WO2001066353A1

WO2001066353A1 - Image processing for colour control

Info

Publication number: WO2001066353A1
Application number: PCT/AU2001/000265
Authority: WO
Inventors: Ian Sherwood Mason
Original assignee: Foxtip Pty Ltd
Priority date: 2000-03-08
Filing date: 2001-03-08
Publication date: 2001-09-13

Abstract

Image processing is effected by processing an input data file having data representative of a scanned image which is required for colour printing, the method including producing an output electronic data file with amendments to apply rules affecting colour dot disposition to correct for what would otherwise be loss of detail and colour distortion in printing a colour image from the original file with transfer ink which is adapted to be printed onto a substrate for subsequent transfer under heat and pressure to the product such as a fabric.

Description

IMAGE PROCESSING FOR COLOUR CONTROL

The present invention relates to the processing of an image so that colour printing may be accomplished. One important application of the invention is to colour printing to produce an ultimate product such as a printed fabric or other substrate.

The printing of fabric has long been established using a method known as silk-screen printing which involves the selective application of suitable coloured materials to function as a dye to create a desired pattern or image on a fabric. However, these techniques are relatively slow, expensive to operate in terms of labour and facilities and there are severe limitations on the type of images which can be produced. Another known approach is to use special inks whereby an image is deposited on a transfer medium, typically a special paper, and under heat and pressure by virtue of sublimation the ink material is transferred and permanently attaches to a fabric. In commercial operations to produce panels and rolls of fabric, direct printing methods are also used, but all these methods for fabric printing have severe limitations on the degree of complexity which can be accommodated and the degree of detail and trueness of image generation thereby severely limiting these techniques.

Printing and in particular colour printing of images onto paper is, however, well established, and the most common process is the four-colour process whereby an image is separated into four monochromatic images from which four printing plates are produced. The product, e.g. paper is then subjected to four printing passes which respectively apply the "screened" image in terms of the four colours, i.e. black, cyan, yellow and magenta. This process is well known to provide high quality images in full colour and great detail. The most common printing process is lithography, but other techniques are used. Conventionally, commercial printing has used half- toning techniques to convert the various tone values of an original image into a geometric arrangement of "half-tone dots" by a process known as "screening" . The half-tone dots are so small as to be not readily discernible by the human eye ordinarily. The overall tone value perceived is proportional to the total surface area covered by the half-tone dots.

Conventionally, half-tone dots are placed in a periodical pattern so the number of dots per unit surface area is constant and tone values are simulated by changing the size of individual dots. This conventional technique is an amplitude-modulation (AM) screening technique. Particular problems can arise when AM screening is used in colour reproduction and an alternative method has been developed and is known as stochastic or frequency- modulation (FM) screening. FM screening is characterised by using equally sized dots whose number per unit surface area is made proportional to the tone and value that must be reproduced. Thus FM screening uses microdots smaller than the typical half-tone dots of AM screening. As it will be described below, embodiments of the present invention preferably use FM screening methodology with specific variations in the methodology to achieve surprisingly, for the first time a significant enhancement in the product when it is to be applied to printing various substrates such as fabric. Most preferably, the present invention is implemented with the use of Agfa CristalRaster technology, but applying this technology in a new way while retaining the superior FM screening techniques inherent in that technology.

Unlike conventional half-tone methods, in_. the Agfa CristalRaster process the microdots are placed randomly rather than being clustered to form a half-tone dot. The process is electronically controlled using algorithms which have been very effective in the production of images and graphics with the microdots for commercial grade colour reproduction typically being 21 microns for the production of an image classified as 2,400 dpi. Since its introduction in November 1993, the Agfa CristalRaster technology has been developed and acknowledged as providing reproduction of a continuous tone colour photograph on a modern offset printer with a very high degree of fidelity.

The present invention manifests itself in various aspects including methods of production, intermediate products and final products, all aimed towards producing a printed substrate such as textile or fabric which reproduces with good fidelity and detailed images, including colour images.

In a first aspect, the invention is provided in a method of producing an output electronic data file from an input electronic data file, the input file having data representative of a scanned image and which would be suitable for a colour printing process onto paper, and amending the file to apply rules affecting colour dot disposition to correct for what would otherwise be loss of detail and colour distortion in printing a colour image from the original file with transfer ink adapted to be transferred in a heat and pressure process to a substrate.

