Using neural networks and abductive modeling for color error reduction in multimedia applications
Although color has appeal for developers and consumers alike, color reproduction poses a major problem in many computer-based applications including multimedia and desktop publishing. The problem arises because of device independence color, and the way each device processes color. A computer display for instance, displays a color image which composition is constituted by an additive mechanism. From the additive mechanism, all colors can be derived by adding red, green and blue (RGB). A printer, however, uses a subtractive mechanism of cyan, magenta, and yellow (CMY) dyes to subtract colors from the image. So, to print a color image seen on the computer screen so it can be perceived as a true-color reproduction when printed on a paper, the red, green, and blue data values that drive the display must be transformed into data that control the amounts of cyan, magenta, yellow and black on the print. Traditional techniques to control the error in transforming color data include some form of calibration using trial and error, polynomial regression, and neural networks. In this thesis, neural networks and abductive modeling were investigated as transformation techniques. The objectives of this research were: 1). to use RGB as a more natural measure of device independence error (DIE) and 2). to compare the DIE capabilities of abductive modeling with neural network.
School Location:USA - Ohio
Source Type:Master's Thesis
Keywords:neural networks abductive modeling color error reduction multimedia applications
Date of Publication:01/01/1995