基于深度学习的多台抽油机运动状态识别方法
陈 鹏 （ 北京石油机械有限公司 ）
https://doi.org/10.37155/2811-0595-0411-39Abstract
抽油机是油田生产的重要组成部分，亟需有效的信息化检测手段。现有运动状态检测算法主要通过对比 前后图像的相似度来判断图像中目标的运动状态变化情况。然而，抽油机的工作环境处于野外，除检测目标外，背景 物体也时常发生变化，导致上述传统算法识别精度低、误报严重。此外，实际工作场景中，一张图片中常常存在多台 处于不同运动状态的抽油机，传统运动状态检测算法无法解决此类多目标难题。针对上述挑战，我们提出了一种基于 深度学习的多台抽油机运动状态检测新方法。其创新点在于：突破了yolo算法只能识别目标无法判断目标运动状态的 局限；实现了野外条件下的多目标运动状态识别，大幅提升了野外工作条件下抽油机启停状态检测准确性。经油田现 场运行证明新方法准确率可达96%。
Keywords
抽油机；运动状态检测；野外复杂场景；多目标Full Text
PDFReferences
[1]管飞,王荣.基于Horn-Schunck光流法的运动目标检
测的研究[J].仪表技术,2015(2):3.
[2]Barnich Olivier and Van Droogenbroeck Marc. Vi
Be: a universal background subtraction algorithm for video
sequences.[J]. IEEE transactions on image processing : a
publication of the IEEE Signal Processing Society, 2011,
20(6):1709-24
[3]栾庆磊,赵为松.动背景下帧差分法与边缘信息融合
的目标检测算法[J].光电工程,2011,38(10):77-83.
[4] Xianfeng Wu and Siyi Wu. Improved Faster RCNN
Object Detection[J]. Advances in Social Sciences, 2016[5]Toseph Redmon, Santosh Divvala, Ross Girshick,et
a1.You Only Look Once: Unified, Real-Time Object
Detection[C].IEEE Conference on Computer Vision and
Pattern Recognition,2016: 779-788.
[6]He K, Zhang X, Ren S, et al. Spatial Pyramid Pooling
in Deep Convolutional Networks for Visual Recognition[C].
european conference on computer vision, 2014: 346-361.
[7]Wang, Zhou; Bovik, A.C.; Sheikh, H.R.; Simoncelli,
E.P.Image quality assessment: from error visibility to
structural similarity：IEEE Transactions on Image Processing
13 (4)，2004-04-01：600–612
[8]曹之君,张良.基于Faster-RCNN的快速目标检测算
法[J].航天控制,2020,38(4):7.
[9]薛俊韬,马若寒,胡超芳.基于 MobileNet 的多目标跟
踪深度学习算法[J].控制与决策,2021,36(8):1991-1996
测的研究[J].仪表技术,2015(2):3.
[2]Barnich Olivier and Van Droogenbroeck Marc. Vi
Be: a universal background subtraction algorithm for video
sequences.[J]. IEEE transactions on image processing : a
publication of the IEEE Signal Processing Society, 2011,
20(6):1709-24
[3]栾庆磊,赵为松.动背景下帧差分法与边缘信息融合
的目标检测算法[J].光电工程,2011,38(10):77-83.
[4] Xianfeng Wu and Siyi Wu. Improved Faster RCNN
Object Detection[J]. Advances in Social Sciences, 2016[5]Toseph Redmon, Santosh Divvala, Ross Girshick,et
a1.You Only Look Once: Unified, Real-Time Object
Detection[C].IEEE Conference on Computer Vision and
Pattern Recognition,2016: 779-788.
[6]He K, Zhang X, Ren S, et al. Spatial Pyramid Pooling
in Deep Convolutional Networks for Visual Recognition[C].
european conference on computer vision, 2014: 346-361.
[7]Wang, Zhou; Bovik, A.C.; Sheikh, H.R.; Simoncelli,
E.P.Image quality assessment: from error visibility to
structural similarity：IEEE Transactions on Image Processing
13 (4)，2004-04-01：600–612
[8]曹之君,张良.基于Faster-RCNN的快速目标检测算
法[J].航天控制,2020,38(4):7.
[9]薛俊韬,马若寒,胡超芳.基于 MobileNet 的多目标跟
踪深度学习算法[J].控制与决策,2021,36(8):1991-1996
Copyright © 2025 陈 鹏
Publishing time:2025-11-30
This work is licensed under a Creative Commons Attribution 4.0 International License
Email
Location