Automated E Billing System

Project Title

Automated E Billing System

Project Area of Specialization Artificial IntelligenceProject Summary

Electricity is one of the most important necessities in today’s world. Life at this age can’t be imagined without it. The billing system of electricity in Pakistan is still using the orthodox approaches leading towards more complaints from the customer’s side. The meter reading system in Pakistan is still too old and involves too much human effort due to which most of the consumers face billing problems every month. The Electricity Corporation of Pakistan hasn’t done too many efforts to improve the billing system. A system was proposed to improve the system by taking pictures of the meters and then attach those pictures along with the consumer bills so that the consumers can verify their readings from the picture. But there are still some errors in that method due to manual entry of readings from those pictures. In this work, an Android-based application will be developed that takes the image of meter as an input, extract the meter reading from it and send it to the central server and save in the database where the web portal will be utilized to generate the bills. Consumers can also track their previous and current bills & readings.

Project Objectives

The main objective of this project is to develop an Automated E-Billing System for the electricity provider companies that not only extract meter readings from the meter but also to generate bills and track the previous record from the web portal. This System should be able

• To extract relevant information from the snapshots of meters.
• To provide a web-based interface that can be used by the consumers & companies to track/print their monthly bills.

Project Implementation Method

To achieve the desired objectives of the project the following will be the major steps:

Step 1: Training a classifier for meter type recognition

A classifier would be trained on two different classes of meters (Digital and Analogue). Classifier would first detect whether it is a digital meter or analog and then according to meter type, further processing would be done.

Step 2: Text recognition using OCR (Optical Character Recognition)

After classifying the meter, some pre-processing steps would be applied to the meter image so that it would be easy for OCR to recognize the characters from it. For character recognition, (Tesseract) OCR library will be used which is an open source project of Google Inc.

Step 3: Development of Mobile and Web Application

In order to develop mobile application, Android OS would be used where the information would be extracted from the meter image and then send those readings to a web portal developed using ASP .NET which is a Microsoft technology for the development of web applications, where consumer bill is generated and consumer can view and print his/her bills.

Benefits of the Project

Our project targets the electricity providing companies in Pakistan. The proposed system will help them:

• To automate their billing system.
• To gain the trust of their consumers by providing an errorless & transparent billing system.
• To reduce manpower & expenditures.

Technical Details of Final Deliverable

The final deliverable will be an Android application and a web portal for it. Android application will scan the meter readings along with their reference numbers while with web portal people can keep track and take print of their previous and current bills.

Final Deliverable of the Project Software SystemType of Industry IT , Energy Technologies Artificial Intelligence(AI), OthersSustainable Development Goals Decent Work and Economic Growth, Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	75000
GPU	Equipment	1	40000	40000
Android Device	Equipment	1	28000	28000
Web Server	Miscellaneous	1	7000	7000