图像特征的CNN提取方法及其应用

被引:6
作者
雷国伟
吕迎阳
纪安妮
吴孙桃
郭东辉
机构
[1] 厦门大学物理与机电学院
关键词
CNN; 并行处理器; 图像识别;
D O I
暂无
中图分类号
TP391.4 [模式识别与装置];
学科分类号
0811 ; 081101 ; 081104 ; 1405 ;
摘要
图像特征的提取是视觉图像识别的重要方法之一,采用细胞神经网络(CNN)并行处理器进行图像特征的提取具有实时快速的优点。该文将介绍CNN并行处理器的基本工作原理及其实现图像特征处理的逻辑组合通用方法,并以图像的纹理分割与识别为例来说明CNN并行处理器应用于视觉图像识别的通用编程方法。
引用
收藏
页码:204 / 206+216 +216
页数:4
相关论文
共 9 条
[1]  
Cellularneuralnetworks:applications. ChuaLO,YangL. CircuitsandSystems,IEEETransactionson . 1998
[2]  
A16x16cellularneuralnetworkuniversalchip:Thefirstcompletesingle-chipdynamiccomputerarraywithdis-tributedmemoryandwithgray-scaleinput-output. CruzJM,ChuaLO. ANALOGIN-TEGRATEDCIRCUITSANDSIGNALPROCESSING . 1998
[3]  
Cellularneuralnetworks:theory. Chua L O,Yang L. CircuitsandSystems,IEEETransactionson . 1988
[4]  
ACE16K:a128/spltimes/128focalplaneanalogprocessorwithdigitalI/OCellularNeuralNetworksandTheirApplications. LinanG,Rodriguez-VazquezA,Espejo,Setal. CNNA2002,Proceedingsofthe20027thIEEEInternationalWorkshopon . 2002
[5]  
TheCNNuniversalmachine:ananalogicarraycomputer. RoskaT,ChuaLO. IEEETransactionsonCircuitsandSystemsII:AnalogandDigitalSignalProcessing .
[6]  
CNN:AVisionofComplexity. JLHuertas W-K,ChenRNMadan. InternationalJournalofbifurcationandCHAOSinappliedsciencesandEngineering . 1997
[7]  
AnewCNNbasedtoolforanautomatedmorphometryanalysisofthecornealendothelium. SalernoM,SargeniF,BonaiutoVetal. CellularNeuralNetworksandTheirApplicationsProceedings,1998FifthIEEEInternationalWorkshopon . 1998
[8]  
NoveltypesofanalogicCNNalgorithmsforrecognizingbank-notes. ZarandyA,WerblinF,Roska,Tetal. CellularNeuralNetworksandtheirApplications,1994,CNNA-94,ProceedingsoftheThirdIEEEInternationalWorkshopon . 1994
[9]  
Imageprocessingusinghystereticcellularneuralnetworks. MateiRP. CircuitsandSystems,2000,ProceedingsISCAS2000Geneva,The2000IEEEInternationalSymposiumon . 2000