講演抄録/キーワード |
講演名 |
2016-11-24 14:50
Hybrid PSO and GSA-based Band Prioritization for Band Selection of High Dimensional Data Sets ○Yang Lang Chang・Yung-Hao Lai・Ming-Xiu Xu・Jyh-Perng Fang(NTUT)・Chihyuan Chu(G-AVE Technology Corporation) SANE2016-64 |
抄録 |
(和) |
With the progress of remote sensing technology, the numbers of bands of hyperspectral images are increased rapidly. However, it leads to the increase of the computational complexity exponentially as problem size increases. The huge data quantity also causes the curse of dimensionality, which leads to the worse accuracy. Therefore, we must reduce the data quantity in order to prevent the curse of dimensionality. In this paper, we propose a novel method, which hybrids the particle swarm optimization (PSO) and gravitational search algorithms (GSA) with impurity function band prioritization (IFBP) method, to reduce and extract the number of bands of hyperspectral images. The experimental results show that the proposed approach can not only work efficiently for reducing the dimension of data sets but also having a better classification accuracy compared to other classifiers. |
(英) |
With the progress of remote sensing technology, the numbers of bands of hyperspectral images are increased rapidly. However, it leads to the increase of the computational complexity exponentially as problem size increases. The huge data quantity also causes the curse of dimensionality, which leads to the worse accuracy. Therefore, we must reduce the data quantity in order to prevent the curse of dimensionality. In this paper, we propose a novel method, which hybrids the particle swarm optimization (PSO) and gravitational search algorithms (GSA) with impurity function band prioritization (IFBP) method, to reduce and extract the number of bands of hyperspectral images. The experimental results show that the proposed approach can not only work efficiently for reducing the dimension of data sets but also having a better classification accuracy compared to other classifiers. |
キーワード |
(和) |
Particle Swarm Optimization / Gravitational Search Algorithm / Impurity Function Band Prioritization / Hyperspectral Images / Multi Criteria Decision Making / Analytic Hierarchy Process / / |
(英) |
Particle Swarm Optimization / Gravitational Search Algorithm / Impurity Function Band Prioritization / Hyperspectral Images / Multi Criteria Decision Making / Analytic Hierarchy Process / / |
文献情報 |
信学技報, vol. 116, no. 319, SANE2016-64, pp. 67-70, 2016年11月. |
資料番号 |
SANE2016-64 |
発行日 |
2016-11-17 (SANE) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
PDFダウンロード |
SANE2016-64 |