WO2016207703A1

WO2016207703A1 - Apparatus and method for detection of pixel or sub-pixel functional defects of an image display

Info

Publication number: WO2016207703A1
Application number: PCT/IB2015/054863
Authority: WO
Inventors: Flávio Pedro GONÇALVES FERNANDES FERREIRA; Boris Paul JEAN BRET; Paulo Manuel FIGUEIRAS FORTE; Paulo Eduardo REIS FELGUEIRAS; Eduardo Jorge NUNES PEREIRA; Michael SCOTT BELSLEY
Original assignee: Bosch Car Multimedia Portugal, S.A.; Universidade Do Minho
Priority date: 2015-06-23
Filing date: 2015-06-29
Publication date: 2016-12-29

Abstract

Method, and apparatus therefor, for detection of display pixel or sub-pixel functional defects of an image display, comprising the steps: outputting a sequence of image test patterns to the image display, wherein each image test pattern comprises image test pattern elements; inputting the images of a test area of the image display through an optical system; defocusing the images in said optical system; acquiring images of the image test patterns by an image sensor optically coupled to the output of said optical system; and processing the images for detecting bright or dark defects with an electronic data processor, wherein the point spread function full width at half maximum of the optical system is smaller than the shortest distance between the images of any two elements of a test pattern, and the point spread function width of the optical system is substantially larger than the pixel pitch of the image sensor.

Description

D E S C R I P T I O N

APPARATUS AND METHOD FOR DETECTION OF PIXEL OR SUB-PIXEL

FUNCTIONAL DEFECTS OF AN IMAGE DISPLAY

Technical field

[0001] The present disclosure relates to a testing system for detecting functional defects, in particular sub-pixel defects, of pixel elements in image displays including, but not being limited to, liquid crystal displays, plasma displays and organic light emitting diode displays.

Background Art

[0002] The increased higher resolution of image displays being developed, together with a stringent demand of the market for the quality standards of these displays, requires ever more sophisticated methods to detect and quantify functional defects at a pixel or even sub-pixel scale. The main problem with an optical inspection of a high resolution display is the difficulty to sample adequately the grid structure of a test display by means of a digital camera. According to the Nyquist criterion, an inspecting monochromatic camera must have a linear resolution at least 6 times the linear resolution of the tested display in order to access its three sub-pixel elements. This problem is aggravated by Moire effects that persist above such limit, since the bandwidth of a sharply focused image of a display is not limited by its resolution. To deal with the aforementioned problem, several methods were disclosed, namely:

[0003] Document U.S. No. 8,428,334 (issued Apr. 23, 2013) describes an automatic optical inspection based on a scanner that scans all pixels in a LCD display, giving a detailed description of the mathematical procedure to identify dark dots and bright dots from thresholds relative to neighbour pixels.

[0004] Document U.S. No. 7,308,157 (issued Dec. 11, 2007) describes an automatic optical inspection system that is based on several high resolution cameras and processing to detect angle misalignment. The optical solution is very expensive and processing is computationally intensive. It is mainly interesting because it categorizes types of defects, cosmetic and functional. It also mentions sub-pixel (colour) resolution, and fractional defects (defects in portions of sub-pixels of an LCD panel). Explicitly mentions at least 4 pixels in inspection CCD are required per single pixel in display under inspection and also mentions the drawback of moire patterns that can make defect detection impossible.

[0005] Document U.S. No. 5,966,458 (issued Oct 12, 1999) aims at overcoming the limitations of using optical defocusing to reduce moire patterns when a display is imaged on an image sensor. This document advocates activating a portion of a grey image of a display and proposes a digital compressing technique to process the image, while maintaining a higher resolution in the CCD sensor compared to the display resolution.

[0006] The abovementioned documents embody the three main methods of current technology aimed to sample a display at a sub-pixel scale, in that order:

• scanning the display with a single movable inspecting camera; sampling the display with a multiplicity of fixed cameras; imaging the display with a higher resolution camera.

[0007] The main disadvantages of the aforesaid methods are, respectively:

• non robustness, since movable mechanical parts are required; expensiveness, which is proportional to the number of cameras;

• unscalability, as the number of display pixels increases.

[0008] These facts are disclosed in order to illustrate the technical problem addressed by the present disclosure.

