Research on a Visual Reading Method for Pointer Instruments Integrating Geometric Constraints
DOI:
https://doi.org/10.54097/4713t075Keywords:
Pointer Meter, Machine Vision, Geometric Constraint, Hough Circle Transform, Automatic ReadingAbstract
A pointer meter automatic reading algorithm integrating geometric constraints is proposed. First, the dial region is located by the Hough circle transform, and the dial centre is taken as the pointer rotation centre. Then, segmentation thresholds are selected adaptively according to the pointer colour, pointer pixels are extracted within a restricted annulus, and the pointer direction is determined by angular histogram voting. Next, the recognition result is screened for geometric consistency using the rotation centre, annulus and confidence constraints, and a rejection state is output when the evidence is insufficient. Finally, the meter reading is calculated from the angular proportion of the pointer between the start and end scale marks. A 0 to 6 bar pressure gauge in a public real-image dataset is taken as the object, and the accepted readings are established by two independent reading rounds and a third-round arbitration. The experimental results show that the mean absolute error of 10 test samples is 0.375 bar with a fiducial error of 6.24 percent; for the 8 samples whose pointer direction is correctly judged, the mean absolute error is 0.100 bar with a fiducial error of 1.66 percent; the recognition confidence of yellow pointers is clearly higher than that of dark and light pointers, so the rejection threshold should be set separately for each pointer colour.
Downloads
References
[1] Li, J., Yuan, L., & Ran, T. (2021). Automatic detection and reading method for pointer meters based on YOLOv4. Mechanical & Electrical Engineering, 38(7), 912–917. (in Chinese)
[2] Dong, Z., Gao, Y., Yan, Y., et al. (2021). Vector Detection Network: An Application Study on Robots Reading Analog Meters in the Wild. IEEE Transactions on Artificial Intelligence, 2(5), 394–403. https://doi.org/10.1109/TAI.2021.3107050. DOI: https://doi.org/10.1109/TAI.2021.3105936
[3] Sun, X., Hou, L., & Qu, H. (2021). Progressive automatic reading method for pointer meters based on convolutional neural networks. Chinese Journal of Sensors and Actuators, 34(10), 1326–1333. (in Chinese)
[4] Howells, B., Charles, J., & Cipolla, R. (2021). Real-Time Analogue Gauge Transcription on Mobile Phone. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (pp. 2369–2377). https://doi.org/10.1109/CVPRW53098.2021.00274. DOI: https://doi.org/10.1109/CVPRW53098.2021.00269
[5] Zhu, B., & Fan, S. (2021). Recognition method for pointer meters in substations under rainy and foggy conditions. Laser & Optoelectronics Progress, 58(24), 2410008. (in Chinese) DOI: https://doi.org/10.3788/LOP202158.2410008
[6] Liu, M., Wang, B., Luo, P., et al. (2024). Intelligent recognition method for pointer-meter readings in low-quality outdoor images. High Voltage Engineering, 50(11), 5034–5046. (in Chinese)
[7] Zhang, S., Wan, J., Wang, H., et al. (2022). Image-reading recognition method for pointer meters based on an attention-enabled convolutional neural network. Electric Power Automation Equipment, 42(4), 218–224. (in Chinese)
[8] Deng, G., Huang, T., Lin, B., et al. (2022). Automatic Meter Reading from UAV Inspection Photos in the Substation by Combining YOLOv5s and DeeplabV3+. Sensors, 22(18), 7090. https://doi.org/10.3390/s22187090. DOI: https://doi.org/10.3390/s22187090
[9] Xu, X., Chen, X., & Mao, Y. (2023). Pointer-meter reading technology based on Deming regression and ROI refinement. Journal of Jiangsu University (Natural Science Edition), 44(3), 324–329. (in Chinese)
[10] Leon-Alcazar, J., Alnumay, Y., Zheng, C., et al. (2024). Learning to Read Analog Gauges from Synthetic Data. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (pp. 8616–8625). https://doi.org/10.1109/WACV57701.2024.00854. DOI: https://doi.org/10.1109/WACV57701.2024.00842
[11] Jin, A., Yuan, L., Zhou, D., et al. (2022). Pointer-meter reading recognition method based on YOLOv5 and U-Net. Instrument Technique and Sensor, (11), 29–33. (in Chinese)
