Title :
Image based water gauge reading developed with ANN Kohonen
Author :
Triantoro, Teguh ; Batubara, F. Rizal ; Fahmi, F.
Author_Institution :
Electr. Eng. Dept., Univ. of Sumatera Utara, Medan, Indonesia
Abstract :
The purpose of this study is to develop a method to read PDAM Tirtanadi water gauge from image captured by a digital camera mobile phone using Kohonen artificial neural network model There are several processes within the application which are image acquisition, preprocessing, segmentation, training ANN Kohonen and digit number identification process. Based on simulation using 90 number digit from 15 water gauge data, we obtained the recognition results with the accuracy rate of 95,56% (per number digit) and 86,67% accuracy rate for the water gauge readings.
Keywords :
cameras; gauges; image recognition; image segmentation; mobile handsets; neural nets; water meters; Kohonen ANN; Kohonen artificial neural network model; PDAM Tirtanadi water gauge; digital camera mobile phone; image acquisition; image based water gauge reading; image capture; image preprocessing; image segmentation; Accuracy; Artificial neural networks; Databases; Image recognition; Image segmentation; Mobile handsets; Training; Artificial Neural Network; Kohonen; Number Recognition; Water Gauge;
Conference_Titel :
Electrical Engineering and Computer Science (ICEECS), 2014 International Conference on
Print_ISBN :
978-1-4799-8477-0
DOI :
10.1109/ICEECS.2014.7045223
