WO2010089597A1

WO2010089597A1 - Displaying image data

Info

Publication number: WO2010089597A1
Application number: PCT/GB2010/050176
Authority: WO
Inventors: Ian John Grimstead; Nicholas John Avis
Original assignee: University College Cardiff Consultants Limited
Priority date: 2009-02-05
Filing date: 2010-02-04
Publication date: 2010-08-12
Also published as: GB0901825D0

Abstract

A method of displaying image data on a display device including obtaining (202) a set of device-specific colour values, each said device-specific colour value in the set being associated with a corresponding non-device specific colour value. The method further includes analysing (502, 504) image data representing an image to identify at least one non-device specific colour value included in the image data, and displaying (206) at least part of the image represented by the image data on the display device using the corresponding device-specific colour value instead of the identified non-device specific colour value.

Description

Displaying Image Data

The present invention relates to displaying image data.

Many types of devices for displaying image data output by a computer or the like are available. As a general rule, the better the quality/resolution of the display device, the higher the cost of the device. For diagnostic imaging in the medical profession, for example, Radiologists use monitors calibrated to meet various DICOM (Digital Imaging and Communications in Medicine) standards

(e.g. DICOM PS 3.14-2007 "Digital Imaging and Communications in medicine

(DICOM) Part 14: Grayscale Standard Display Function). Such monitors are relatively expensive compared with standard VDU commonly used with desktop

PCs. Such DICOM-calibrated monitors are calibrated to show levels of grey at a set intensity, as set by the DICOM standard. However, it is noted that (in the UK at least) there is no official approach; just a recommendation by the Royal

College of Radiologists regarding the type of monitor to use and the like when reporting (i.e. analysing X-Ray images and reporting on patients). The

"reporting" quality monitor has to show sufficient differentiation between levels of grey to enable a Radiologist to correctly view an image.

Embodiments of the present invention are intended to allow lower-bit capacity, e.g. 8-bit, display devices (for use in situations such as the example given above) to present greyscale images with an improved perceived image quality compared with the pure grey shades normally used by the display device.

According to a first aspect of the present invention there is provided a method of displaying image data on a display device, the method including: obtaining a set of device-specific colour values, each said device-specific colour value in the set being associated with a corresponding non-device specific colour value; analysing image data representing an image to identify at least one non- device specific colour value included in the image data, and displaying at least part of the image represented by the image data on the display device using the corresponding device-specific colour value instead of the identified non-device specific colour value.

The device-specific colour values and the non-device specific colour values may correspond to greyscales/shades of grey. The device-specific colour values may comprise a subset of a PseudoGrey sequence.

The analysing of the image data to identify at least one non-device specific colour value may include identifying a pure grey pixel.

The method may include: displaying at least one colour on the display device, the colour being identified by a non-device specific colour value; recording a device-specific colour value representing at least one characteristic of the displayed colour as recorded by a light analysis device, and storing data associating the device-specific colour value with the non- device specific colour value. The at least one colour may be part of a test image display. The at least one characteristic may include intensity/brightness and/or chromaticity/hue.

The steps of displaying at least one colour, recording the device-specific colour value and storing the data associating the device-specific colour value may be performed at regular intervals or continuously. The storing of data associating the non-device specific colour value for the colour with the device-specific colour value may include: identifying minimum and maximum values of said recorded device- specific colour values; calculating a range of greyscale values between the minimum value and the maximum value; searching the device-specific colour values for values that best match each said greyscale value in the range, and storing data associating the best match device-specific colour value with the corresponding greyscale value in the range.

The minimum and maximum values may correspond to measured intensity or chromaticity, for example. The calculated range of greyscale values may comprise a pure linear curve between the minimum value and the maximum value. According to another aspect of the present invention there is provided a method of creating a set of device-specific colour values for use in displaying image data on a display device, the method including: displaying at least one colour on the display device, the colour being identified by a non-device specific colour value; recording a device-specific colour value representing at least one characteristic of the displayed colour as recorded by a light analysis device, and storing data associating the device-specific colour value with the non- device specific colour value.

