Open AccessLeaf feature extraction using glcm, moment invariant and shape morphology for indonesian medicinal plants recognition
Author(s) -
Hermawan Syahputra,
Zulfahmi Indra,
Didi Febrian,
Dhea Putri Adriani
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1317/1/012008
Subject(s) - artificial intelligence , pattern recognition (psychology) , feature extraction , invariant (physics) , moment (physics) , computer science , computer vision , image processing , texture (cosmology) , mathematics , image (mathematics) , physics , classical mechanics , mathematical physics
This study aims to determine the extraction of GLCM texture features, shape morphology and moment invariant features on the leaf image of medicinal plants and determine the accuracy of plant recognition based on these three features by using Artificial Neural Network Classifiers. The procedure performed to classify medicinal plants based on their leaf image is image acquisition, image pre-processing, feature extraction, image classification and calculating the accuracy of test results. The introduction had tested for ten Indonesian medicinal plant samples, namely: Bangun-Bangun, Binahong, Jarak, Kemuning, Mangkokan, Mengkudu, Pegagan, Sambiloto, Sambung Nyawa, and Sirih. Based on the test results, obtained 97% accuracy with GLCM features, 69% with Shape Morphological features, 86% with GLCM and Shape Morphological features and 79% with moment invariant features.