[12] Zou, L., Wang, K., Wang, X., et al. (2023). Automatic Recognition Reading Method of Pointer Meter Based on YOLOv5-MR Model. Sensors, 23(14), 6644. https://doi.org/10.3390/s23146644. DOI: https://doi.org/10.3390/s23146644
[13] Huang, S., Fan, S., & Wang, Z. (2022). Pointer-meter reading recognition method for substations based on CenterNet and DeepLabv3+. Journal of Electric Power, 37(3), 232–243. (in Chinese)
[14] Zhao, W., Chen, H., Guo, L., et al. (2023). Adaptive recognition method for substation meter readings based on YOLO-E and improved OCRNet image segmentation. Electric Power Construction, 44(11), 75–85. (in Chinese)
[15] Tan, L., Wu, W., Ding, J., et al. (2024). A Robust Pointer Meter Reading Recognition Method Based on TransUNet and Perspective Transformation Correction. Electronics, 13(13), 2436. https://doi.org/10.3390/electronics13132436. DOI: https://doi.org/10.3390/electronics13132436
[16] Liu, Y., Li, Y., Zhang, T., et al. (2023). Pointer-meter detection and reading recognition based on improved YOLOv5s. Journal of Beijing Information Science and Technology University (Natural Science Edition), 38(3), 9–14. (in Chinese)
[17] Chen, C., Wang, Q., & Yao, W. (2022). Automatic reading algorithm for pointer meters based on machine vision. Transducer and Microsystem Technologies, 41(10), 130–133. (in Chinese)
[18] Tao, J., Lin, W., Zeng, L., et al. (2023). Pointer-meter reading recognition based on YOLOv4-tiny and Hourglass. Journal of Electronic Measurement and Instrumentation, 37(5), 1–10. (in Chinese)
[19] Gong, Q., Bie, B., Fan, X., et al. (2023). Pointer-meter reading algorithm based on keypoint detection. Journal of Electronic Measurement and Instrumentation, 37(3), 66–73. (in Chinese)
[20] Liu, Y., Tusongjiang Kari, Yilihamu Yarmaimaiti, et al. (2024). Fully automatic pointer-meter reading method based on VCA-UNet. Instrument Technique and Sensor, (2), 36–43. (in Chinese)
[21] Wang, J., Zhang, S., Wang, X., et al. (2025). Meter-reading recognition based on improved U2Net. Microelectronics & Computer, 42(7), 93–102. (in Chinese)
[22] Hou, L., Wang, S., Sun, X., et al. (2023). A pointer meter reading recognition method based on YOLOX and semantic segmentation technology. Measurement, 218, 113241. https://doi.org/10.1016/j.measurement.2023.113241. DOI: https://doi.org/10.1016/j.measurement.2023.113241
[23] Wang, J., Zeng, J., Kang, B., et al. (2025). A Keypoint-Driven Robust Pointer Meter Reading Method With Enhanced Structure for Complex Environments. IEEE Transactions on Instrumentation and Measurement, 74, 1–16. https://doi.org/10.1109/TIM.2025.3464414. DOI: https://doi.org/10.1109/TIM.2025.3552871
[24] Yan, F., Xu, W., Huang, Q., et al. (2022). Fully automatic recognition of pointer-meter readings based on a lightweight semantic-segmentation model. Laser & Optoelectronics Progress, 59(24), 2410001. (in Chinese) DOI: https://doi.org/10.3788/LOP202259.2410001
[25] Qu, H., & Zhang, W. (2023). Pointer-meter detection and recognition based on YOLOv7+U2-Net. Computer Systems & Applications, 32(11), 276–285. (in Chinese)
[26] Zuo, Z., Li, Y., Chen, L., et al. (2024). Automatic pointer-meter reading method based on machine vision. Instrument Technique and Sensor, (6), 37–40. (in Chinese)
[27] Su, Y., Si, H., Tang, X., et al. (2025). Intelligent reading method for pointer meters based on lightweight YOLOv5s and U²-Net. Journal of Jishou University (Natural Sciences Edition), 46(4), 44–51. (in Chinese)
[28] Lu, X., Zhu, S., & Lu, B. (2024). Pointer Meter Reading Method Based on YOLOv8 and Improved LinkNet. Sensors, 24(16), 5288. https://doi.org/10.3390/s24165288. DOI: https://doi.org/10.3390/s24165288
[29] Lin, K., & Han, F. (2025). Automatic recognition of aircraft-instrument information based on image processing and computer vision. Automation & Instrumentation, (8), 78–82. (in Chinese)
[30] Liu, Q., & Shi, L. (2025). A pointer meter reading method based on human-like reading sequence and keypoint detection. Measurement, 248, 116994. https://doi.org/10.1016/j.measurement.2025.116994. DOI: https://doi.org/10.1016/j.measurement.2025.116994
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Frontiers in Computing and Intelligent Systems
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
