Open AccessClassification accuracy improvement of laser-induced breakdown spectroscopy based on histogram of oriented gradients features of spectral images
Author(s) -
Jiujiang Yan,
Ping Yang,
Zhongqi Hao,
Ran Zhou,
Xiangyou Li,
Shisong Tang,
Yun Tang,
Xiaoyan Zeng,
Yongfeng Lu
Publication year - 2018
Publication title -
optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.394
H-Index - 271
ISSN - 1094-4087
DOI - 10.1364/oe.26.028996
Subject(s) - laser induced breakdown spectroscopy , histogram , spectroscopy , materials science , histogram of oriented gradients , optics , speckle pattern , pattern recognition (psychology) , laser , artificial intelligence , remote sensing , computer science , analytical chemistry (journal) , chemistry , physics , geology , chromatography , image (mathematics) , quantum mechanics
To improve the classification accuracy of laser-induced breakdown spectroscopy (LIBS), image histogram of oriented gradients (HOG) features method (IHFM) for materials analysis was proposed in this work. 24 rice (Oryza sativa L.) samples were carried out to verify the proposed method. The results showed that the classification accuracy of rice samples by the full-spectra intensities method (FSIM) and IHFM were 60.25% and 81.00% respectively. The classification accuracy was obviously improved by 20.75%. Universality test results showed that this method also achieved good results in the plastics, steel, rock and minerals classification. This study provides an effective method to improve the classification performance of LIBS.