General Description

[0009] The disclosure describes an apparatus and method for the optical inspection of functional defects, in particular of R, G and B sub-pixel elements, of digital image displays. The technical features of the invention aim to overcome the aliasing problem permitting the imaging of a given display with an image sensor that does not necessarily have sufficient resolution to sample the display grid structure. Herein, alternatives and preferences of the apparatus and method of the present invention are summarized.

[0010] The apparatus comprises a test display, a digital image sensor, an optical system to transfer images from the test display to the digital image sensor and a computer system to control the test display and to analyse the digital images. In a preferred embodiment of the invention, the optical system includes an RGB colour filter facing the image sensor in order to produce RGB digital images.

[0011] The method for detecting functional defects by means of the aforesaid apparatus is divided into three steps according to an embodiment:

1. The display is addressed with test patterns, each pattern comprising a sub-set of the display grid and each pattern element being addressed with a high greyscale level, while the remaining display is addressed with a low greyscale level. In the preferred embodiment of the method, the test patterns comprise pixel elements equally spaced in both the horizontal and vertical directions. In an alternative embodiment of the invention, the elements of the test patterns are sub-pixels of one of the R, G or B classes;

2. A defocused image of each addressing pattern is transferred by means of the optical system to the image sensor. In a preferred embodiment of the invention, the point spread function width of the optical system is smaller than the period of the test patterns but substantially larger than the pixel pitch of the image sensor. In particular, the pixel pitch of the image sensor is at least 2 times smaller than the point spread function width of the optical system;

3. The digital images corresponding to the defocused optical images of the display test patterns are analysed. Dark defects can be detected by a local lowering of digital values, while bright defects can be detected by a similar lowering after subtraction of an image of the display with all its pixels addressed at the lowest grey value. In an embodiment of the invention, digital values are values of a RGB triplet obtained by a linear transformation between the optical system colour space and the display colour space and defects are detected by Fourier analysis of the perturbations on the spatial regularity of the images.

[0012] It is disclosed a method for detection of display pixel or sub-pixel functional defects of an image display, comprising the steps of:

outputting a sequence of image test patterns to the image display, wherein each image test pattern comprises image test pattern elements;

inputting the images of a test area of the image display through an optical system; defocusing the images in said optical system;

acquiring images of the image test patterns by an image sensor optically coupled to the output of said optical system; and

processing the images for detecting bright or dark defects with an electronic data processor,

wherein the full width at half maximum of the point spread function of the optical system is smaller than the shortest distance between the images of any two image test pattern elements of an image test pattern, and

the point spread function width of the optical system is substantially larger than the pixel pitch of the image sensor.

[0013] It is also disclosed an apparatus for detection of display pixel or sub-pixel functional defects of an image display, comprising:

an optical system for inputting the image of a test area of the image display;

an image sensor optically coupled to the output of said optical system;

an electronic data processor arranged for outputting a sequence of image test patterns to the image display and for processing the images acquired by the image sensor of the image test patterns for detecting bright or dark defects;

wherein the optical system is a defocusing optical system and the image test pattern has spatial periodicity over the test area of the image display .

[0014] In an embodiment, each image test pattern of the test area comprises test elements which are a sparse sub-set of the display elements within the test area having a first intensity level, and

remaining display elements of the test area which have a second intensity level wherein the first intensity level is different than the second intensity . [0015] In an embodiment, the first intensity level is the highest intensity level of the display and the second intensity level is the lowest intensity level of the display.

[0016] In an embodiment, the first intensity level is the highest greyscale level of the display and the second intensity level is the lowest greyscale level of the display..

[0017] In an embodiment, each test pattern is a regular pattern of said test elements, suitably spaced, both in the horizontal and vertical directions, by a fixed multiple of the display pixel pitch.

[0018] In an embodiment, each said element is a pixel.

[0019] In alternative embodiment, each said element is a sub-pixel element (i.e. a dot).

[0020] In an embodiment, the sequence of image test patterns is such that the full pixel grid of the test area is addressed with both first and second intensity levels.

[0021] In an embodiment, the display test area is the full display area.

[0022] In an embodiment, the optical system is arranged such that:

the optical point spread function width of the optical system is smaller than the spatial period of the image test patterns; and the point spread function width of the optical system is substantially larger than the pixel pitch of the image sensor.

[0023] In an embodiment, the point spread function width of the optical system is at least larger than 2x2 pixels of the image sensor.

