Low-Cost Smart Metering Using Deep Learning

Farhan Khan; Sarmad Rafique; Gul Muhammad Khan

Authors

Farhan Khan Department of Electrical Engineering University of Engineering and Technology Peshawar, Pakistan
Sarmad Rafique Department of Computer Systems Engineering University of Engineering and Technology Peshawar, Pakistan
Gul Muhammad Khan Department of Electrical Engineering University of Engineering and Technology Peshawar, Pakistan

Keywords:

Yolo-v8, Paddle OCR, Meter Detection, Automatic Meter Recognition, Low-cost Smart Metering.

Abstract

Utility services like electricity, water, and gas are essential for modern living, and their demand has been rising worldwide. However, traditional manual meter reading is a standard procedure for billing purposes. This is not only labor and time-intensive but also prone to mistakes, which results in incorrect billing and revenue losses. In the era of advanced AI, leveraging cutting-edge technology to automate meter readings has become increasingly viable. However, Existing AI-based meter reading systems have limitations in detecting and recognizing meters from a distance. This research addresses these problems by presenting a novel system that utilizes the YOLOv8 model to detect meter screens from a distance. In addition, the system uses a fine-tuned Paddle OCR to recognize meter readings. A Novel dataset curated for the meter screen detection, recognition, and end-to-end OCR tasks related to electricity, gas, and water utility meters has been presented, containing up to 8,044 images. The proposed system was trained and extensively tested on the proposed dataset to gauge its performance. The system achieved an exceptional mean Average Precision (mAP) of 0.995 for both analog and digital meters on the detection task; furthermore, the system achieved an accuracy of 96.92% in the recognition task, which is 70% better than the accuracy of Pre-trained Paddle OCR. Moreover, an all-encompassing evaluation that combines detection and recognition using Paddle OCR and YOLOv8, i.e., the end-to-end OCR task, achieved an accuracy of 97.8%. Lastly, the system achieved an inference speed of up to 6 frames per second, guaranteeing real-time effectiveness.

References

“India’s power sector. (2012, August 9). World Bank”, [Online]. Available: https://www.worldbank.org/en/news/feature/2010/04/19/india-powersector

T. B. Smith, “Electricity theft: a comparative analysis,” Energy Policy, vol. 32, no. 18, pp. 2067–2076, Dec. 2004, doi: 10.1016/S0301-4215(03)00182-4.

“annual report - Tenaga Nasional Berhad.” Accessed: May 06, 2024. [Online]. Available: https://www.yumpu.com/en/document/view/51051742/annual-report-tenaga-nasional-berhad

M. Waqar, M. A. Waris, E. Rashid, N. Nida, S. Nawaz, and M. H. Yousaf, “Meter Digit Recognition Via Faster R-CNN,” 2019 Int. Conf. Robot. Autom. Ind. ICRAI 2019, Oct. 2019, doi: 10.1109/ICRAI47710.2019.8967357.

A. Cooper, “Electric Company Smart Meter Deployments: Foundation for A Smart Grid,” 2021.

C. M. Tsai, T. D. Shou, S. C. Chen, and J. W. Hsieh, “Use SSD to Detect the Digital Region in Electricity Meter,” Proc. - Int. Conf. Mach. Learn. Cybern., vol. 2019-July, Jul. 2019, doi: 10.1109/ICMLC48188.2019.8949195.

Li. Chunshan, Yukun Su, Rui Yuan, Dianhui Chu, and Jinhui Zhu. “Light-Weight Spliced Convolution Network-Based Automatic Water Meter Reading in Smart City”, [Online]. Available: https://ieeexplore.ieee.org/document/8917620

R. Laroca, Victor Barroso, Matheus A. Diniz, Gabriel R. Gonçalves, William R. Schwartz, and David Menotti. “Convolutional neural networks for automatic meter reading”, [Online]. Available: https://www.spiedigitallibrary.org/journals/journal-of-electronic-imaging/volume-28/issue-1/013023/Convolutional-neural-networks-for-automatic-meter-reading/10.1117/1.JEI.28.1.013023.short

C. Zhang and Y. Lu, “Study on artificial intelligence: The state of the art and future prospects,” J. Ind. Inf. Integr., vol. 23, p. 100224, Sep. 2021, doi: 10.1016/J.JII.2021.100224.

Z.-Q. Zhao, P. Zheng, S.-T. Xu, and X. Wu, "Object Detection With Deep Learning: A Review," in IEEE Transactions on Neural Networks and Learning Systems, vol. 30, no. 11, pp. 3212-3232, Nov. 2019, doi: 10.1109/TNNLS.2018.2876865.

P. Jiang, D. Ergu, F. Liu, Y. Cai, and B. Ma, “A Review of Yolo Algorithm Developments,” Procedia Comput. Sci., vol. 199, pp. 1066–1073, Jan. 2022, doi: 10.1016/J.PROCS.2022.01.135.

A. C. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., Berg, “SSD: Single Shot MultiBox Detector”, [Online]. Available: https://link.springer.com/chapter/10.1007/978-3-319-46448-0_2

R. Girshick, “Fast R-CNN,” Proc. IEEE Int. Conf. Comput. Vis., vol. 2015 Inter, pp. 1440–1448, 2015, doi: 10.1109/ICCV.2015.169.

