颜色恒常性[1]是人类视觉系统的一个重要特征,它能帮助我们在不同照明条件下,一 定程度地消除干扰,对物体的本来颜色做出正确判断。但是显然,机器并没有这样的功 能,它们无法像人类一样做出这样的调节,这样在不同的照明条件下,机器所采集的图 像的颜色和物体本来的颜色多多少少存在偏差,这种偏差即为偏色[2]。图像的偏色现象对 于许多基于图像色彩特征的研究十分不利,会降低研究中算法的性能。78635
本论文实现了对图像的这种偏色现象的检测和校正算法,使用数字图像处理技术, 最终的成果可以导入和保存图片,得到图像的偏色因子以对图像是否偏色做出判断,以 及对偏色图像的多种校正等。图像的偏色检测采用基于图像等效圆的检测,算出偏色因 子来判断偏色程度;偏色的校正分别采用灰度世界、结合灰度世界和完美反射、直方图 拉伸三个校正算法,争取每个偏色图像能找到适合它的方法。
关键词 偏色检测 偏色校正
毕 业 设 计 说 明 书 外 文 摘 要
Title Color Cast Detection and Correction
Abstract
Human visual system has color constancy, so that we in the light of different illumination conditions, are able to eliminate interference to a certain extent, the original color of the object to make correct judgment。But, of course, the machine does not have such a feature, they cannot like human made such regulation, therefore the acquisition of color of the true colors of images and the object in different light illumination will has a deviation, the deviation is called color cast。 The color image's partial color phenomenon is very disadvantageous to the research of many image color characteristics, which will reduce the performance of the algorithm in the research。
This paper proposed to image the partial color phenomena detection and correction algorithm using digital image processing techniques, the final results can import and save the picture, image is obtained by partial color factor to the image whether partial color to make judgments, and to a variety of color image correction。Image color cast detection use image equivalent circle detection based on calculated partial color factor to determine the color deviation degree; partial color correction use three methods :gray world, combined with the gray world and perfect reflection, histogram stretching correction algorithm to find a suitable method of each partial color image。
Keywords colour cast detection color cast correction
本科毕业设计说明书 第 I 页
目 次
1 引言 1
1。1 概述 1
1。2 国内外研究状况及发展趋势 1
1。3 生产需求状况 2
1。4 课题设计背景和意义 2
1。5 本文结构 2
2 偏色检测与校正常用方法 3
2。1 偏色现象 3
2。1。1 偏色产生原因 3
2。1。2 简单判别偏色 3
2。2 偏色检测技术介绍 4
2。3 偏色校正技术介绍 4
2。3。1 局部校正 图像的偏色检测及校正:http://www.youerw.com/jisuanji/lunwen_90677.html