Image-based Automatic Dial Meter Reading in Unconstrained Scenarios. (30th November 2022)

• Record Type: Journal Article
• Authors: Salomon, Gabriel; Laroca, Rayson; Menotti, David
• Is Part Of: Measurement. Volume 204(2023)
• Publication Date: 2022-11-30
• Subjects: Automatic Meter Reading -- Dial meters -- Pointer-type meters -- Deep learning -- Public dataset; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2022.112025
• Languages: English
• ISSNs: 0263-2241

Abstract

Abstract: The replacement of analog meters with smart meters is costly, laborious, and far from complete in developing countries. The Energy Company of Paraná (Copel) (Brazil) performs more than 4 million meter readings (almost entirely of non-smart devices) per month, and we estimate that 850 thousand of them are from dial meters. Therefore, an image-based automatic reading system can reduce human errors, create a proof of reading, and enable the customers to perform the reading themselves through a mobile application. We propose novel approaches for Automatic Dial Meter Reading (ADMR) and introduce a new dataset for ADMR in unconstrained scenarios, called UFPR-ADMR-v2. Our best-performing method combines YOLOv4 with a novel regression approach (AngReg), and explores several post-processing techniques. Compared to previous works, it decreased the Mean Absolute Error (MAE) from 1343 to 129 and achieved a meter recognition rate (MRR) of 98.9% — with an error tolerance of 1 Kilowatt-hour (kWh). Highlights: New approaches for Automatic Dial Meter Reading in unconstrained scenarios. A benchmark with 20 different approaches for ADMR. The proposed system achieves state-of-the-art results while being quite efficient. A novel publicly available dataset for ADMR in unconstrained scenarios. We tackle common ADMR issues with new post-processing correction methods.