“Repository of GitHub - JaidedAI/EasyOCR: Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin, Chinese, Arabic, Devanagari, Cyrillic and etc in May 2024.” Accessed: May 06, 2024. [Online]. Available: https://gitpiper.com/resources/python/computervision/JaidedAI-EasyOCR

“keras-ocr — keras_ocr documentation.” Accessed: May 06, 2024. [Online]. Available: https://keras-ocr.readthedocs.io/en/latest/

“GitHub - tesseract-ocr/tesseract: Tesseract Open Source OCR Engine (main repository).” Accessed: May 06, 2024. [Online]. Available: https://github.com/tesseract-ocr/tesseract

“GitHub - ultralytics/ultralytics: NEW - YOLOv8 ???? in PyTorch > ONNX > OpenVINO > CoreML > TFLite.” Accessed: May 06, 2024. [Online]. Available: https://github.com/ultralytics/ultralytics

“GitHub - PaddlePaddle/PaddleOCR: Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices).” Accessed: May 06, 2024. [Online]. Available: https://github.com/PaddlePaddle/PaddleOCR

A. Azeem, W. Riaz, A. Siddique, and U. A. K. Saifullah, “A Robust Automatic Meter Reading System based on Mask-RCNN,” Proc. 2020 IEEE Int. Conf. Adv. Electr. Eng. Comput. Appl. AEECA 2020, pp. 209–213, Aug. 2020, doi: 10.1109/AEECA49918.2020.9213531.

R. Laroca, A. B. Araujo, L. A. Zanlorensi, E. C. de Almeida, and D. Menotti, “Towards Image-based Automatic Meter Reading in Unconstrained Scenarios: A Robust and Efficient Approach,” IEEE Access, vol. 9, pp. 67569–67584, Sep. 2020, doi: 10.1109/ACCESS.2021.3077415.

S. Zhuo, Xiaoming Zhang, Ziyi Chen, Wei Wei, Fang Wang, Q. Li and Y. Guan, “DAMP-YOLO: A Lightweight Network Based on Deformable Features and Aggregation for Meter Reading Recognition”, [Online]. Available: https://www.mdpi.com/2076-3417/13/20/11493

W. Lin, Z. Zhao, J. Tao, C. Lian, and C. Zhang, “Research on Digital Meter Reading Method of Inspection Robot Based on Deep Learning,” Appl. Sci. 2023, Vol. 13, Page 7146, vol. 13, no. 12, p. 7146, Jun. 2023, doi: 10.3390/APP13127146.

M. L. W. Concio, F. S. Bernardo, J. M. Opulencia, G. L. Ortiz, and J. R. I. Pedrasa, “Automated Water Meter Reading Through Image Recognition,” IEEE Reg. 10 Annu. Int. Conf. Proceedings/TENCON, vol. 2022-November, 2022, doi: 10.1109/TENCON55691.2022.9977678.

R. Carvalho, J. Melo, R. Graça, G. Santos, and M. J. M. Vasconcelos, “Deep Learning-Powered System for Real-Time Digital Meter Reading on Edge Devices,” Appl. Sci. 2023, Vol. 13, Page 2315, vol. 13, no. 4, p. 2315, Feb. 2023, doi: 10.3390/APP13042315.

M. Imran, H. Anwar, M. Tufail, A. Khan, M. Khan, and D. A. Ramli, “Image-Based Automatic Energy Meter Reading Using Deep Learning,” Comput. Mater. Contin., vol. 74, no. 1, pp. 203–216, 2023, doi: 10.32604/CMC.2023.029834.

A. K. Sharma and K. K. Kim, “Lightweight CNN based Meter Digit Recognition,” J. Sens. Sci. Technol., vol. 30, no. 1, pp. 15–19, Jan. 2021, doi: 10.46670/JSST.2021.30.1.15.

D. Liu, C. Deng, H. Zhang, J. Li, and B. Shi, “Adaptive Reflection Detection and Control Strategy of Pointer Meters Based on YOLOv5s,” Sensors 2023, Vol. 23, Page 2562, vol. 23, no. 5, p. 2562, Feb. 2023, doi: 10.3390/S23052562.

A. Naim, A. Aaroud, K. Akodadi, and C. El Hachimi, “A fully AI-based system to automate water meter data collection in Morocco country,” Array, vol. 10, Jul. 2021, doi: 10.1016/j.array.2021.100056.

“Water Meters Dataset, 1244 Photos & Masks.” Accessed: May 06, 2024. [Online]. Available: https://www.kaggle.com/datasets/tapakah68/yandextoloka-water-meters-dataset.

K. Kanagarathinam and K. Sekar, “Text detection and recognition in raw image dataset of seven segment digital energy meter display,” Energy Reports, vol. 5, pp. 842–852, Nov. 2019, doi: 10.1016/J.EGYR.2019.07.004.

A. Iqbal, A. Basit, I. Ali, J. Babar, and I. Ullah, “Automated Meter Reading Detection Using Inception with Single Shot Multi-Box Detector,” Intell. Autom. Soft Comput., vol. 27, no. 2, pp. 299–309, Jan. 2021, doi: 10.32604/IASC.2021.014250.