KWh-meter Number Recognition using Normalized Cross-Correlation Technique


Downloads per month over past year

Kartika Firdausy, KTF (2014) KWh-meter Number Recognition using Normalized Cross-Correlation Technique. In: International Conference on Green World in Bussiness and Technology (ICGWBT) 2014, 29 March 2014, Yogyakarta, Indonesia.

Kartika Firdausy_ KWh-meter Number Recognition using Normalized Cross-Correlation Technique.pdf

Download (434kB) | Preview


Electricity demand has become a major requirement for modern society. Many complaints forwarded by the customer's electrical system is recording kWh-meter that sometimes less accurate. This study aims to design an application that can detect the number shown on the kWh-meter automatically, to then be sent by the registrar to the office in a text format. Process starting from the introduction of image acquisition, image conversion into grayscale format, thresholding, cropping, normalization, segmentation, template matching, and identification of patterns of numbers. System implementation using Microsoft Visual C++ 2012 with OpenCV 2.4.6. Testing the accuracy of the resulting application includes image recognition rate, which compares the process of automatically and manually. From the experimental results, the average recognition error is 20.7%, hence the accuracy of recognition using this technique is 79.3%.

Item Type: Conference or Workshop Item (Paper)
Additional Information: ISSN: 2355-553X
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisi / Prodi: Faculty of Industrial Technology (Fakultas Teknologi Industri) > S1-Electrical Engineering (S1-Teknik Elektro)
Depositing User: Kartika Firdausy
Date Deposited: 02 Apr 2018 01:34
Last Modified: 02 Apr 2018 01:34

Actions (login required)

View Item View Item