In a second aspect, the invention consists in a method including the following steps: a) taking a selected non-paper substrate such as a textile or fabric intended to be finally used for receiving a colour image in a process of sublimation under heat and pressure to transfer the image from printed transfer paper, b) taking a transfer paper having printed thereon with the selected inks intended to be used in the final process a test pattern comprising a series of discrete zones for each of the colours to be printed and graded substantially from zero to 100% colour dot coverage, c) applying heat and pressure to transfer the test pattern to the substrate, d) using an optical method to measure colour density for at least some of the discrete zones and recording the values observed, and e) comparing the values measured with that expected for faithful colour reproduction and using calibration calculations to define altered dot coverage percentages for at least some of the values measured to provide instructions adapted to be used to amend the data file for the method described in the previous paragraph.

Preferably all the test pattern areas are measured and required corrections for the particular fabric and inks deduced and most preferably the areas are in about 5% dot percentage increments.

The invention also extends to such an amended electronic data file as well as a method of production and a product being of the nature of a printed transfer sheet utilising inks for sublimation printing onto fabrics. The invention also consists in substrates such as fabrics printed by such a process. Practical implementation of the invention through to the printing of a substrate such as a fabric preferably has the following steps:

1. A compensation curve for each colour ink is derived for the particular substrate to be used by conducting an experiment using selected inks to determine the reaction of the particular inks to the particular substrate with measurement of colour density, in the manner generally outlined above in the second aspect of the invention and preferably as exemplified further herein after.

2. An original work is scanned to establish an electronic data file preferably conforming to normal standards used for electronic pre-press printing processes.

3. The data file is amended to apply data from the compensation curves for the ink-substrate combination to be used.

4. An image corresponding to the amended data file is produced on film.

5. Lithographic plates are produced from the film typically in a known manner.

6. The lithographic plates are used to print onto transfer paper a colour image with appropriate control of ink selection and control of ink densities onto paper.

7. Heat and pressure is applied to transfer the image from the paper to the selected substrate.

The invention is best applied where FM screening technology is used for production of the lithographic film. AM screening may be feasible where it is acceptable to operate at a coarse level but the best colour reproduction and detail is achievable with FM techniques. To derive the compensation curves, the arbitrary scale of a conventional reflective densitometer can be used and the test pattern can be derived from a conventional electronically loaded data file Known calibration software for colour printing can be applied to use the data derived from the reflective densitometer with a comparison made for the value of each dot percentage with that expected and program adjustments made so that the conventionally derived data file defining the image to be printed is adjusted. For example this procedure may determine that where colour corresponding to a 100% density of cyan is required, the dots must be reduced to 90% of the normal value to compensate. A similar analysis is done for each of the steps in the test pattern and each colour. It has been found that advantageously existing Agfa calibration programs supplied with image setters can be used in applying the present inventive concepts.

Despite systems such as the Agfa systems and software having been available for some time, and despite the well- established technology of heat transfer inks for printing fabrics, the result has been of poor quality in terms of detail and accuracy of colour reproduction. Despite the long standing market demand for better quality, there have been no proposals known to the present inventor which solve the problem. The present inventor has achieved, at least with preferred embodiments, a most significant advance over prior art methods and products yet the solution to the problem is elegant and can be applied using existing modern colour pre-press equipment and software .

Using the known Agfa CristalRaster screening methodology, for conventional printing a D_max is around 4.4 yet it has been found that the radically different control steps used by the present invention typically has a D_max of around 1.7 with a D_mιn of about 0.72.

A significant realisation behind the present invention and the novel methodology is the appreciation that the different inks in a colour printing process, when applied to fabric printing, have most significant differences in behaviour on the fabric relative to one another. In contrast to conventional wisdom and the trade literature, there has been no suggestion that this factor exists and no suggestion that via appropriate compensation curves the problem of good quality colour printing on fabric or similar substrates can be solved.

Preferably the heat transfer step is conducted at a temperature of around 200°C and pressure of about 70lbs per square inch.

Lithographic printing is the preferred process and to obtain good results carefully controlled standards need to be preserved with regard to ink densities for the four standard colours; preferably the order of printing should ensure that black is printed last. However, systems other than standard 4-colour systems could be used.

It is paramount to achieving the best results of the invention to derive accurately and correctly the required compensation curves for the ink-fabric combination.