According to yet another aspect of the present invention there is provided apparatus adapted to display image data on a display device, the apparatus including: a component configured to obtain a set of device-specific colour values, each said device-specific colour value in the set being associated with a corresponding non-device specific colour value; a component configured to analyse image data representing an image to identify at least one non-device specific colour value included in the image data, and a component configured to modify a video signal to the display device so that the display device displays at least part of the image represented by video signal using the corresponding device-specific colour value instead of the identified non-device specific colour value.

The apparatus may comprise a modifier device connected between a computer and the display device. Alternatively, the apparatus may comprise a computing device that generates the video signal .

The apparatus may include: a component configured to display at least one colour on the display device, the colour being identified by a non-device specific colour value; a light analysis device configured to record a device-specific colour value representing at least one characteristic of the displayed colour, and a component configured to store data associating the device-specific colour value with the non-device specific colour value.

The light analysis device may be spaced apart from the display device so that ambient light affects readings taken by the light analysis device. The light analysis device can hence record both ambient light levels and output level of the display device, rather than requiring two separate analysis devices.

The apparatus may include a plurality of the light analysis devices, each of the light analysis devices being configured to receive light from different portions (e.g. corners) of the display device. The apparatus may be configured to process measurements taken by the plurality of light analysis devices to calculate averaged ambient light. The apparatus may be configured to produce different device-specific colour values for different portions of the display device.

Intensity measurements taken by the plurality of light analysis devices can be used to perform a linear, quadratic or other interpolation of ambient light and/or colour skew.

According to a further aspect of the present invention there is provided a computer program product comprising a computer readable medium, having thereon computer program code means, when the program code is loaded, to make the computer execute a method of displaying image data on a display device substantially as described herein.

According to yet another aspect of the present invention there is provided a video signal modified in accordance with a method substantially as described herein. According to a further aspect of the present invention there is provided a method of creating a set of device-specific colour values for use in displaying image data on a display device, the method including: displaying at least one colour using a display device; measuring a characteristic of light produced by the display device when displaying the shade; comparing the measured characteristic and a desired value for light produced when displaying the colour, and storing data associating the measured characteristic and the desired value.

The method may further include using the stored data to modify display of the at least one colour by the display device.

The method may be repeated automatically at intervals, or continuously.

Whilst the invention has been described above, it extends to any inventive combination of features set out above or in the following description.

Although illustrative embodiments of the invention are described in detail herein with reference to the accompanying drawings, it is to be understood that the invention is not limited to these precise embodiments. As such, many modifications and variations will be apparent to practitioners skilled in the art. Furthermore, it is contemplated that a particular feature described either individually or as part of an embodiment can be combined with other individually described features, or parts of other embodiments, even if the other features and embodiments make no mention of the particular feature. Thus, the invention extends to such specific combinations not already described. The invention may be performed in various ways, and, by way of example only, embodiments thereof will now be described, reference being made to the accompanying drawings in which:

Figure 1 is a block diagram showing a computer connected to a light analysis device and a display device; Figure 2 is a flowchart illustrating schematically steps performed using the computer and the analysis and display devices;

Figure 3 illustrates schematically steps performed during a test image process; Figure 4 demonstrates differences between inputs and outputs of a conventional display device;

Figure 5 illustrates schematically steps performed during a greyscale data set creation process;

Figure 6 illustrates schematically steps performed during a device-specific image display process, and

Figure 7 is a block diagram showing an example of an implementation of the device-specific image display process.

Referring to Figure 1 , a computing device 102 is shown connected to a display device 104. The computer 102 may be a conventional desktop PC or the like. The display device 104 may comprise one of several types of display devices, e.g. a monitor, mobile telephone/device screen or projector, implementing any suitable display technology, e.g. CRT, LCD, some electronic ink replacements, etc, and is capable of displaying image data based on a video signal transmitted from the computer. The video signal comprises a certain number of bits determined by the manufacturer of the display device/computer.

