基于神经网络的测色仪器系统误差分析和校正  

Analysis and Correction of Error for Colorimeter System based on Neural Network

作　　者：徐小力[1] 郑维娜[1] 谷玉海[1] 

机构地区：[1]北京机械工业学院机电系统测控北京市重点实验室,北京100085

出　　处：《自动化仪表》2006年第2期25-27,共3页Process Automation Instrumentation

摘　　要：为了提高工业生产过程中广泛应用的光电积分测色仪的测试精度,分析、研究仪器误差成为仪器设计的重要内容之一。在分析探测器的光谱三刺激非线性误差的基础上,进行了测色仪原理误差的研究,提出了基于人工神经网络的非线性滤波器的方法,并建立了一个基于神经网络的非线性滤波器黑箱模型。实验表明:用此三刺激值计算的色品坐标、色差等参数均达到了设计要求。In order to increase the measurement accuracy of widely used photoelectric integral colorimeters,analyzing and studying the error of the instrument has become one of the important tasks for designing the instrument. On the basis of analysis of spectral tristimulus non-linear error of the detector,the research on principle error of colorimeters is achieved and the method of non-linear filter based on artificial neural network is proposed, and a non-linear filter black box based on neural network is established. The experiment shows that the parameters calculated by tristimulus, e. g. chromaticity coordinates, color difference ,etc. , meet the requirements of design.

关 键 词：色差计 三刺激值 神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]