[0024] In an embodiment, the defocusing optical system comprises an objective lens placed in front of the image sensor arranged such that it produces out-of focus images of the test area to the image sensor with a predefined point spread function width.

[0025] In an embodiment, the image sensor is a colour sensor and the image test patterns have equal intensity levels for the sub-pixel elements of the display test area.

[0026] In an embodiment, the image sensor is a colour sensor and the data processor is arranged to carry out a colour transformation of the acquired images,

such that the spectral crosstalk between colour channels is eliminated. [0027] In an embodiment, the image sensor is monochrome and the optical system includes a colour filter for acquiring sensor images separating, spatially or temporally, the individual colour image components.

[0028] In an embodiment, the colour filter is a colour filter wheel, a liquid crystal tuneable filter, or a Bayer filter.

[0029] In an embodiment, the image sensor is monochrome and data processor is arranged that the image test patterns activate sequentially the sub-pixels of each colour of the display test area.

[0030] In an embodiment, the processing of the acquired digital images for detecting a dark defect comprises detecting a local intensity decrease in the images acquired by the image sensor of the image test patterns.

[0031] In an embodiment, the processing of the acquired digital images for detecting a bright defect comprises detecting a local intensity lowering of the images acquired by the image sensor of the image test patterns after subtraction of an image acquired by the image sensor of the display test area with all its pixels having the second intensity value.

[0032] In an embodiment, the data processor is arranged to detect perturbations in the spatial periodicity of the acquired images for detecting bright or dark defects.

[0033] In an embodiment, detecting perturbations in the periodicity of the acquired images for detecting bright or dark defects comprises Fourier filtering the acquired images for removing the spatial periodicity of the test patterns and subsequently applying a detection threshold.

Brief description of the drawings

[0034] The following figures provide preferred embodiments for illustrating the description and should not be seen as limiting the scope of invention.

[0035] Fig. 1 is a schematic diagram of the apparatus wherein, and according to an embodiment of the invention, the optical system may contain RGB filters. [0036] Fig. 2 illustrates the relationship between a regular test pattern of pixels addressing the display (2a) and its unfocused optical image (2b), according to the preferred embodiment of the invention.

[0037] Fig. 3 illustrates the relationship between a colour channel of a given test pattern addressing a defected display and its unfocused optical image, and the detection of dark defects coincident with any element of the test pattern.

[0038] Fig. 4 illustrates the relationship between a colour channel of a display having bright defects and its unfocused optical image when all pixels of the display are addressed at the lowest greyscale level.

[0039] Fig. 5 illustrates the detection of bright defects coincident with any element of the test pattern addressing the display.

[0040] Fig. 6 is a schematic diagram of the image processing algorithm for measuring dark and bright defects, according to the preferred embodiment of the invention.

[0041] Fig. 7 shows an example of raw images of equal sized sub-areas of a test display containing dark defects and bright defects.

[0042] Fig. 8 demonstrates the detection of dark defects within an area of a test display and according to the preferred embodiment of the invention.

[0043] Fig. 9 demonstrates an example of the detection of bright defects within an area of a test display and according to the preferred embodiment of the invention.

Detailed Description

[0044] The present invention refers to an apparatus and method for the analysis of functional pixel or sub-pixel defects of any display device containing pixel elements or sub-pixel elements including, but not being limited to, LCD devices. Hereafter embodiments of the present invention are described in detail; wherein like parts of the invention are designated by like numbers in the accompanying drawings.

[0045] The invention includes an apparatus, shown schematically on Fig. 1, comprising a test display 11 to be inspected, an image sensor 13, an optical system 12 to transfer images of the test display 11 to the image sensor 13, a digital camera interface 14 to acquire the images captured by the image sensor 13, a display adapter 16 to address the test display 11, and a digital computer 15 to process images received from the digital camera interface 14 and to generate test patterns to be sent to the display adapter 16. The area of test display 11 captured by the image sensor 13 may be a fraction or the whole of the test display 11 area. The optical system 12 may comprise lenses, prisms, mirrors, apertures, filters, or any other convenient optical devices that, in combination, transfer an optical image of the test area of the test display 11 onto the image sensor 13.

[0046] In the preferred embodiment of the apparatus, the optical system 12 contains a colour filter set, such as a Bayer mosaic against the image sensor 13, for capturing digital images in the red, green, and blue (RGB) wavelength regions.

