Ensemble of One-Dimensional Classifiers for Hyperspectral Image Analysis

被引:0
作者
Ksieniewicz, Pawel [1 ]
Krawczyk, Bartosz [1 ]
Wozniak, Michal [1 ]
机构
[1] Wroclaw Univ Technol, Dept Syst & Comp Networks, Wybrzeze Wyspianskiego 27, PL-50370 Wroclaw, Poland
来源
DATA MINING AND BIG DATA, DMBD 2016 | 2016年 / 9714卷
关键词
Ensemble learning; Hyperspectral imaging; Computer vision; Feature extraction; Dimensionality reduction; Image classification; FEATURE-EXTRACTION; CLASSIFICATION;
D O I
10.1007/978-3-319-40973-3_52
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Remote sensing and hyperspectral data analysis are areas offering wide range of valuable practical applications. However, they generate massive and complex data that is very difficult to be analyzed by a human being. Therefore, methods for efficient data representation and data mining are of high interest to these fields. In this paper we introduce a novel pipeline for feature extraction and classification of hyperspectral images. To obtain a compressed representation we propose to extract a set of statistical-based properties from these images. This allows for embedding feature space into fourteen channels, obtaining a significant dimensionality reduction. These features are used as an input for the ensemble learning based on minimal-distance classifiers. We introduce a novel method for forming ensembles simple one dimensional classifiers. They are constructed independently on a low-dimensional representation - a single classifier for each extracted feature. Then a voting procedure is being used to obtain the final decision. Extensive experiments carried on a number of benchmarks images prove that using proposed feature extraction and ensemble of simple classifiers can offer a significant improvement in terms of classification accuracy when compared to state-of-the-art methods.
引用
收藏
页码:513 / 520
页数:8
相关论文
共 14 条
[1]   Combined 5 x 2 cv F test for comparing supervised classification learning algorithms [J].
Alpaydin, E .
NEURAL COMPUTATION, 1999, 11 (08) :1885-1892
[2]   Hyperspectral image nonlinear unmixing and reconstruction by ELM regression ensemble [J].
Ayerdi, Borja ;
Grana, Manuel .
NEUROCOMPUTING, 2016, 174 :299-309
[3]  
Cyganek B, 2010, LECT NOTES COMPUT SC, V6475, P191, DOI 10.1007/978-3-642-17691-3_18
[4]   High- resolution landcover classification using Random Forest [J].
Hayes, Matthew M. ;
Miller, Scott N. ;
Murphy, Melanie A. .
REMOTE SENSING LETTERS, 2014, 5 (02) :112-121
[5]  
Krawczyk B, 2014, LECT NOTES COMPUT SC, V8480, P274
[6]   A novel hyperspectral segmentation algorithm-concept and evaluation [J].
Ksieniewicz, Pawel ;
Jankowski, Dariusz ;
Ayerdi, Borja ;
Jackowski, Konrad ;
Grana, Manuel ;
Wozniak, Michal .
LOGIC JOURNAL OF THE IGPL, 2015, 23 (01) :105-120
[7]  
Lasota T., 2013, LNCS LNAI, V7828, P1
[8]   A novel approach to hyperspectral band selection based on spectral shape similarity analysis and fast branch and bound search [J].
Li, Shijin ;
Qiu, Jianbin ;
Yang, Xinxin ;
Liu, Huan ;
Wan, Dingsheng ;
Zhu, Yuelong .
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2014, 27 :241-250
[9]   RETRACTED: A novel method of feature extraction and fusion and its application in satellite images classification (Retracted Article) [J].
Lin, Da ;
Xu, Xin .
REMOTE SENSING LETTERS, 2015, 6 (09) :687-696
[10]   SCALE-SPACE AND EDGE-DETECTION USING ANISOTROPIC DIFFUSION [J].
PERONA, P ;
MALIK, J .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1990, 12 (07) :629-639