摘要
为了解决当前着装规范性技术存在的一些问题,减少电力作业安全事故发生概率,提出了基于视频图像的着装规范性识别方法。首先采集电力施工监理视频图像,并对视频图像进行预处理,然后从电力施工监理视频图像提取HOG特征和LBP特征,利用支持向量机算法识别作业人员的着装,最后进行了仿真实验。仿真结果表明,所提方法可以实现高精度的着装识别,实现了工作人员规范穿戴问题的实时跟踪、报警,具有较高的实际应用价值。
In order to solve some problems existing in the current dress code technology and reduce the probability of power operation safety accidents,a method of clothing standardization recognition based on video image is proposed.Firstly,the video image of power construction supervision is collected,and the video image is preprocessed.Then,hog feature and LBP feature are extracted from the video image of power construction supervision.Then,the support vector machine algorithm is used to identify the operator's clothing.Finally,the simulation experiment is carried out.The results show that the proposed method can realize high-precision dress recognition and realize the real-time follow-up of staff's standard wearing tracking and alarming a high practical application value.
作者
施蔚青
刘洪兵
何四平
Shi Weiqing;Liu Hongbing;He Siping(Training and Evaluation Center,Yunnan Power Grid Co.,Ltd.,Yunnan Kunming,650221,China)
出处
《机械设计与制造工程》
2022年第5期121-124,共4页
Machine Design and Manufacturing Engineering
关键词
着装识别
图像识别
方向梯度直方图特征
局部二值模式特征
支持向量机
dress recognition
image recognition
histogram of oriented gradient feature
local binary pattern feature
support vector machine