[0047] In a preferred embodiment, the test area captured by the image sensor 13 comprises the entire area of the test display 11 and the major axis of the image sensor 13 grid is rotationally aligned to the major axis of the image of the test display 11 grid.

[0048] The disclosure includes a method to address the test display with test patterns, wherein each test pattern comprises a sparse sub-set of the display elements within the display test area, and wherein each test element may be any combination of sub- pixels, addressed with a fixed high greyscale level, while the remaining display elements within the test area are addressed with a fixed low greyscale level. Displaying sequentially the whole set of patterns the full pixel grid of the display test area can be scanned. Fig. 2a illustrates a region of a display addressed with an instance of a regular test pattern, according to the preferred embodiment of the present invention.

[0049] In this embodiment of the addressing method, the test display 11 is addressed with a test pattern which is a sub-grid of the full display grid, comprising a rectangular periodic grid of pixels 21, all being addressed with the highest greyscale level, while the remaining elements of the display are addressed with the lowest greyscale level 22. In this embodiment, by shifting the test patterns, vertically and horizontally, all the display elements can be activated after a number of iterations equal to the square of the test pattern period. Fig. 2a shows an instance of this embodiment wherein the period of the test pattern is the double of the pixel pitch. In this instance, the addressed part of the display contains 4 times less elements than the original display.

[0050] The invention includes an imaging method to transfer defocused optical images of the test display area to the image sensor. Herein, defocusing is any optical means of varying the point spread function (PSF) width of the optical system 12. Defocusing allows correcting the aliasing in the imaging of the test pattern addressing the display by varying the number of pixels in the image sensor 13 detecting the optical image spot corresponding to each element of the test pattern addressing the display. A spot is a region of the image sensor corresponding to the point spread function of one pixel of the display.

[0051] In an embodiment of the imaging method, defocusing is set by adjusting an objective lens placed in front of the image sensor 13 in order to produce out-of focus images of the test display 11.

[0052] Fig. 2b illustrates the optical image of the pattern shown on Fig. 2a that is incident on the imaging sensor, whereby the optical PSF 23 width of the optical system is larger than the image sensor 13 pixel pitch. In the illustration of Fig. 2b, and according to the preferred embodiment of the invention, the PSF 23 width matches the test pattern period of the regular test pattern addressing the test display. For the instance of Fig. 2b, the defocused image of the display addressed by the regular pattern imitates an active display with a 2-fold reduced linear spatial resolution.

[0053] The disclosure includes a detection method to measure functional defects of the test display 11, whereby digital values of the digital images detected by the image sensor 13 are analysed. In this method, each of the test patterns addressing the test display is converted into corresponding digital spot patterns, each spot being a digital image of the corresponding element of the test pattern after convolution with the PSF 23 of the optical system.

[0054] In the preferred embodiment of the detection method, the digital images comprise three RGB channels whose digital values are one of the three elements of a RGB triplet obtained by means of a colour transformation matrix from the optical system RGB colour space into the display RGB colour space. This transformation eliminates the spectral crosstalk between colour channels, allowing the exact measuring of the light intensity emitted by each sub-pixel class of the test display. The colour transformation matrix can be calibrated as follows. Each RGB channel of a portion of the display is turned on separately and the RGB triplet of average digital values detected by the image sensor is recorded in a corresponding matrix column. The inverse of the resulting matrix is the colour transformation matrix.

[0055] Fig. 3a illustrates a region of a defected display, wherein one of the channels, in the instance the green channel, is addressed with a regular test pattern and contains two adjacent dark defects, 31 and 32, and two adjacent bright defects, 33 and 34. By definition, only the dark defect 32 coincident with an element of the test pattern and only the bright defect 34 not coincident with any element of the test pattern are visible, compared to the image capture from a non-defected display. Fig. 3b shows the optical image, incident on the image sensor 13, corresponding to the pattern shown on Fig. 3a. The dark defect 32 coincident with an element of the test pattern can be detected by a decrease of the peak or integrated digital values in digital image spot corresponding to the optical image spot 35, relative to the corresponding digital values for a non-defected display. However, the bright defect 33 coincident with an element of the test pattern cannot be directly measured and, furthermore, is confused with the neighbouring bright defect 34 lying in the dark area of the test pattern, since the distance between the two bright defects, 33 and 34, is smaller than the PSF width of the optical image.

[0056] In order to detect a bright defect coincident with an element of the test pattern, while not being perturbed by a neighbour bright defect in the dark region of the pattern, the invention includes a subtraction method described hereafter.

