Inspection behavior recognition of underground power distribution room based on improved two-stream CNN method
-
摘要: 井下配电室监控视频持续时间较长且行为类型复杂，传统双流卷积神经网络（CNN）法对此类行为识别效果较差。针对该问题，对双流CNN法进行改进，提出了一种基于改进双流法的井下配电室巡检行为识别方法。通过场景分析，将巡检行为分为站立检测、下蹲检测、走动、站立记录、坐下记录5种类型，并制作了巡检行为数据集IBDS5。将每个巡检行为视频等分为3个部分，分别对应巡检开始、巡检中和巡检结束；对3个部分视频分别随机采样，获取代表空间特征的RGB图像和代表运动特征的连续光流图像，并分别输入空间流网络和时间流网络进行特征提取；对2个网络的预测特征进行加权融合，获取巡检行为识别结果。实验结果表明，以ResNet152网络结构为基础，且权重比例为1∶2的空间流和时间流双流融合网络具有较高的识别准确度，Top-1准确度达到98.92%；本文方法在IBDS5数据集和公共数据集UCF101上的识别准确率均优于3D-CNN、传统双流CNN等现有方法。Abstract: The monitoring video of underground power distribution room has a long duration and complex behavior types, and the traditional two-stream convolutional neural network (CNN) has poor recognition effect on such behaviors. In view of the problem, the two-stream CNN method was improved, and a method of inspection behavior recognition of underground power distribution room based on improved two-stream CNN was proposed. Through scene analysis, the inspection behaviors are divided into five types: standing detection, squatting detection, walking, standing record, and sitting down record, and the inspection behavior dataset IBDS5 is produced. Each inspection behavior video is divided into three parts, corresponding to the start of inspection, middle inspection and end of inspection; RGB images representing spatial features and continuous optical flow images representing motion features are obtained by random sample from three parts of the video, and the images are input to spatial flow network and time flow network respectively for feature extraction; weighted fusion of predicted features of the two networks are performed to obtain inspection behavior recognition results. The experimental results show that the spatial-temporal and dual-stream fusion network based on ResNet152 network structure with a weight ratio of 1∶2 has high recognition accuracy, and Top-1 accuracy reaches 98.92%;the recognition accuracy of the proposed method on the IBDS5 dataset and the public dataset UCF101 are better than existing methods such as 3D-CNN and traditional two-stream CNN.
-
-
期刊类型引用(10)
1. 杨艺，杨艳磊，王田，王科平. 基于多重信息自注意力的综采工作面目标行为识别. 煤炭学报. 2025(02): 1425-1442 . 
百度学术
2. 马天，姜梅，杨嘉怡，张杰慧，丁旭涵. 基于多特征融合时差网络的带式输送机区域违规行为识别. 工矿自动化. 2024(07): 115-122 . 本站查看
3. 李占利，权锦成，靳红梅. 基于3D-Attention与多尺度的矿井人员行为识别算法. 国外电子测量技术. 2023(07): 95-104 . 百度学术
4. 潘德泰，李贵亮，何启远，祁鸣露，陈其超，吴川彬. 基于计算机视觉的电网输变配环节配电线路巡检系统. 电子设计工程. 2023(17): 85-89 . 百度学术
5. 程德强，寇旗旗，江鹤，徐飞翔，宋天舒，王晓艺，钱建生. 全矿井智能视频分析关键技术综述. 工矿自动化. 2023(11): 1-21 . 本站查看
6. 王宇，于春华，陈晓青，宋家威. 基于多模态特征融合的井下人员不安全行为识别. 工矿自动化. 2023(11): 138-144 . 本站查看
7. 杨春雨，张鑫. 煤矿机器人环境感知与路径规划关键技术. 煤炭学报. 2022(07): 2844-2872 . 百度学术
8. 饶天荣，潘涛，徐会军. 基于交叉注意力机制的煤矿井下不安全行为识别. 工矿自动化. 2022(10): 48-54 . 本站查看
9. 黄瀚，程小舟，云霄，周玉，孙彦景. 基于DA-GCN的煤矿人员行为识别方法. 工矿自动化. 2021(04): 62-66 . 本站查看
10. 党伟超，史云龙，白尚旺，高改梅，刘春霞. 基于条件变分自编码器的井下配电室巡检行为检测. 工矿自动化. 2021(12): 98-105 . 本站查看
其他类型引用(15)
计量
- 文章访问数: 67
- HTML全文浏览量: 16
- PDF下载量: 17
- 被引次数: 25
下载:
