Comparison Analysis of the Artificial Neural Network Algorithm and K-Means Clustering in Gorontalo Herbal Plant Image Identification System

Yulita Salim; Mukhlisulfatih Latief; Novri Kandowangko; Rampi Yusuf

doi:10.1109/EIConCIT.2018.8878665

Comparison Analysis of the Artificial Neural Network Algorithm and K-Means Clustering in Gorontalo Herbal Plant Image Identification System

Closed

Yulita Salim, Mukhlisulfatih Latief, Novri Kandowangko, Rampi Yusuf

2018 Proceedings - 2nd East Indonesia Conference on Computer and Information Technology: Internet of Things for Industry, EIConCIT 2018 Conference paper Cited by 1

Abstract

The objective of this study was to analyze the comparison between artificial neural network algorithm and k-means clustering to see the extent of the effectiveness of this algorithm on the identification of Gorontalo herbal plant image. This study uses a digital imaging processing method with segmentation and extraction techniques. Segmentation proses used thresholding method. The next process was extraction process of the characteristics of the image of the herbal plant using the shape and color characteristics to obtain the metric, eccentricity, hue, saturation, and value of the plant was carried out. These five parameters were used as parameters to identify the herbal plant image. This study used 91 images which consisted of 80 imagery training and 11 test images. The study revealed that k-means clustering accuracy was 27.27% whereas the artificial neural network algorithm accuracy was 54.54%. In this case artificial neural networks had better accuracy than K-means. © 2018 IEEE.

Affiliations

Faculty of Computer Science, Universitas Muslim Indonesia, Makassar, Indonesia; Department of Informatics Engineering, State University of Gorontalo, Gorontalo, Indonesia

View full text (DOI) →

Research at a Glance

Research at a Glance

Topics & SDG Alignment

Collaboration

Author Profile (Selected)

References Overview

Journal & Source

Metadata & Integrity