For instance, in an 8-bit display device, each Red, Green, Blue (RGB) value that represents the intensity of each of those colours in order to determine the displayed colour of each pixel in the display comprises 8 bits. Thus, the displayed colour of each pixel will be determined by one of 256 possible values for Red, one of 256 possible values for Green and one of 256 possible values for Blue, which means that each pixel can be displayed as one of 16 million (256 x 256 x 256) colours. High-quality image data, such as that used in Radiography, can be represented using a higher number of bits, e.g. 12-bits per channel. A light analysis device 106, such as a light meter or spectrometer, is also connected to the computer 102. The computer is configured to execute code that displays test patterns/images comprising shades of grey on the display device 104 and to record properties of the resulting display as received by the analysis device 106. An overview of this process is shown in Figure 2. At step 202, the computer 102 displays a test image(s) based on a set of greyscales on the display device 104 and also records a value representing the intensity received by the analysis device 106 when each greyscale is displayed. The test images in the specific example cycle through 1 ,735 levels of PseudoGrey whilst triggering the analysis device in order to form a comparison between measured and expected values. This process effectively records which greyscales can be best displayed by the device 104 and the recorded values can be considered to be display device-specific greyscale values.

At step 204, the computer creates data associating the device-specific greyscale values recorded by the analysis device 106 with non-device specific greyscale values, the non-device specific values being selected from a range of intensities between the minimum and maximum intensities identified in a manner as will be described below. At step 206, the computer uses the data created at step 204 to display a (non-test) image. Figure 3 details operations performed during step 202 of Figure 2. It will be appreciated that all the steps described herein are exemplary only and some of them can be omitted or re-ordered in alternative embodiments. The process starts at step 302 and at step 304 the analysis device 106 is focussed on the display device 104. In many embodiments, the analysis device is spaced apart from the screen of the display device, which means that ambient light as well as light projected from the screen will affect the reading taken by the analysis device. For example, the analysis device may be a simple photo/light meter (such as a Konica Minolta CS-200 Chroma Meter) and can be spaced apart from the screen by a meter or two, e.g. a space typical of the distance between the screen and a human viewer during normal use. In alternative embodiments, the analysis device may be fitted/clipped on to the display device, with a small gap (e.g. 3 - 10 mm) between its sensor input and the screen, which can still allow ambient light to affect its readings. The main body of the photometer can be fitted to the rear of the monitor, with a thin "light pipe" extending from the screen to the sensor itself for minimal intrusion.

In yet another embodiment a plurality of the analysis devices are used, each fitted adjacent different portions (e.g. corners) of the screen. The use of multiple sensors can be used to effectively detect averaged ambient light, which can obviate errors that may be introduced by a slight shadow on a portion of the screen but which would not be detected by a single sensor. Use of multiple sensors can also allow varying of PseudoGrey mapping across the screen. Given four sensors, for instance, the calibration algorithm can be written to blend the intensity readings across the screen. An individual Pseudo Grey look-up table can be provided for each sensor and the hardware/software implementation of the processes described herein can use pixel position information to determine the weight from each look-up table and hence produce a non-uniform adjustment across the display. For instance, one side of the monitor could be in slight shade, which means that darker shades are visible on that side compared to the other side.

The intensity measurements taken at the four corners can be used to perform a linear, quadratic or other interpolation of ambient light and/or colour skew. The correction factor so produced can be varied over the screen to produce a corrected display that is visible above ambient light and produces consistent lighting levels across the display.

At step 306, a test image is displayed on the device 104. The test image will typically comprise a particular sequence of greyscales. In one embodiment, the greyscale values are based on a set known as "PseudoGrey", which is a sequence where there is a +/-1..n intensity step from pure grey on each colour channel (red, green, blue). This can produce slight off-white tints, which appear white to the eye, but can introduce additional greyscale intensity levels. For example, a standard LCD monitor can be used to display a subset of 256 of the 1 ,785 shades of grey that are available using the PseudoGrey concept of +/- 1 to pure grey. An additional 6 off-white shades of grey can be perceived as pure grey by the observer, but exhibit additional levels of intensity. An example comparison between conventional/pure grey shades and PseudoGrey is given below:

Pure grey: Red = green = blue (R=G=B) Sequence: 0,0,0; 1 ,1 ,1 ; 2,2,2; 3,3,3; ... ; 255,255,255

PseudoGrey: R=G=B; R=G, B=G+1 ; B=G, R=G+1 ; ...