The applicants have found that it is preferred to use the Agfa Cristal Raster methodology and in this embodiment it has been found that it is best to use a film that has a density of 4.4 and is produced in a positive form. Furthermore there should be present on the film a printing colour bar which should be on the "grip" edge of the film so that product quality can be appraised. Preferably the creation of the compensation curves utilises a test pattern having a series of graduated tints of each of the colours using the conventional four- colours, namely cyan, black, magenta and yellow but other colour combinations could also be used.

The next step is to enter the information into a calibration program which calculates a compensation curve that will allow for the spread of ink on the material . Thus the compensation curve is different for each colour on each material.

For illustrative purposes only, an embodiment of the invention will now be described with reference to the accompanying diagrams of which:

Figure 1 is a black and white reproduction of a typical test pattern as applied to a sample test fabric using sample inks;

Figure 2 is a table of test data using a reflective densitometer on a particular test pattern with a particular ink and fabric; and Figures 3, 4 and 5 are respective compensation curves derived from data measured on respectively three different test fabrics and in this case indicating the compensation required for black ink.

Referring first to Figure 1 the test pattern has four test pattern strips for cyan, yellow, magenta and black.

For each colour across the fabric there are a series of squares labelled zero to 100 in increments of 5 running left to right and a second series of squares as a check for uniformity across the fabric and running right to left. This corresponds with the theoretical dot percentage. Thus each 100% square in the printing process has dots of such density that there is 100% lay down of ink in that square and e.g. in the 50% square, half the area if viewed microscopically would be white space and the other half with colour dots.

Along the upper edge, the test sample also has printed colour bars which can be used to check uniformity of the process across the width of the fabric.

A reflective densitometer is then used to measure each of the squares and a value derived as shown in figure 2 and the data supplied to a colour calibration program to derive compensation curves. The sample compensation curves of Figures 3, 4 and 5 are for different fabrics but with the same inks. The interpretation of the curve can now be explained in the following terms taking the example of a desired 50% dot coverage. The desired curve is the diagonal straight line. The upper curve represents measured data and shows that a 92% value has been measured in the 50% square of the transferred test patent. The correction program has calculated that to achieve the appearance of 50% dot coverage in the final product the colour black needs to be reduced to 14% dot coverage so that it will appear as a 50% dot coverage when the transfer printing has been concluded.

By comparing figures 3, 4 and 5 it will be seen significantly different compensation is required for different fabrics and in this case for the colour black some very substantial corrections are required.

When this process is repeated for all four colours and printing has been effected, surprisingly good rendition of colour faithfulness and detail has been achieved. Preferably the calibration program used is the Agfa Calibrator Version 4 but a significant and novel approach of the present inventors compared with any prior art methods is the realisation that measurement of performance of individual inks is both possible and necessary to achieve calibration to achieve good results in the printing of fabric.

Claims

1. A method of producing an output electronic data file from an input electronic data file, the input file having data representative of a 'scanned image and which would be suitable for a colour printing process onto paper, and amending the file to apply rules affecting colour dot disposition to correct for what would otherwise be loss of detail and colour distortion in printing a colour image from the original file with transfer ink adapted to be transferred in a heat and pressure process to a substrate having ink take up characteristics different to paper.

2. A method as defined in claim 1 wherein the file amendment is effected using compensation curves derived for each of the coloured inks to be used and derived from a test pattern applied to a sample of the substrate.

3. A method including the following steps : a) taking a selected non-paper substrate intended to be finally used for receiving a colour image in a process of sublimation under heat and pressure to transfer the image from printed transfer paper, b) taking a transfer paper having printed thereon with the selected inks intended to be used in the final process a test pattern comprising a series of discrete zones for each of the colours to be printed and graded from zero to

100% colour dot coverage, c) applying heat and pressure to transfer the test pattern to the substrate, d) using an optical method to measure colour density for at least some of the discrete zones and recording the values observed, and e) comparing the values measured with that expected for faithful colour reproduction and using calibration calculations to define altered dot coverage percentages for at least some of the values measured to provide instructions adapted to be used to amend the data file for the method of claim 1.

4. A method is claimed in claim 3 , wherein the test pattern areas are measured and required corrections for the particular fabric and inks deduced and most preferably the areas are in about 5% dot percentage increments.

5. A method is claimed in claim 3 or claim 4, and further including producing on film an image for each colour separation using FM screening technology.

6. A data film for producing images resulting from the method of any one of claims 1-4.

7. Film colour separations of an image produced by the method of any one of claim 1-5.