识别准确度达96.78%,在验证集上识别准确度达95.70%。该文算法模型对于不同训练集批次的猪只以及不良照明条件下依然能准确识别侵略性行为,算法模型泛化性能良好。与C3D模型进行对比,该文提出的网络模型准确率高出43.47个百分点,单帧图像处理时间为0.50 s,可满足实时检测的要求。研究结果可为猪场养殖环境中针对猪只侵略性行为检测提供参考。Pigs like to fight with each other to form a hierarchy relationship in groups.Aggressive behaviors,mostly fighting,are frequently found in intensive pig raising facilities.Strong aggressive behaviors can cause other pigs lack of food and water,growing slowly,wounds,sick and even dead in serious situation.This considerably reduces health and welfare of pigs and further decreases economic benefits of pig industries.Monitoring and recognizing aggressive behaviors among pig group is the first step to manage the aggressive behaviors in group pigs effectively.Traditional human recording method is time-consuming and labor-intensive.This method can’t be used 24 hours a day,7 days a week.Machine vision technique brings an automatic monitoring method to solve this problem.In this paper,we introduced a new method for aggressive behaviors monitoring based on deep learning.The experiments were held under controlled environments,which were achieved in an environment-controlled chamber designed previously.The details of the chamber were depicted in a published paper written by our research group.Nursery pigs were fed under three different concentration levels of NH3 gas,which were<3.80,15.18,37.95 mg/m^3,with a suitable temperature of around 27℃ and the comfortable humidity between 50%-70%.Each nursery group had six pigs and were weight around 9.6 kg.During each 28 days’experiment of three concentration levels of NH3,videos were taken from the top of the chamber.An end-to-end network,named 3D CONVNet,was proposed for aggressive behavior recognition of group pigs in this paper,which based on a C3D network and built with 3D convolution kernels.The network structure of the 3D CONVNet was improved in both width and depth dimensions.The number of main convolutional layers was increased to 19,extra batch normalization and dropout layers were added to deepen the network.Furthermore,the multi-scale feature fusion method was introduced to widen the network.This improvement had bettered the performance of the algorithm considerably

关 键 词：卷积神经网络 机器视觉 模型 行为识别 侵略性行为 深度学习 群养猪 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]