全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于图谱分解和概率神经网络的图像分类

DOI: 10.11834/jig.200605104

Keywords: 图谱,图像分类,概率神经网络

Full-Text   Cite this paper   Add to My Lib

Abstract:

为了准确地对不同学习样本数的图像进行分类,首先讨论了对属于不同类别的图像序列如何进行正确有监督分类的问题,由于解决这类问题首先要选取适合图像分类的图像特征作为分类的依据,为此先用图像角点来构成Delaunay图,然后将由Delaunay图的谱特征形成的分类特征矢量作为分类的依据;其次,由于分类器的选取也直接影响分类结果,为此采用了学习效率高的概率神经网络分类器来进行分类.经过大量分类实验表明,图谱特征很好地保持了图像的结构特征,是理想的图像分类特征;经过与其他相关分类器的分类比较实验表明,基于概率神经网络的分类器可以准确地进行图像分类;通过不同学习样本数的比较,证实了概率神经网络在进行图像分类时,对于学习样本数并不敏感,并具有一定稳定性.

References

[1]  Joseph P H,David A L.Classification of remote sensing Images having high spectral resolution[J].Remote Sensing Environment,1996,57(3):119 ~ 126.
[2]  Giorgio Giacinto,Fabio Roli,Lorenzo Bruzzone.Combination of neural and statistical algorithms for supervised classification of remotesensing images[J].Pattern Recognition Letters,2000,21 (5):385 ~ 397.
[3]  Dennis P C,Susan P M,Stanley M D.Machine classification of dental images with visual search[J].Academic Radiology,2001,8(12):1239 ~ 1246.
[4]  Soo B P,Jae W L,Sang K K.Content-based image classification using a neural network[J].Pattern Recognition Letters,2004,25(3):287 ~300.
[5]  Luo Bin,Wilson R C,Hancock E R.Spectral embedding of graphs[J].Pattern Recognition,2003,36 (10):2213 ~ 2223.
[6]  Faruq A Al-Omari,Omar Al-Jarrah.Handwritten Indian numerals recognition system using probabilistic neural networks[J].Advanced Engineering Informatics,2004,18 (1):9 ~ 16.
[7]  Richard O D,Peter E H,David G S.Pattern Classification,Second Edition[M].Etobicoke,Canada:John Wiley&Sons,2001:163 ~ 173.
[8]  Farid Melgani,Sebastiano B S.A statistical approach to the fusion of spectral and spatio-temporal contextual information for the classification of remote-sensing images[J].Pattern Recognition Letters,2002,23 (9):1053 ~ 1061.
[9]  James R C.Spectral and textural classification of single and multiple band digital images[J].Computer&Geosciences,1996,22 (8):849 ~ 865.
[10]  Smith Yoav,Gershom Zajicek,Micheal Werman,et al.Similarity measurement method for the classification of architecturally differentiated images[J].Computer and Biomedical Research,1999,32(1):1 ~12.
[11]  John R S,LI Chung-sheng.Image classification and querying using composite region templates[J].Computer vision and image understanding,1999,75(1):165 ~ 174.
[12]  Harris C G,Stephens M.A combined corner and edge detector[A].In:Proceedings fourth Alvey Vision Conference (AVC88)[C],Manchester,UK,1988:147 ~ 151.
[13]  Specht D F.Probabilistic neural networks[J].Neural Networks,1990,3(1):109 ~118.
[14]  Emanuel Parzen.On estimation of a probability density function and mode[J].Annals of Mathematical Statistics,1962,33 (3):1065 ~ 1076.
[15]  Luo Bin,Wilson R C,Hancock E R.Graph Spectral Approach for Learning View Structure[A].In:Proceedings of 16th International Conference on Pattern Recognition (ICPR2002)[C],Quebec,Canada,2002:785 ~ 788.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133