Article Per Year (5 Year)
p-Index From 2019 - 2024
0
P-Index
This Author published in this journals
All Journal Jurnal Informatika
Widi Hapsari
Program Studi Teknik Informatika Fakultas Teknologi Informasi UKDW Yogyakarta
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
PENERAPAN OPTICAL CHARACTER RECOGNITION (OCR) UNTUK PEMBACAAN METERAN LISTRIK PLN Robert Gunawan; Sri Suwarno; Widi Hapsari
Jurnal Informatika Vol 10, No 2 (2014): Jurnal Teknologi Komputer dan Informatika
Publisher : Universitas Kristen Duta Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (819.168 KB) | DOI: 10.21460/inf.2014.102.331

Abstract

This paper will discuss an automatic recognition system to facilitate the activities of reading and recording of the electricity kWh meter. The system will analyze a captured image of the electricity kWh meter display using character recognition method. Prior to applying this method, each photo image of the kWh meter display will be pre-processed (i.e. converting the image to grayscale and determining the threshold). To detect the kWh meter display area, a smearing approach will be applied which connected the component labeling to character segmentation. To recognize each number, a template matching will be used. This study show that these steps were not yields good result in recognizing the characters in the electricity kWh meter. This is due to several factors, such as the persistence of noise that interfered with the character recognition, camera angles, lighting effect and the effect of the smearing limit.
ANALISIS TEKSTUR PADA CITRA MOTIF BATIK UNTUK KLASIFIKASI MENGGUNAKAN K-NN Kristian Adi Nugraha; Widi Hapsari; Nugroho Agus Haryono
Jurnal Informatika Vol 10, No 2 (2014): Jurnal Teknologi Komputer dan Informatika
Publisher : Universitas Kristen Duta Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1693.096 KB) | DOI: 10.21460/inf.2014.102.332

Abstract

Indonesian’s Batik is one of culture heritage that recognized around the world. Batik has many variations of pattern based on their region. In this research, Batik would be used as subject for texture feature extraction. The value of this feature extraction would be used for classification using K-Nearest Neighbor (K-NN) method. Texture Feature Extraction components that used in this research were Entropy, Correlation, Homogeneity, and Energy. This research will investigate which component would give dominant effect for Batik’s pattern recognition. Batik pattern used in this research is pattern from Yogyakarta region. There are four patterns namely Ceplok, Parang, Semen, and Nitik. The result showed that there was no component from Texture Feature Extraction that gave dominant effect (average = 53,96%). Component with the highest value of accuracy is Correlation with a percentage of 55,83%. Whereas for K-NN classification, the best accuracy is 60% for K = 5.
Co-Authors Kristian Adi Nugraha Nugroho Agus Haryono Robert Gunawan Sri Suwarno