计算机科学 ›› 2015, Vol. 42 ›› Issue (4): 292-296.doi: 10.11896/j.issn.1002-137X.2015.04.060

• 图形图像与模式识别 • 上一篇    下一篇

一种基于背景自学习的高光谱图像生物信息提取方法

张玉香,高旭杨,王 挺,张乐飞,杜 博   

  1. 武汉大学遥感信息工程学院 武汉430079,武汉大学计算机学院 武汉430072,武汉大学测绘遥感信息工程国家重点实验室 武汉430079,武汉大学计算机学院 武汉430072,武汉大学计算机学院 武汉430072
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受国家“973计划”资助

Background Self-learning Framework for Bio Information Extraction from Hyperspectral Images

ZHANG Yu-xiang, GAO Xu-yang, WANG Ting, ZHANG Le-fei and DU Bo   

  • Online:2018-11-14 Published:2018-11-14

摘要: 为了提取指纹、癌变区域等重要的生物信息,传统方法一般是使用物理、化学手段直接作用在信息载体上,这不仅需要较长时间,容易对原有信息及载体造成破坏,而且提取过程不可重现、精度较低。高光谱成像技术避免了获取信息时物理接触造成的破坏,能多次稳定获取图像,成为了一种优秀的生物信息采集途径。在此介绍一种基于背景自学习的高光谱图像信息提取方法,它解决了传统非结构化背景模型适应性不强的问题,利用空间光谱信息进一步提升了信息提取精度。实验证明,该方法能有效对背景信息进行估计,提取完整的生物目标信息,精度优于传统目标信息提取方法。

关键词: 高光谱图像,生物信息提取,背景自学习

Abstract: Physical or chemical methods are commonly used to extract certain bio information,such as fingerprint extraction,tumor region detection,etc.These methods may not only be time-consuming,but also possibly damage the entire bio information carrier.Meanwhile,the process cannot be recurred and reach a satisfactory accuracy.A new technique,hyperspectral imaging,can be adopted for the information extraction,by which the origin information will not be contaminated and can be able to be acquired from the image repeatedly.We proposed an information extraction method from hyperspectral images based on a background self-learning framework.In the conventional unstructured background models,it may be difficult to accurately estimate the background statistics,neither in a global nor local way.The proposed method can avoid this problem.Considering the spatial spectral information,its performance can be further improved.It is designed to extract fingerprint and tumor region from hyperspectral bio images.The experimental results show the validity and the superiority of our method for the bio information extraction from hyperspectral images.

Key words: Hyperspectral image,Bio information extraction,Background self-learning

[1] 王崇文,李见为.指纹取像与指纹识别[J].计算机工程,2002,28(4):10-12
[2] Sudiro S A,Paindavoine M,Kusuma T M.Simple fingerprint mi-nutiae extraction algorithm using crossing number on valley structure[C]∥2007 IEEE Workshop on Automatic Identification Advanced Technologies.IEEE,2007:41-44
[3] Lee C J,Wang S D.Fingerprint feature extraction using Gabor filters[J].Electronics Letters,1999,35(4):288-290
[4] Zhang Q,Yan H.Fingerprint classification based on extraction and analysis of singularities and pseudo ridges[J].Pattern Re-cognition,2004,37(11):2233-2243
[5] Abbas Q,Celebi M E,García I F.Skin tumor area extractionusing an improved dynamic programming approach[J].Skin Research and Technology,2012,18(2):133-142
[6] Gil J,Wu Hai-shan,Wang B Y.Image analysis and morphometry in the diagnosis of breast cancer[J].Microscopy Research and Technique,2002,59(2):109-118
[7] 张兵,高连如,等.高光谱图像分类与目标探测[M].北京:科学出版社,2011
[8] 黄远程.高光谱影像混合像元分解的若干关键技术研究[D].武汉:武汉大学,2010
[9] Manolakis D,Shaw G.Detection algorithms for hyperspectralimaging applications[J].Signal Processing Magazine,IEEE,2002,19(1):29-43
[10] Harsanyi J C,Farrand W H,Chang C I.Detection of subpixelsignatures in hyperspectral image sequences[C]∥Proceedings of the American Society for Photogrammetry and Remote Sensing.1994:236-247
[11] Scharf L L,McWhorter L T.Adaptive matched subspace detectors and adaptive coherence estimators[C]∥Conference Record of the Thirtieth Asilomar Conference on Signals,Systems and Computers,1996.IEEE,1996:1114-1117
[12] Park B,Windham W R,Lawrence K C,et al.Contaminant classification of poultry hyperspectral imagery using a spectral angle mapper algorithm[J].Biosystems Engineering,2007,96(3):323-333
[13] Chang C I.Orthogonal subspace projection (OSP) revisited:acomprehensive study and analysis[J].IEEE Transactions on Geoscience and Remote Sensing,2005,43(3):502-518
[14] 刘凯,张立福,杨杭,等.面向对象分析的非结构化背景目标高光谱探测方法研究[J].光谱学与光谱分析,2013,33(6)
[15] Fuhrmann D R,Kelly E J,Nitzberg R.A CFAR Adaptive-Matched Filter Detector[J].IEEE Transaction on Aerospace and Electronic Systems,1992,28(1):208-216
[16] Matteoli S,Acito N,Diani M,et al.An automatic approach to adaptive local background estimation and suppression in hyperspectral target detection[J].IEEE Transactions on Geoscience and Remote Sensing,2011,49(2):790-800
[17] Bioucas-Dias J M,Nascimento J M P.Hyperspectral subspaceidentification[J].IEEE Transactions on Geoscience and Remote Sensing,2008,46(8):2435-2445
[18] 王千,王成,冯振元,等.K-means聚类算法研究综述[J].电子设计工程,2012,20(7):21-24
[19] DeLong E R,DeLong D M,Clarke-Pearson D L.Comparing the areas under two or more correlated receiver operating characteri-stic curves:a nonparametric approach[J].Biometrics,1988,44(3):837-845

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!