Automatic KWh Recording Using OCR and Raspberry Pi 3 with Firebase Cloud Database

This research aims to develop an automatic recording system for electricity consumption in $\mathbf{k W h}$ using Optical Character Recognition (OCR) technology and Raspberry Pi 3. The background of the problem is that manual recording by officers often faces obstacles such as the homeowner’s absence or obstructed access to the meter, which can lead to inaccuracies in electricity billing. The method used in this research is the prototype method which involves several steps which is Communication Stage, Quick Planning Stage, Quick Modeling, construction, and evaluation, where the system is equipped with a camera to capture images of the total kWh on the electricity meter. These images are then processed using OCR to convert visual data into numeric data readable by a computer. The results can be monitored through a designed web monitoring system. The novelty of this research lies in the development of an automatic system utilizing OCR and Raspberry Pi 3 for recording electricity kWh, which has not been widely implemented in Indonesia. The results show that this system can provide ease and reduce errors in recording electricity kWh, with a recording accuracy rate of $\mathbf{8 0 \%} \% \mathbf{- 1 0 0 \%}$. The benefits of this system include increased efficiency and accuracy in recording electricity consumption, which can reduce billing errors and improve energy management processes. The contribution of this research is expected to provide an innovative solution for more accurate and efficient electricity kWh recording.