Key Technology Of Color Management Cie L ~ * ~ * B. A ~ * And Cmyk Transform Algorithm

Color reproduction is always the focused technology in information transmitting, printing publishing, computer vision and image processing, ect. Color management is used to solve the color conversion problem in different devices and to keep color consistent and correct in the whole image processing .And color conversion is the important part in color management technology and has great significance on image reproduction.In this dissertation color conversion algorithm of CEE L~*a~*b~*and CMYK is researched, which is the key technology in output device characterization and color management module. And the study could be used in on-line image detecting, computer color matching, image color separation, digital proofing and image impainting,ect. Hence the research work is valuable both in theory and application.Lots of helpful exploratory works have been done in color conversion theory and method. The main contributions of this dissertation are as follows:1. The theory, method, workflow, key technology, system construction and application of color management technology have been analyzed and the academic fundamental, implementing process and characteristics of common used color space transformation methods are compared.2. The standard color target ECI2002 is selected to study in CMYK four color printings where the dot colorimetric values of one color(e.g.C) and K are settled while the others(e.g.M and Y) are variable, lie on a flat plane. And then variables' linear pertinence and plane prominence of regressed plane equations are verified based on mathematic statistical method.3. The conversion of CMY and L~*a~*b~* is built under colorimetric plane theory which has higher precision. Further research has been done to build up the conversion from CMYK to L~*a~*b~* based on mathematic statistical method and numerical analyzing method. Then the factors influencing precision of these equations are analyzed. 4. The generation mechanism of the plane theory is firstly analyzed by CIE1931Yxy chromaticity system and it is found that there is some existing rules included in the colorimetric values of CMYK four color halftone dots.5. Sample data have been grouped according to L~* value to train BP network. And the result shows that the one hidden layer is not suitable in the study while a BP network with two hidden layers for conversion L~*a~*b~* to CMYK, which has a relative quicker speed and higher precision, has been trained by combination different neuron numbers of two hidden layers.6. GRNN is present first time to study conversion from L~*a~*b~* to CMYK in this dissertation. Grouped data according to L~* value and non-grouped data have been used separately to set up GRNN. The result shows that the GRNN gains an advantage over BP network weather in training conveniency ,speed or precision. So,GRNN is much more suitable for color conversion research.