摘要
深度学习中卷积神经网络在行为识别领域有着良好的识别效果,但是由于深度学习需要较大数据集训练模型,而现今公开数据集中危险行为识别相关方向没有大量数据集。针对危险行为识别领域样本少、无法进行深度学习训练等问题,建立了危险行为识别数据集,并采用迁移学习方法对C3D网络模型进行迁移训练。结果表明,迁移学习后C3D网络模型对危险行为识别数据集平均识别率达到了83.2%,可以有效识别危险行为动作。
The convolutional neural network in deep learning performs excellently in the field of behavior recognition,and a large data set is necessary for deep learning to train the model.However,there is no large data set related with dangerous behavior recognition in the open data sets at the moment.Aiming at the problems such as lacking samples in the field of dangerous behavior recognition and being unable to carry out training in deep learning,the dangerous behavior recognition data set was established. The transfer learning method was adopted to carry out transfer training for the C3D network model.It turns out that,after transfer learning,the average recognition rate of the C3D network model on the dangerous behavior recognition data set is up to 83.2%,which means C3D model can effectively recognize dangerous behaviors.
作者
李辰政
张小俊
朱海涛
张明路
LI Chen-zheng;ZHANG Xiao-jun;ZHU Hai-tao;ZHANG Ming-lu(School of Mechanical Engineering,Hebei University of Technology,Tianjin 300130,China;Anguo Power Supply Branch of State Grid Hebei Electric Power Co.,LTD,Anguo 071200,China)
出处
《科学技术与工程》
北大核心
2019年第16期187-192,共6页
Science Technology and Engineering
基金
国家重点研发计划(2017YFC0806503)资助
关键词
危险行为识别
深度学习
迁移学习
卷积神经网络
identification of dangerous behaviors
deep learning
transfer Learning
convolutional neural network