Sequence: 0,0,0; 0,0,1 ; 1 ,0,0; 1 ,0,1 ; 0,1 ,0; ... ; 100,101 ,100; ... ; 255,255,255 It will be understood that alternatives to the example PseudoGrey sequence above can be used if desired. Modifying the sequence away from +/- 1 from grey levels allows other colour changes to be performed. For example, in alternative embodiments colours other than grey can be detected and swapped for a preferred colour, rather than the technique being restricted to pure shades of grey. An example is correcting shades of green to ensure they contain the minimal contribution of other colours (such as red and blue), which can be useful for chroma keying with digital video compositing (for instance, where the colour green is replaced in an image, which is an issue a projected image is used). Alternatively, pure green shades could be augmented towards blue or red, so the green shade will not be removed during the chroma key process. In such embodiments, the light meter is used to sample chromaticity (or another colour/light characteristic) instead of intensity as in the PseudoGrey example described above. Intensity shades that reduce the colour gamut/chromaticity range can then be selected, e.g. to produce "purer" white, instead of a range of greyscales.

If a display device is scanned for all the available intensity levels using a test based on the PseudoGrey sequence, 256 shades of grey can be selected from this range and used to replace pure grey colours found in image data. This can then produce 256 discernable shades of grey that are linear rather than lie on a gamma curve, or produce any other desired intensity curve (such as DICOM, or a curve intended to match the intensity curve of a printing mechanism/medium). A standard monitor exhibits a "gamma curve", whereby 2x the logical intensity does not produce 2x the physical intensity. Figure 4 shows a monitor 402 with a gamma curve of 2.5 and the non-linear relationship between input 404 and output 406. This could be corrected in a software driver, resulting in a straight line intensity plot, but at the expense of available shades of grey. Some shades would be discarded using such an approach and so there is a potential reduction to around 200 shades out of 256 available. This is unacceptable for diagnostics because the source data is often 12-bit (4096 shades) and reducing the colour set even further could hide medical issues/artefacts.

Embodiments of the system described herein are further capable of taking into account ambient lighting so that the darkest shade of grey will be 0,0,0, followed by the first shade of grey that can be seen above that at the ambient light level. This may be found to be 25,25,25, for example, rather than the next, conventional pure grey shade of 1 ,1 ,1 , which may not be visible above pure black 0,0,0.

At step 308 the analysis device samples the light projected from the display device (in combination with the ambient light). A value representing the intensity is recorded by the analysis device. Due to display device-specific characteristics (and ambient light) the intensity recorded by the analysis device may not match the RGB values used by the computer to represent the greyscale/colour displayed in the test image. At step 310 a signal representing the intensity recorded by the analysis device (i.e. the device-specific intensity as actually visible on the display device 104) is output to the computer 104 and stored and associated with a value identifying the greyscale that was being displayed. This can be achieved using any suitable data structure, e.g. a look- up table.

At step 312 a question is asked whether there are more greyscales to display as part of the test image process. In alternative embodiments, each test image may include more than one grey shade and the analysis device(s) may be configured to sense different shades. If the answer to the question is negative then the process terminates at step 314, otherwise at step 316 the next greyscale in the test image process/sequence is selected and control passes back to step 306. This process records the light intensity (or colour gamut) of the displayed test pattern and this information is then used to calculate the best 256 shades of PseudoGrey on the specific display device being used instead of standard pure grey shades in image data, as shown in Figure 5. It will be understood that the process can be altered for use with display devices other than 8-bit monitors.

Turning to Figure 5, operations performed during step 204 of Figure 2 are shown. The process starts at step 500 and a search is carried out at step 502 for the minimum and maximum intensity values that were recorded during step

