An Advanced Automatic Remote Meter Detection Method Based on Machine Vision

Abstract:

There are a great amount of electronic meters equipped in the distribution substations, which were traditionally monitored by operators. On-site monitoring for risk assessment of these meters is very important. In this paper, we presented an advanced machine vision based automatic meter detection method toward the development of an online automatic meter reading intelligent inspection robot in substation. Firstly, the image received from the inspection robot was enhanced using histogram equalization. Then, the image was segmented into two parts based on the threshold obtained by Otsu’s method. Using these two parts, and the whole enhanced image, circular Hough transformation was applied on these three images and detected the circle with highest probability on them. The normalized correlation coefficients were calculated between the corresponding areas of those three circles from three images and the template image of SF₆ meter. Finally, the circle with highest correlation coefficient, which was higher than a certain threshold, was determined to be the meter. If it is lower than the threshold, the algorithm would decide that no meter was found in the image. The method was tested with 222 images obtained in one substation in Xi’an, Shanxi, China, and an 87.4% accuracy was achieved using these images, which indicated the potential of this method.

• Periodical: Applied Mechanics and Materials (Volumes 568-570)
• Pages: 994-1000
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.568-570.994
• Online since: June 2014
• Authors: Han Li, Shao Jun Liu*, Ku Wang, Xia Liu
• Keywords: Automatic Meter Reading, Circular Hough Transform, Machine Vision, Normalized Correlation Coefficients, Substation

References

[1] S. Lu, Q. Qian, and B. Zhang, et al. Development of a Mobile Robot for Substation Equipment Inspection. Automation of electric power systems, 2006, (13): 94-98.

[2] L. Zhou, Y. Zhang, and Y. Sun, et al. Development and Application of Equipment Inspection Robot for Smart Substations. Automation of electric power systems, 2011, (19): 85-88.

[3] S. Fu, W. Li, and Y. Zhang, et al. Structure-constrained obstacles recognition for power transmission line inspection robot[C]/Intelligent Robots and Systems, 2006 IEEE/RSJ International Conference on. IEEE, 2006: 3363-3368.

DOI: 10.1109/iros.2006.282513

[4] S. Zhao, B. Li, and J. Yuan, et al. Research on remote meter automatic reading based on computer vision[C]/Transmission and Distribution Conference and Exhibition: Asia and Pacific, 2005 IEEE/PES. IEEE, 2005: 1-4.

DOI: 10.1109/tdc.2005.1546972

[5] R. Kurte, C. Beyer, and H. Heise, et al. Application of infrared spectroscopy to monitoring gas insulated high-voltage equipment: electrode material-dependent SF6 decomposition [J]. Analytical and bioanalytical chemistry, 2002, 373(7): 639-646.

DOI: 10.1007/s00216-002-1280-4

[6] N. Otsu. A threshold selection method from gray-level histograms,. IEEE Trans. Sys., Man., Cyber. 1979, 9 (1): 62–66.

DOI: 10.1109/tsmc.1979.4310076

[7] R. O. Duda, and P. E. Hart. Use of the Hough Transform to detect lies and curves in pictures. Comm. ACM, 1972, 15(1), 11-15.

DOI: 10.1145/361237.361242

[8] J. P. Lewis,. Fast normalized cross-correlation. Vision interface. 1995, 10(1), 120-123.