SIA OpenIR  > 工业控制网络与系统研究室
一种基于遗传算法选择最优光谱谱段的方法及其应用
Alternative TitleGenetic algorithm based optimal spectrum segments selecting method
孙兰香; 于海斌; 张鹏; 丛智博; 辛勇
Department工业控制网络与系统研究室
Rights Holder中国科学院沈阳自动化研究所
Patent Agent沈阳科苑专利商标代理有限公司 21002
Country中国
Subtype发明
Status有权
Abstract本发明涉及一种基于遗传算法选择最优光谱谱段的方法及其应用,具体步骤为:1)获得LIBS光谱数据;2)参数编码,形成遗传算法初始种群;3)主成分分析;4)训练人工神经网络模型;5)评价网络;6)形成新种群;7)重复3)~6)至满足适应度指标,输出最优分段及对应的最优分类器;8)应用分类器对未知样品进行分类。本方法训练出的分类器可对训练样品对应种类的待测样品进行准确分类,从而定性分析样品成分组成。
Other AbstractThe method involves obtaining training sample spectrum data. Training sample spectrum data wavelength and intensity range determining process is carried out. Spectrum segmentation parameter coded bits determining process is carried out. A main component information extracting process is carried out by using artificial neural network algorithm. A network training set is provided with multiple training samples. Original population re-inserting process is carried out to form an optimal population. An optimal spectral outputting process is carried out in an optimal classification network.
PCT Attributes
Application Date2015-05-19
2017-01-04
Date Available2019-04-23
Application NumberCN201510259959.1
Open (Notice) NumberCN106295667A
Language中文
Contribution Rank1
Document Type专利
Identifierhttp://ir.sia.cn/handle/173321/19745
Collection工业控制网络与系统研究室
Affiliation中国科学院沈阳自动化研究所
Recommended Citation
GB/T 7714
孙兰香,于海斌,张鹏,等. 一种基于遗传算法选择最优光谱谱段的方法及其应用[P]. 2017-01-04.
Files in This Item: Download All
File Name/Size DocType Version Access License
CN201510259959.1.pdf(598KB)专利 开放获取CC BY-NC-SAView Download
CN201510259959.1授权.p(468KB)专利 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[孙兰香]'s Articles
[于海斌]'s Articles
[张鹏]'s Articles
Baidu academic
Similar articles in Baidu academic
[孙兰香]'s Articles
[于海斌]'s Articles
[张鹏]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[孙兰香]'s Articles
[于海斌]'s Articles
[张鹏]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: CN201510259959.1.pdf
Format: Adobe PDF
File name: CN201510259959.1授权.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.