308/310 described above. Next, at step 504 a range of greyscale values between the identified minimum and maximum intensity values is calculated. Given that an 8-bit RGB signal can produce 256 shades of grey, there are many more shades of grey available via the PseudoGrey sequence (or other off-shade sequences) than the conventional R=G=B pure grey shades. Hence, it is possible to select a sub-set of the PseudoGrey set of grey shades to produce an intensity gradient that best matches the range of grey shades displayable by the display device 104 (under the present ambient light conditions), as found during step 202. The range can be in the form of a pure linear curve (a straight line), in which case the 256 PseudoGreys can be selected for their linear increase intensity to produce 256 equal increments. Alternatively, the grey shades in the range may be selected to produce a DICOM curve with Just-Noticeable Differences between the minimum and maximum intensities that the monitor and computer produced. In yet another alternative embodiment the brightest and darkest shades could be identified. It is then possible to search, e.g. a binary search, through the grey range to find the colour combination required to produce a desired intensity rather than linearly displaying and testing each available grey. At step 506 the first greyscale in the calculated range is selected and at step 508 a search is carried out amongst the stored device-specific greyscale values for the one that best matches the first greyscale in the range, e.g. the one that has an intensity value identical or closest to it. At step 510 data associating the first greyscale in the range with the closest matching device- specific greyscale value is stored, e.g. in a look-up table. At step 512 a question is asked whether all the greyscales in the range have been processed. If the answer is positive then the process ends at step 514, otherwise at step 516 the next greyscale in the range is selected and control is passed back to step 508.

Given the calibration information, namely a mapping from grey scale value (range 0...255) to an equivalent PseudoGrey shade that produces the desired intensity gradient, it is possible to "correct" greyscale imagery and an algorithm for achieving this is presented in Figure 6 (which illustrates operations performed during step 206 of Figure 2). The process starts at step 600, where image data (which can be in any suitable bitmap format) is received, e.g. from a file. The image (either in its entirety or just a selected portion that is to be displayed) is scanned pixel-by-pixel and processed in accordance with the steps of Figure 6. At step 602 data representing the pixel to be displayed is selected. At step 604 a check is performed as to whether the colour of the pixel is deemed to be pure grey shade (e.g. has R=G=B values, or within a given threshold). If so then at step 606 the device-specific greyscale value corresponding to that greyscale value, as stored in the table created at step 510, is selected and used to display the pixel. In one implementation the decode sections of UltraVNC Java source code were modified to exchange pure grey pixels with the device- specific greyscale values. If the question is answered in the negative then no change to the pixel intensity is made (step 608). The processing of the individual pixel data then ends at step 610.

Figure 7 shows schematically an example of hardware configured to implement processes described above. The computer 702 may be connected to the display device 704 by means of a signal modifier device 706, e.g. a dongle.

A (unmodified) video signal is transferred to the signal modifier from the computer, but data representing the device-specific greyscale values is also received by/stored in the modifier device, following execution of processes similar to the ones described above for the monitor 704. The modifier device can then apply a process similar to that shown in Figure 6 to correct/modify the greyscales that are displayed on the monitor 704.

The output from the modifier device can contain fractional values in addition to producing PseudoGrey-based intensities. If the display device 704 can utilise fractional steps between 256 intensity levels (such as analogue circuitry or higher bit-depth technology), then the modifier device could output, for example, 10-bit data from input 8-bit data. This would enable a 256-level display to produce a corrected display whilst retaining 256 levels of pure grey rather than using PseudoGrey. This in turn could map each red, green and blue channel to an alternative bit depth, meaning that a 24-bit colour display could be corrected to produce a 36-bit colour display corresponding to a required intensity curve, e.g. linear, DICOM, or other.

A software-based implementation of the above PseudoGrey mapping algorithm can be provided, which examines each colour entering the graphics card/display controller, altering colours where required, and then display them conventionally using the underlying graphics card on the display device. In addition to a software-based approach, hardware-based embodiments can be implemented, which can also reveals an alternative, higher-quality approach:

1. True 12-bit imaging modifier device/dongle The modifier device can be connected to the VGA/DVI socket of the PC, augmenting the video signal before passing it on to the monitor. The process operating in the dongle firmware may be as follows: i. Dongle inputs 24-bit colour ii. Dongle tests if input colour has the same intensity on red, green and blue (hence is greyscale). Alternatively, the dongle could test if input colour has a corresponding entry in a device-specific lookup table", and an optimisation for when greyscale is being processed is to check if the input colour channels are equal (R=G=B) rather than scan a table. iii. If so, dongle swaps colour with greyshade (or device-specific colour) from internal palette iv. Dongle then outputs (potentially) replaced colour

If the dongle has higher fidelity/accuracy than 8-bit, namely it can output 1 ,024 voltage steps, then the dongle is hardware-capable of 10-bit rendering (or higher, depends on what we implement). If the connected monitor is capable of displaying fractional values then the result can be true 10-bit resolution. 2. Full colour correction via dongle