[0057] Fig. 4a illustrates the same region of the display depicted on Fig. 3a with all the pixels of the display addressed with the lowest greyscale value, wherein the bright defects, 41 are visible. Fig. 4b shows the corresponding optical image, incident on the image sensor, wherein the adjacent bright defect are imaged as two overlapping spots 42, since their distance is smaller than the PSF width of the image. Hereon, the image obtained with all elements of the display addressed at the lowest greyscale value is called «black image». Fig. 5 shows the result of subtracting the black image of Fig. 4b to the image of the pattern addressed display of Fig. 3b. The resulting image shows a lowering of the intensity within the area 51 corresponding to the bright defect coincident with an element of the test pattern, which was not present before the subtraction, therefore allowing the detection of the bright defect.

[0058] A method for defect detection is shown schematically on Fig. 6, whereby defects are measured as perturbations in the periodicity of the digital images corresponding to periodic test patterns addressing the display. In the preferred embodiment of the method, digital images are analysed by Fourier filtering, using multi-band notch filters, in order to eliminate the periodic structure of the images while revealing the perturbations corresponding to defects; dark defects are negative perturbations. Applying the same analysis to the images subtracted beforehand by the above mentioned black image of the display, one obtains a differential perturbation; bright defects are measured by subtracting the previously detected dark defects from this differential perturbation.

[0059] Hereafter, experimental examples of the detection of dark defects and bright defects in a real display are presented. In accordance with the preferred embodiment of the invention, the experimental apparatus comprised an imaging camera consisting of a colour image sensor with an objective lens of 25mm focal length and a test display addressed through a display video adapter. Also according to an embodiment of the invention, the method consisted in addressing the test display with a set of regular (or periodic) pixel patterns, capturing defocused images of said patterns and Fourier corresponding digital images by means of digital notch filters, after a previous colour transformation to separate the exact RGB components emitted by the test display.

[0060] It will be appreciated that unless otherwise indicated herein, the particular sequence of steps described is illustrative only and can be varied without departing from the disclosure. Thus, unless otherwise stated the steps described are so unordered meaning that, when possible, the steps can be performed in any convenient or desirable order.

[0061] The test display was an LCD having a resolution of 1920*720 RGB pixels and the image sensor was a colour CCD. The full display was imaged onto 1344*504 pixels of the sensor. This full-field imaging of the display corresponded to 0.7 pixel of the camera per pixel of the display - an insufficient spatial resolution to resolve two dots of the same colour at adjacent pixels. Therefore, neither dark defects could be seen nor could joint bright defects be resolved before the application of the method of the present invention.

[0062] In the following discussion any of the R, G, or B sub-pixel elements of the display is named a dot and its functional defect classified either as a dark dot or a bright dot. Defects which appear half bright or half dark are referred as fractional and defects of the same type occurring in adjacent pixels are described as joint. In the experiment, both dark dots and bright dots of a particular colour channel were emulated by imposing an arbitrary addressing level to the corresponding dots of the display.

[0063] Fig. 7a shows a focussed region of the test display comprising 90*70 display pixels, addressed uniformly in white, where the presence of green dark dots is not visible due to the insufficient resolution of the camera; Fig. 7b shows an equal sized region of the display, addressed uniformly in black, having a green bright dot 71 and two joint fractional bright green dots 72; although bright dots can be seen directly, joint bright dots 72 cannot be told apart, since the PSF of the imaging system at the object plane is larger than the pixel pitch of the display.

[0064] Fig. 8 illustrates the detection of dark dots by applying the method of an embodiment of the disclosure. The test patterns consisted of periodic grids of white pixels on a black background, with a period of 11 display pixels in both horizontal and vertical directions, making a total of 11*11=121 test patterns in order to scan the whole display. Fig. 8a shows the defocused image of the same region of the display shown on Fig. 7a, with the display addressed with one of said test patterns. The presence of two green dark dots showed up as two slightly dark magenta spots in an otherwise regular grid of white spots, each spot having the width of the PSF of the imaging system. The defocusing of the objective was set in a way that the PSF width matched the grid spacing, maximizing the sampling of each spot while avoiding the overlap of neighbouring spots. Each spot of the image was sampled by about 64 pixels of the image sensor. Fig. 8b shows the result of the application of a notch Fourier filter to the green component of the image of Fig. 8a, after colour transformation from the camera RGB space to the display RGB space. Fourier filtering allows isolating the dark dots from the surrounding grid of spots. In this case one can see the detection of a dark dot 81 and also the detection of a fractional dark dot 82.

