Automated and Reliable Meter Reading System

Abstract

This project aims to automate the existing process of electric meter reading system and make it reliable. Currently, meter readers take pictures of meters and manually enters data into their system. Current System is not automated, it is a very time consuming, costly and can be riddled with errors. This project considers Lahore Electric Supply Company (LESCO) as a case study and for testing the proposed system. Our main objective is to facilitate the company employees, save them from hardships of entering data and avoid from the possible human mistakes. This application will not only facilitate the company but also improve the trust of customers and significantly decrease the illegal use of electricity along with the easy monitoring of line losses and theft of electricity. The system will use machine learning, image processing techniques to detect the regions of interest and extract current meter reading. Image will be captured through mobile app and will be processed to get the number of units consumed by the customer.