If each channel (R, G, B) is calibrated in isolation then it is possible to store three respective look-up tables in the dongle: one each for red, green and blue channel. An input value in the range 0-255 (integer) is output as a voltage, but fractional values (such as 10-bit would give us steps of % rather than integer) can be introduced to enable a colour-calibrated monitor at 10-bit resolution, thereby converting a 24-bit colour monitor to display 30-bit colour (or higher), of which only 24-bits are "user-selectable". This approach is dependent on the monitor (e.g. VGA or Analogue DVI, depending on the exact implementation) being capable of displaying fractional intensities.

Alternatively (if the monitor cannot accept fractional intensities - such as digital DVI), the dongle could have multiple output ports, where the output is spread between the ports. For instance, where DVI is limited to 8 bits per R₁G₁B channel, one DVI port could contain 8 bits, whilst the other port contains the extra 2 bits per channel. This results in 24 bits on the first DVI port, and the remaining 6 bits on the second DVI port (in the case of 30 bit colour transmission). The receiving display device then combines the two signals back into 30-bit colour depth.

Further, if later display technologies are used (e.g. DisplayPort), then the dongle could convert the incoming 8 bits per channel RGB analogue (or digital) signal into a higher bit depth digital signal, which is subsequently encoded in an alternative digital format to be forwarded to the targeted display device.

3. PseudoGrey imaging dongle

Integer values in the range 0-255 for red, green blue are still output, but upon input of 24-bit colour, a check is performed for pure grey shades (red = green = blue) and if any are found then the output signal is replaced with PseudoGrey (or any desired colour pattern) from a pre-defined palette.

4. Software implementation of the PseudoGrey imaging dongle

Instead of using a dongle to process the colour transformation, it is possible to carry out the operations at the device driver level. For instance, an "intercept" driver can be produced that captures calls to the graphics driver, adjusts any colours (if necessary) and then passes on the data to the graphics card. In some cases, e.g. where a colour bitmap is provided, a local copy of the driver with the PseudoGrey (device-specific) colours may need to be made. The local copy could then be forwarded to the graphics card.

The process of Figure 3 may be performed manually or may be performed at least partially-automatically by code executing on the computer and/or a modifier device. For example, the test pattern and associated processes may be triggered daily to verify that the display system is still correctly calibrated, or may occur at more frequent intervals (e.g. every hour), or even continuously. A continuous test enables the display to react to changes in ambient light, ensuring the maximum perceived range of the display is in use.

With continuous execution of the processes described herein the monitor should effectively never go out of calibration during normal use. As the monitor ages it would be continuously adjusted. However, once the monitor is incapable (or nearing this state) of displaying a calibrated intensity (such as insufficient range of brightness), then a visual/audible warning can be produced. An alternative approach would be annual calibration (in combination with a larger, more accurate photometer), which can involve sampling 100 pixels in the centre of the monitor. Alternatively, sampling a few pixels may be considered sufficient (e.g. with a clip-on photometer) and this would be preferable because no additional hardware would be required.

Alternatively, a light sensor may be used to detect changes in ambient light and initiate the calibration processes when a change within a certain threshold is sensed. Further, the calibration may use a weighted average of calibration results in order to avoid suddenly "jumping" when a shadow drifts across the screen and so should only react to moderately static changes in lighting. Such changes can then be processed gradually to avoid a sudden onscreen flash that may distract the user.

Claims

1. A method of displaying image data on a display device, the method including: obtaining (202) a set of device-specific colour values, each said device- specific colour value in the set being associated with a corresponding non- device specific colour value; analysing (502, 504) image data representing an image to identify at least one non-device specific colour value included in the image data, and displaying (206) at least part of the image represented by the image data on the display device using the corresponding device-specific colour value instead of the identified non-device specific colour value.

2. A method according to claim 1 , wherein the device-specific colour values and the non-device specific colour values correspond to greyscales/shades of grey.

3. A method according to claim 2, wherein the device-specific colour values comprise a subset of a PseudoGrey sequence.

4. A method according to claim 2 or 3, wherein the analysing (502, 504) of the image data to identify at least one non-device specific colour value includes identifying a pure grey pixel.