[0065] Fig. 9 illustrates the detection of bright dots, including discriminating joint bright dots, by applying the method of an embodiment of the disclosure. As earlier described in the discussion of Fig. 6, the method for detecting bright dots is similar to the one used for detecting dark dots, except for the following: each image of any test pattern is subtracted beforehand by a black image obtained with the display addressed uniformly in black, or the lowest greyscale level, and the resulting Fourier filtered images are subtracted by the previously detected dark dots. The area of analysis of the display was the one shown on Fig. 7b. Fig. 9a shows the detection of a bright dot 91 and a fractional bright dot 92, both defects coincident with some element of the test pattern. However, as mentioned in the description of Fig. 7b, there is a joint bright defect next to 92. Fig. 9b shows the result obtained when processing the images obtained with the test pattern shifted one pixel to the side. The fractional joint bright dot 93 is revealed.

[0066] In conclusion, the presented experimental results confirm the ability of the apparatus and method of the embodiments of the disclosure to detect, localize and measure both dark and bright dots by means of a camera having an insufficient resolution to sample the grid of dots of a test display. It is noticeable that the same camera can be used to analyse displays with arbitrary resolutions albeit the number of test patterns, and the corresponding processing time, scales with the square of the pattern spacing. In order for defects to be detectable, the minimum spacing of the periodic test patterns can be estimated as follows. Assuming the typical situation of a display having a higher format ratio compared to the camera, a full view imaging preferably requires aligning their major axis. In that situation, being N_c and Nd the linear resolution along the major axis of the camera and the display, respectively, the period Δη of the test patterns should obey Δη > 2*Nd/N_c. In the illustrated case Δη > 2.7 and the minimum spacing would be 3, corresponding to 9 patterns. For the precise measurement of fractional defects one should operate above said limit. With the same inspecting camera, doubling the linear resolution of the test display implies doubling the period of the test patterns, hence quadrupling the number of test patterns in order to scan the whole grid structure of the display.

[0067] The term "comprising" whenever used in this document is intended to indicate the presence of stated features, integers, steps, components, but not to preclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

[0068] The disclosure should not be seen in any way restricted to the embodiments described and a person with ordinary skill in the art will foresee many possibilities to modifications thereof.

[0069] The above described embodiments are combinable.

[0070] The following claims further set out particular embodiments of the disclosure.

Claims

C L A I M S

1. A method for detection of display pixel or sub-pixel functional defects of an image display, comprising the steps of:

wherein the point spread function full width at half maximum of the optical system is smaller than the shortest distance between the images of any two image test pattern elements of an image test pattern, and

2. The method of the previous claim, wherein each image test pattern of the test area comprises:

test elements which are a sparse sub-set of the display elements within the test area having a first intensity level, and

remaining display elements of the test area which have a second intensity level wherein the first intensity level is different than the second intensity level.

3. The method of the previous claim, wherein each test pattern is a regular pattern of said test elements, equally spaced, both in the horizontal and vertical directions, by a fixed multiple of the display pixel pitch.

4. The method of the previous claim, wherein each said element is a pixel.

5. The method of claim 3, wherein each of said elements is a sub-pixel element.

6. The method of any of the previous claims, wherein the sequence of image test patterns is such that the full pixel grid of the test area is addressed with both first and second intensity levels.

7. The method of any of the previous claims, wherein the display test area is the full display area.

8. The method of any of the previous claims, wherein the point spread function width of the optical system is at least larger than 2x2 pixels of the image sensor.

9. The method of any of the previous claims, wherein the image sensor is a colour sensor and the image test patterns have equal intensity levels for the sub-pixel elements of the display test area.

10. The method of any of the previous claims, wherein the image sensor is a colour sensor and the method comprises carrying out a colour transformation of the acquired images or a colour transformation to the image test patterns to be displayed,

such that the spectral crosstalk between colour channels is eliminated.

11. The method of any of the claims 1-8, wherein the image sensor is monochrome and the method comprises colour filtering by the optical system for acquiring sensor images separating, spatially or temporally, the individual colour image components.

12. The method of any of the claims 1-8, wherein the image sensor is monochrome and the method comprises activating sequentially the sub-pixels of each colour of the display test area.

