一种基于GA-BP神经网络结合PCA的LCD显示器光谱特征化模型  

Spectral characterization model of liquid crystal display based on GA-BP neural network and PCA

作　　者：于海琦[1] 刘真[1] 田全慧[2] 

机构地区：[1]上海理工大学出版印刷与艺术设计学院,上海200093 [2]上海出版印刷高等专科学校,上海200093

出　　处：《液晶与显示》2016年第2期201-207,共7页Chinese Journal of Liquid Crystals and Displays

基　　金：国家自然科学基金-青年科学基金(No.61301231)~~

摘　　要：为实现LCD显示器的光谱特征化,本文提出一种基于遗传算法优化(Genetic Algorithm,GA)的BP神经网络(GABP)结合PCA(Principal component analysis)的光谱特征化模型。首先对显示器色空间进行子空间划分,同时采用PCA对光谱数据进行降维,接着在各子空间中采用遗传算法对BP神经网络的权值阈值进行优化,建立显示器驱动值与光谱数据之间的神经网络模型,实现了显示器的光谱特征化。实验结果表明子空间划分后,在子空间中进行模型参数的优化有利于模型整体精度的提高,GA的优化有效改善了BP神经网络的极值问题,提高了模型的精度,PCA在不影响模型精度的同时提高了算法的运行效率。由此说明该模型是一种高精度显示器特征化模型。To achieve spectral characterization of LCD,a spectral characterization model based on GABP neural network and PCA was proposed in this paper.Firstly,color space of monitor was divided into subspaces and principal component analysis（PCA）was used to dimensionality reduction of the spectrum reflectance.Then genetic algorithm（GA）was adopted for optimizing the weight threshold of BP neural network in subspaces.A BP neural network was established to describe the transformation model of driver value and spectral reflectance.Lastly,the spectral characteristics model of LCD was realized.Experimental results show that model parameter optimization in the subspace is beneficial to the improvement of the model of overall after subspace division.The extremum problem of BPneural network can be improved by GA effectively.PCA can improve the operating efficiency of algorithm under the circumstances of maintaining accuracy.This implies that it is a high-precision display characteristic model.

关 键 词：BP神经网络 遗传算法 光谱特征化 液晶显示器 主成分分析 

分 类 号：TP394.1[自动化与计算机技术—计算机应用技术] TH691.9[自动化与计算机技术—计算机科学与技术]