5. A method according to any one of the preceding claims, further including: displaying (306) at least one colour on the display device (104), the colour being identified by a non-device specific colour value; recording (308) a device-specific colour value representing at least one characteristic of the displayed colour as recorded by a light analysis device, and storing (310) data associating the device-specific colour value with the non-device specific colour value.

6. A method according to claim 5, wherein the at least one colour comprises part of a test image display.

7. A method according to claim 5 or 6, wherein the at least one characteristic includes intensity/brightness and/or chromaticity/hue.

8. A method according to any one of claims 5 to 7, wherein the steps of displaying at least one colour, recording the device-specific colour value and storing the data associating the device-specific colour value are performed automatically at regular intervals or continuously.

9. A method according to any one of claims 6 to 8, wherein the storing of data associating the non-device specific colour value for the colour with the device-specific colour value includes: identifying (402) minimum and maximum values of said recorded device- specific colour values; calculating (404) a range of greyscale values between the minimum value and the maximum value; searching (408) the device-specific colour values for values that best match each said greyscale value in the range, and storing (410) data associating the best match device-specific colour value with the corresponding greyscale value in the range.

10. A method according to claim 9, wherein the minimum and maximum values correspond to intensity or chromaticity recorded by the light analysis device.

11. A method according to claim 9 or 10, wherein the calculated range of greyscale values comprises a pure linear curve between the minimum value and the maximum value.

12. A method of creating a set of device-specific colour values for use in displaying image data on a display device, the method including: displaying at least one colour on the display device, the colour being identified by a non-device specific colour value; recording a device-specific colour value representing at least one characteristic of the displayed colour as recorded by a light analysis device, and storing data associating the device-specific colour value with the non- device specific colour value.

13. Apparatus (706) adapted to display image data on a display device (704), the apparatus including: a component configured to obtain a set of device-specific colour values, each said device-specific colour value in the set being associated with a corresponding non-device specific colour value; a component configured to analyse image data representing an image to identify at least one non-device specific colour value included in the image data, and a component configured to modify a video signal to the display device so that the display device displays at least part of the image represented by video signal using the corresponding device-specific colour value instead of the identified non-device specific colour value.

14. Apparatus according to claim 13, wherein the apparatus comprises a modifier device (706) connected between a computer (702) and the display device.

15. Apparatus according to claim 13 or 14, including: a component configured to display at least one colour on the display device, the colour being identified by a non-device specific colour value; a light analysis device (106) configured to record a device-specific colour value representing at least one characteristic of the displayed colour, and a component configured to store data associating the device-specific colour value with the non-device specific colour value.

16. Apparatus according to claim 15, wherein the light analysis device (106) is spaced apart from the display device so that ambient light affects readings taken by the light analysis device.

17. Apparatus according to claim 15 or 16, including a plurality of the light analysis devices, each of the light analysis devices being configured to receive light from different portions (e.g. corners) of the display device.

18. Apparatus according to claim 17, wherein the apparatus is configured to process measurements taken by the plurality of light analysis devices to calculate averaged ambient light.

19. Apparatus according to claim 17 or 18, wherein the apparatus is configured to produce different device-specific colour values for different portions of the display device.

20. Apparatus according to claim 19, wherein intensity measurements taken by the plurality of light analysis devices are used to perform a linear, quadratic or other interpolation of ambient light and/or colour skew.

21. A computer program product comprising a computer readable medium, having thereon computer program code means, when the program code is loaded, to make the computer execute a method of displaying image data according to any one of claims 1 to 14.

22. A method of creating a set of device-specific colour values for use in displaying image data on a display device, the method including: displaying at least one colour using a display device; measuring a characteristic of light produced by the display device when displaying the shade; comparing the measured characteristic and a desired value for light produced when displaying the colour, and storing data associating the measured characteristic and the desired value.

23. A method according to claim 22, further including using the stored data to modify display of the at least one colour by the display device.

24. A method according to claim 22 or 23, wherein the method is repeated automatically at intervals, or continuously.

25. A method of displaying image data substantially as described herein and/or with reference to the accompanying drawings.

26. Apparatus for displaying image data substantially as described herein and/or with reference to the accompanying drawings.