基于主成分分析的彩色图转灰度图方法的研究

摘要

主要讲述利用主成分分析方法将彩色图像转变为灰度图像.传统的彩色图转灰度图是由R、G、B三元色按照一定的权重进行加权而得.一般的转化公式对于所有RGB或RGBA的图像都适用,它虽然符合心理学研究结果,但是没有反映出不同图像本身的特点.阐述利用主成分分析方法对彩色图像转化为灰度图.该方法既能根据不同图像有不同的灰度,又能增强灰度图的对比度,适合作为深度学习的图像输入.%This paper mainly describes the transformation of color images into grayscale images using prin－cipal component analysis.The traditional color map grayscale is weighted by R,G and B,and is weighted ac－cording to certain weights.The general transformation formula is applicable to all images of RGB or RGBA,al－though it conforms to the results of psychological research,but does not reflect characteristics of different images themselves.In this paper,the color image is converted to grayscale by principal component analysis.This meth－od can be used as image input for deep learning,which can be used to improve the contrast of gray scale image according to different images.Abstract: This paper introduces the key technology and working principle of the robot tool switching de－vice which can significantly improve the working ability and the efficiency of the tool switching for the robot.A new tool switching device and its control system are designed to meet the requirements of its use for the existing 3P3R robots.Before the tool switching,the position and posture error of robot is corrected through error correc－tion control program to improve the accuracy of robot.By using the 3P3R robot tool switching device,the 3P3R robot can perform various,diversified and multi kinds of work to improve the efficiency and flexibility.