Open AccessGTCLC: leaf classification method using multiple descriptors
Author(s) -
Kalyoncu Cem,
Toygar Önsen
Publication year - 2016
Publication title -
iet computer vision
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.38
H-Index - 37
eISSN - 1751-9640
pISSN - 1751-9632
DOI - 10.1049/iet-cvi.2015.0414
Subject(s) - artificial intelligence , local binary patterns , hue , pattern recognition (psychology) , mathematics , linear discriminant analysis , classifier (uml) , contextual image classification , feature extraction , computer vision , computer science , histogram , image (mathematics)
The authors propose Geometric, texture and color based leaf classification, a novel leaf classification method using a combination of geometric, shape, texture and colour features that are extracted from the photographic image of leaves. This method combines features that complement each other to define the leaf. A new local binary pattern (LBP) variant, namely sorted uniform LBP (LBP P,R su2 ), is also proposed for leaf texture description. The experiments show that LBP P,R su2 has a higher accuracy in leaf texture classification compared with classical rotation invariant LBP variants. In addition, mean and deviation of hue channel International Commission on Illumination – lightness, chroma, hue colour models – are used to describe the colour tone of the leaf. The proposed feature descriptors require a classifier that can prioritise different features. In this study, linear discriminant classifier (LDC) is employed because of its prioritisation and generalisation abilities. The results show that the proposed combination of features classified with LDC outperforms all well‐known leaf classification methods.