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.
出处
《自动化仪表》
CAS
2006年第2期25-27,共3页
Process Automation Instrumentation
关键词
色差计
三刺激值
神经网络
Color-difference meter Tristimulus values Neural network
