Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (10): 3191-3199.DOI: 10.11772/j.issn.1001-9081.2023101496

• Multimedia computing and computer simulation • Previous Articles     Next Articles

YOLOv7-MSBP target location algorithm for character recognition of power distribution cabinet

Cheng WANG1(), Yang WANG1, Yingjiao RONG2   

  1. 1.School of Internet of Things Engineering,Jiangnan University,Wuxi Jiangsu 214122,China
    2.Science and Technology on Near?surface Detection Laboratory,Wuxi Jiangsu 214035,China
  • Received:2023-11-06 Revised:2024-01-11 Accepted:2024-01-17 Online:2024-10-15 Published:2024-10-10
  • Contact: Cheng WANG
  • About author:WANG Yang, born in 1998, M. S. His research interests include image processing, deep learning
    RONG Yingjiao, born in 1978, M. S., engineer. Her research interests include signal processing, target detection.
  • Supported by:
    Fund of Key Laboratory of National Defense(6142414220203)

面向配电柜字符识别的YOLOv7-MSBP目标定位算法

王呈1(), 王炀1, 荣英佼2   

  1. 1.江南大学 物联网工程学院,江苏 无锡 214122
    2.近地面探测技术重点实验室,江苏 无锡 214035
  • 通讯作者: 王呈
  • 作者简介:王呈(1983—),男,江苏无锡人,副教授,博士,主要研究方向:非线性系统建模与控制、机器学习、数据挖掘 Wangc@jiangnan.edu.cn
    王炀(1998—),男,山东烟台人,硕士,主要研究方向:图像处理、深度学习
    荣英佼(1978—),女,江苏无锡人,工程师,硕士,主要研究方向:信号处理、目标检测。

Abstract:

Accurately locating the instrument position of power distribution cabinet through machine vision is the key to realize intelligent identification of instruments. Aiming at the problem of low target positioning accuracy caused by complex background of power distribution cabinet, various character scales and small camera pixels, a YOLOv7-MSBP target location algorithm for character recognition of power distribution cabinet was proposed. Firstly, a Micro-branch detection branch was designed and the initial anchor box laying interval was changed to improve the detection accuracy for small targets. Secondly, Bi-directional Feature Pyramid Network (BiFPN) was introduced to fuse the feature values of different layers across scales, thereby improving the situations of the loss of detailed features and insufficient feature fusion caused by downsampling. Meanwhile, Synchronous Convolutional Block Attention Module (Syn-CBAM) was designed, channel and spatial attention features were fused with weights, then the feature extraction ability of the algorithm was improved. And a Partial Convolution (PConv) module was introduced in the backbone network to reduce model redundancy and delay, and increase detection speed. Finally, the positioning results of YOLOv7-MSBP were sent to Paddle OCR (Optical Character Recognition) model for character recognition. Experimental results show that the mean Average Precision (mAP) of YOLOv7-MSBP algorithm reaches 93.2%, which is 4.3 percentage points higher than that of YOLOv7 algorithm. It can be seen that the proposed algorithm can locate and recognize the characters of the power distribution cabinet quickly and accurately, which verifies the effectiveness of the proposed algorithm.

Key words: YOLOv7 algorithm, instrument identification, attention mechanism, Bi-directional Feature Pyramid Network (BiFPN), machine vision

摘要:

通过机器视觉算法精确定位配电柜仪表的位置是实现仪表智能化识别的关键。针对配电柜背景复杂、字符尺度多样和相机像素低而导致的目标定位精度不高问题,提出一种面向配电柜字符识别的YOLOv7-MSBP目标定位算法。首先,设计Micro-branch检测分支,改进初始锚框铺设间隔,从而提高对小目标的检测精度。其次,引入双向特征金字塔网络(BiFPN)跨尺度融合不同层特征值,以改善因下采样造成的细节特征丢失、特征融合不充分的现象;同时,设计同步混合阈卷积注意力模块(Syn-CBAM),加权融合通道和空间注意力特征,以提升算法的特征提取能力;并且,在主干网络引入部分卷积(PConv)模块,以降低算法冗余和延迟,提高检测速度。最后,将YOLOv7-MSBP的定位结果送入Paddle OCR(Optical Character Recognition)模型识别字符。实验结果表明,YOLOv7-MSBP算法的平均精度均值(mAP)达到93.2%,与YOLOv7算法相比提高了4.3个百分点,可见所提算法能够快速准确定位识别配电柜字符,验证了所提算法的有效性。

关键词: YOLOv7算法, 仪表识别, 注意力机制, 双向特征金字塔, 机器视觉

CLC Number: