仿生模式识别与多权值神经元

内容简介

[

　　this book is the second one after the first book named “first
step to multi-dimensional space biomimetic informatics”（in
chinese）， which are both illuminating the novel biomimetic
high-dimensional space geometry computing theory， but this book is
more detailed and systemic. this book consists of three parts，
statistical pattern recognition， biomimetic pattern recognition and
multi-weight neuron. biomimetic pattern recognition and
multi-weight neuron are proposed by academician shoujue wang at the
start of representing digital data over hundreds of dimensionality
as points， and developed for five years with many applications in
many fields so far.

]

目录

part i review of statistics pattern recognitionchapter 1 introduction of pattern recognition1.1 pattern recognition concept1.2 pattern recognition system biasic processes1.3 a brief survey of pattern recognition appro aches1.4 scope and organizationchapter 2 kernel of statistical pattern recognition andpre-precessing2.1 question arise2.1.1 question expression2.1.2 empirical risk minimization2.1.3 generalization ability and complexity2.2 kernel of statistical pattern recognition2.2.1 vapnik-chervonenkis dimension2.2.2 the bounds of generalization ability2.2.3 the minimization of structural risk2.3 preprocessin92.4 feature extraction and feature selection2.4.1 curse of dimensionality2.4.2 feature extraction2.4.3 feature selection2.5 support vector manchine2.5.1 the optimal hyperplane under linearly separable2.5.2 the soft spacing under linearly nonseparable2.5.3 the kernel function under non-linear case2.5.4 support vector machine’s traits and advantagesreferencespart ii biomimetic pattern recognitionchapter 3 introductionchapter 4 the foundation of biomimetic pattern recognition4.1 overview of high-dimensional biomimetic informatics4.1.1 the proposal of the problem of computer imaginalthinking4.1.2 the principle of high-dimensional biomimeticinformatics4.2 basic contents of high-dimensional biomimetic informatics4.3 main features of high-dimensional giomimetic informatics4 4 concepts and mathematical symbols in high-dimensionalbiomimetic informatics4.4.1 concepts and definitions4.4.2 mathematical symbols4.4.3 symbolic computing methods in resolving geometry computingproblems4.4.4 several applications in solving complicated geometrycomputing problems4.5 some applications4.5.1 blurred image restoration4.5.2 uneven lighting image correction4.5.3 removing facial makeup disturbanceschapter 5 the theory of biomimetic pattern recognition5.1 the concept of biomimetic pattern recognition5.2 the choice of the name5.3 the developments of biomimetic pattern recognition5.4 covering.the concept of recognition in biomimetic patternrecognition5.5 the principle of homology-continuity： the starting point ofbiomimetic pattern recognition5.6 expansionary product5.7 experiments5.7.1 the architecture of the face recognition system5.7.2 umist face data5.7.3 pre-treatment5.7.4 the realization of svm face recognition algorithms5.7.5 the realization of bpr face recognition algorithms5.7.6 experiments results and analyzes5.8 summarychapter 6 applications6.1 object recognition6 2 a multi-camera human-face personal identification system6.3 a recognition system for speaker-independent continuousspeech6.4 summaryreferencespart ⅲ multi-weight neurons and networkschapter 7 history and definations of artificial neuralnetworks7.1 from biological neural networks to artificial neural networksand its development7.2 some definitions and concepts of artificial neuralnetworks7.3 unifications and divergences between array-processors andneural networks7.4 artificial neural networks’ effects on nanoelectronicalcomputational technologychapter 8 geometric concepts of artificial neurons8.1 mathematical expressions of common neurons and their geometricconcepts8.2 general mathematical model of common neurons and its geometricconcept8.3 direction basis function neuron and its geometric concept8.4 multi-threshold neurons and networkschapter 9 multi-weight neurons and their applications9.1 general mathematical expression of multi-weight neurons’functions9.2 interchangeabilities of points， vectors， hyper planes inhigh-dimensional space9.3 effect of high-dimensional point distribution ana

封面

书名:仿生模式识别与多权值神经元

作者:王守觉　等著

页数:167

定价:¥48.0

出版社:国防工业出版社

出版日期:2012-12-01

ISBN:9787118080810

PDF电子书大小:106MB 高清扫描完整版