13. The method of any of the previous claims, wherein the first intensity level is the highest greyscale level of the display and the second intensity level is the lowest greyscale level of the display.

14. The method of any of the previous claims wherein the processing of the acquired digital images for detecting a dark defect comprises detecting a local intensity decrease in the images acquired by the image sensor of the image test patterns.

15. The method of any of the previous claims wherein the processing of the acquired digital images for detecting a bright defect comprises detecting a local intensity lowering of the images acquired by the image sensor of the image test patterns after subtraction of an image acquired by the image sensor of the display test area with all its pixels having the second intensity value.

16. The method of any of the previous claims wherein the processing of the acquired digital images for detecting bright or dark defects comprises detecting perturbations in the spatial periodicity of the acquired images .

17. The method of any of the previous claims wherein detecting perturbations in the periodicity of the acquired images for detecting bright or dark defects comprises Fourier filtering the acquired images for removing the spatial periodicity of the test patterns and subsequently applying a detection threshold.

18. An apparatus for detection of display pixel or sub-pixel functional defects of an image display, comprising:

an optical system for inputting the image of a test area of the image display;

an image sensor optically coupled to the output of said optical system;

wherein the optical system is a defocusing optical system and the image test pattern has spatial periodicity over the test area of the image display.

19. The apparatus of the previous claim, wherein each image test pattern of the test area comprises

test elements which are a sparse sub-set of the display elements within the test area having a first intensity level, and remaining display elements of the test area which have a second intensity level, wherein the first intensity level is different than the second intensity level.

20. The apparatus of the previous claim, wherein each of said elements is a pixel.

21. The apparatus of claim 19, wherein each of said elements is a sub-pixel element.

22. The apparatus of any of the claims 18-21, wherein the sequence of image test patterns is such that the full pixel grid of the test area is addressed with both first and second intensity levels.

23. The apparatus of any of the claims 28-22, wherein the display test area is the full display area.

24. The apparatus of any of the claims 18-23, wherein the optical system is arranged such that:

the optical point spread function width of the optical system is smaller than the spatial period of the image test patterns; and

25. The apparatus of any of the claims 18-24, wherein the point spread function width of the optical system is at least larger than 2x2 pixels of the image sensor.

26. The apparatus of any of the claims 18-25 wherein the defocusing optical system comprises an objective lens placed in front of the image sensor arranged such that it produces out-of focus images of the test area to the image sensor with a predefined point spread function width.

27. The apparatus of any of the claims 18-26, wherein the image sensor is a colour sensor and the image test patterns have equal intensity levels for the sub-pixel elements of the display test area.

28. The apparatus of any of the claims 18-27, wherein the image sensor is a colour sensor and the data processor is arranged to carry out a colour transformation of the acquired images,

such that the spectral crosstalk between colour channels is eliminated.

29. The apparatus of any of the claims 18-28, wherein the image sensor is monochrome and the optical system includes a colour filter for acquiring sensor images separating, spatially or temporally, the individual colour image components.

30. The apparatus of the previous claim, wherein the colour filter is a colour filter wheel, a liquid crystal tuneable filter, or a Bayer filter.

31. The apparatus of any of the claims 18-26, wherein the image sensor is monochrome and data processor is arranged that the image test patterns activate sequentially the sub-pixels of each colour of the display test area.

32. The apparatus of any of the claims 18-31, wherein the first intensity level is the highest greyscale level of the display and the second intensity level is the lowest greyscale level of the display.

33. The apparatus of any of the claims 18-32 wherein the processing of the acquired digital images for detecting a dark defect comprises detecting a local intensity decrease in the images acquired by the image sensor of the image test patterns.

34. The apparatus of any of the claims 18-33 wherein the processing of the acquired digital images for detecting a bright defect comprises detecting a local intensity lowering of the images acquired by the image sensor of the image test patterns after subtraction of an image acquired by the image sensor of the display test area with all its pixels having the second intensity value.

5. The apparatus of any of the claims 18-34 wherein the data processor is arranged to detect perturbations in the spatial periodicity of the acquired images for detecting bright or dark defects.

The apparatus of any of the claims 18-35 wherein detecting perturbations in the periodicity of the acquired images for detecting bright or dark defects comprises Fourier filtering the acquired images for removing the spatial periodicity of the test patterns and subsequently applying a detection threshold.