Acquisition and Reproduction of Colour Images: Colorimetric and Multispectral Approaches
A dissertation submitted in partial fulfillment of the degree of
”Docteur de l’Ecole Nationale Supérieure des Télécommunications”,
Paris, France, 1999.
Colour imaging, colorimetry, colorimetric characterisation, multispectral imaging,
spectral characterisation, filter selection, spectral reconstruction.
First edition published in 1999 by Ecole Nationale Supérieure des
Télécommunications, 46, rue Barrault, F-75634 Paris Cedex 13, France, under the title "Acquisition et reproduction d'images couleur : approches colorimétrique et multispectrale," Publication ENST 99 E 021.
Second edition published in 2001 by Universal Publishers / dissertation.com, 7525 NW 61 Terrace, Suite 2603, Parkland, FL 33067-2421. The second edition is available in paperback and e-book (pdf) format from dissertation.com.
Changes from the first edition include layout, adding of a List of Tables and a List of Figures, correction of miscellaneous typographical errors, translation to American English (you know, color instead of colour, etc.) modifications to some Figures, to the numbering of Figures and Equations, and adding a new preface.
More information about the described research can be obtained
Prof. Francis Schmitt,
Dr. Hans Brettel,
Dr. Jon Yngve Hardeberg,
or through the address below.
Ecole Nationale Supérieure des Télécommunications
46, rue Barrault
F-75634 Paris Cedex 13
The goal of the work reported in this dissertation is to develop
methods for the acquisition and reproduction of high quality digital
colour images. To reach this goal it is necessary to understand and
control the way in which the different devices involved in the entire
colour imaging chain treat colours. Therefore we addressed the problem
of colorimetric characterisation of scanners and printers,
providing efficient and colorimetrically accurate means of conversion
between a device-independent colour space such as the CIELAB space,
and the device-dependent colour spaces of a scanner and a printer.
First, we propose a new method for the colorimetric characterisation of colour
scanners. It consists in applying a non-linear correction to the
scanner RGB values followed by a 3rd order 3D polynomial regression
function directly to CIELAB space. This method gives very good
results in terms of residual colour differences.
The method has been successfully applied to
several colour image acquisition devices, including digital
cameras. Together with other proposed algorithms for image quality
enhancements it has allowed us to obtain very high quality digital
colour images of fine art paintings.
An original method for the colorimetric characterisation of a printer is then proposed. The method is based on a computational geometry approach. It uses a 3D triangulation technique to build a tetrahedral partition of the printer colour gamut volume and it generates a surrounding structure enclosing the definition domain. The characterisation provides the inverse transformation from the device-independent colour space CIELAB to the device-dependent colour space CMY, taking into account both colorimetric properties of the printer, and colour gamut mapping.
To further improve the colour precision and colour fidelity we have
performed another study concerning the acquisition of multispectral
images using a monochrome digital camera together with a set of K>3
carefully selected colour filters.
Several important issues are
addressed in this study. A first step is to perform a spectral
characterisation of the image acquisition system to establish the
spectral model. The choice of colour chart for this characterisation
is found to be very important, and a new method for the design of an
optimised colour chart is proposed.
Several methods for an optimised selection of colour filters are then proposed, based on the spectral properties of the camera, the illuminant, and a set of colour patches representative for the given application.
To convert the camera output signals to device-independent data, several approaches are proposed and tested. One consists in applying regression methods to convert to a colour space such as CIEXYZ or CIELAB. Another method is based on the spectral model of the acquisition system. By inverting the model, we can estimate the spectral reflectance of each pixel of the imaged surface.
Finally we present an application where the acquired multispectral images are used to predict changes in colour due to changes in the viewing illuminant. This method of illuminant simulation is found to be very accurate, and working on a wide range of illuminants having very different spectral properties.
The proposed methods are evaluated by their theoretical properties, by simulations, and by experiments with a multispectral image acquisition system assembled using a CCD camera and a tunable filter in which the spectral transmittance can be controlled electronically.
Web design and content by Jon Y. Hardeberg.
Last modified on July 2, 2001. Disclaimer: The contents of this web page represent my personal view of the world, and I am the only one that can be blamed for any blunders.