王雷雄, 王波, 马富齐, 董旭柱, 姚良忠. 基于双目立体匹配和场景元素识别的变电人员近电安全距离检测方法研究[J]. 电网技术, 2023, 47(3): 1010-1021. DOI: 10.13335/j.1000-3673.pst.2022.1763
引用本文: 王雷雄, 王波, 马富齐, 董旭柱, 姚良忠. 基于双目立体匹配和场景元素识别的变电人员近电安全距离检测方法研究[J]. 电网技术, 2023, 47(3): 1010-1021. DOI: 10.13335/j.1000-3673.pst.2022.1763
WANG Leixiong, WANG Bo, MA Fuqi, DONG Xuzhu, YAO Liangzhong. Research on the Detection Method of the Approach Distance Between Substation Workers and the Live Equipment Based on Binocular Stereo Matching and Scene Element Recognition[J]. Power System Technology, 2023, 47(3): 1010-1021. DOI: 10.13335/j.1000-3673.pst.2022.1763
Citation: WANG Leixiong, WANG Bo, MA Fuqi, DONG Xuzhu, YAO Liangzhong. Research on the Detection Method of the Approach Distance Between Substation Workers and the Live Equipment Based on Binocular Stereo Matching and Scene Element Recognition[J]. Power System Technology, 2023, 47(3): 1010-1021. DOI: 10.13335/j.1000-3673.pst.2022.1763

基于双目立体匹配和场景元素识别的变电人员近电安全距离检测方法研究

Research on the Detection Method of the Approach Distance Between Substation Workers and the Live Equipment Based on Binocular Stereo Matching and Scene Element Recognition

  • 摘要: 考虑到变电站带电设备周边环境中存在的强电场、暂态电击、稳态电击以及静电感应等安全影响因素,近电安全距离作为风险防护的重要指标对保障电力设备安全稳定运行及作业人员生命安全至关重要。为此,该文提出了一种基于双目立体匹配和场景元素识别的变电人员近电安全距离检测方法。该方法首先通过基于通道注意力模块的多阶段双目立体匹配模型获取作业场景空间信息,其次通过场景元素高效识别模型区分人与带电设备,然后基于空间信息和识别结果计算元素空间表面中心,最后基于空间表面中心计算人员近电安全距离,实现人员与带电设备间实时高精度距离检测。实际数据表明,该文方法实现了对现场作业人员与带电设备近电安全距离的实时准确测量。

     

    Abstract: Considering the Security influencers such as strong electric field, transient electric shock, steady-state electric shock and electrostatic induction in the live device environment of substation, minimum approach distance as an important indicator of risk protection is crucial to ensure the safe and stable operation of electrical equipment and the safety of personnel life. Therefore, this paper proposes a method for detecting the approach distance between the substation workers and the live equipment based on binocular stereo matching and scene element recognition. This method first obtains the spatial information of the operation scene by the multi-stage binocular stereoscopic matching model based on the channel attention module, and then distinguishes between people and live equipment through the efficient identification model of the substation scene elements, and then calculates the space surface center of the element based on the spatial information and recognition results, and finally calculates the approach distance of the personnel based on the space surface center, so as to realize the real-time high-precision distance detection between the personnel and the live equipment... The actual data show that the proposed method realizes the real-time accurate measurement of approach distance between workers and the live equipment.

     

/

返回文章
返回