GB2275854A

GB2275854A - Image analysis apparatus

Info

Publication number: GB2275854A
Application number: GB9304431A
Authority: GB
Inventors: Alan Macdonald; Ian Miller
Original assignee: International Business Machines Corp
Current assignee: International Business Machines Corp
Priority date: 1993-03-04
Filing date: 1993-03-04
Publication date: 1994-09-07
Also published as: GB9304431D0

Abstract

Image analysis apparatus comprises a two dimensional array of photodetectors (80) for receiving light from pixels of a display screen (90). The intensity values generated by the photodetectors (80) in response to incident light from the pixels are stored in a memory (40). A processor (50) determines one or more performance parameters of the display screen (90) as a function of the intensity values stored in the memory (40) and of the distance between the centres of adjacent photodetectors in the array (80). Calibration and correlation problems associated with prior art measurement techniques, such as those mentioned earlier, are avoided by mapping the spacing of the array of photodetectors onto the known geometry of the display pixel structure. Also, the apparatus can be conveniently incorporated into a handset and therefore does not require complicated positioning jigs in use. <IMAGE>

Description

IMAGE ANALYSIS APPARATUS The present invention relates to image analysis apparatus for measuring the front of screen performance of a display device.

An example of conventional image analysis apparatus for measuring front of screen performance comprises an array of fixed cameras. Each camera is equipped with its own lens. Calibration data varies from one camera to another as a function of the tolerance of the constituent components and of the position of the cameras relative to the target.

Calibration is generally performed by placing a calibration target carrying a test image in place of the normal target. A computer system under the control of computer software then analyses the detected test images to determine the spatial calibration data for each camera/lens combination.

Another example of conventional image analysis apparatus for measuring front of screen performance comprises a camera which is moveable relative to the target to select different areas of the target.

Yet another example of such apparatus comprises a moveable lens system for selectively directing different portions of the target towards a fixed camera. In both cases, the spatial calibration is again performed by placing a calibration target carrying a test image in place of the normal target, and analysing the detected test image to determine the calibration data for the camera.

Conventional image analysis tools, such as those described above, are too large for use as hand-held instruments. The user cannot therefore select different areas of the target for analysis without the use of complex and relatively slow positioning jigs.

In accordance with the present invention, there is now provided image analysis apparatus comprising: a two dimensional array of photodetectors for receiving light from pixels of a display screen; a memory for storing intensity values generated by the photodetectors in response to incident light from the pixels; and a processor adapted to determine one or more performance parameters of the display screen as a function of the intensity values stored in the memory and of the distance between the centres of adjacent photodetectors in the array.

Conventional calibration and correlation problems are avoided in accordance with the present invention by mapping the spatial geometry of the array of photodetectors onto the known pixel geometry of the display prior to measurement. Furthermore, because the apparatus of the present invention can be conveniently incorporated into a portable handset, complicated positioning jigs are not required. The apparatus can thus be used with ease by relatively unskilled personnel. The present invention thus provides fast, low cost, high resolution measurement of front of screen parameter suitable for use in high volume manufacturing environments. The processor is preferably adapted to determine one or more of the following performance parameters of the display.

a) the inner character contrast ratio of characters displayed by the display; b) the line width generated by the display; c) the spot size generated by the display; d) the spot shape produced by the display; e) the spot profile produced by the display; and, f) the beam convergence of the display.

The array of photodetectors are preferably in the form of a charge-coupled device package that can be conveniently accommodated in a handset along with the processor and the memory. It will hence be appreciated that the present invention extends to a portable instrument hand-set comprising image analysis apparatus as described in the preceding paragraph.

A preferred embodiment of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 is a block diagram of image analysis apparatus of the present invention; Figure 2 is a simplified view of a CCD array for image analysis apparatus of the present invention; Figure 3 is a flow chart illustrating a method of calibrating the image analysis apparatus of the present invention; Figure 4 is a plan view of a CRT display screen; Figure 5 is a plan view of a CRT display screen overlayed with the CCD array; Figure 6 is a flow chart illustrating convergence measurement in accordance with the present invention; Figure 7 is a flow chart illustrating inner character contrast ratio measurement in accordance with the present invention;; Figure 8 is a flow chart illustrating line width measurement in accordance with the present invention; Figure 9 is a flow chart illustrating spot size measurement in accordance with the present invention; Figure 10 is a flow chart illustrating spot shape determination in accordance with the present invention; and Figure 11 is a flow chart illustrating spot profile measurement in accordance with the present invention.

Referring first to Figure 1, an example of image analysis apparatus of the present invention comprises a data processor 10 such as a personal computer. The data processor 10 comprises a central processing unit (CPU) 50, random access memory (RAM) 40, read only storage (ROS) 30, and large capacity store 20, such as a tape streamer or hard disk drive for example, all interconnected by bus architecture 60. Bus architecture 60 is connected to a colour display under test 90 via a display adaptor 100.

It will however be appreciated from the following that the present invention can also be used to analyse the performance of monochrome displays. A monochrome charge-coupled device (CCD) array 80 is connected, via an analogue to digital convertor (ADC) 70, to the bus architecture 60 of the data processor 110. CCD array 80 is in the form of a handset connected to ADC 70 via a flexible cable. The flexible cable allows CCD array to be held in position against the screen of the display under test 90 by hand.

Referring now to Figure 2, CCD array 80 comprises a two dimensional array of photosensitive elements 85 formed on the surface of a silicon substrate. The elements 85 are arranged in columns spaced from each other by opaque aluminium register shields 87. Each element 85 has an area of 12um X 18um. In operation, a packet of electrical charge is accumulated at each element as a function of incident light intensity and exposure time. By application of control signals, each packet of charge can be shifted out of the array and converted into a voltage. The serial data stream produced by shifting out the charge packets is clocked into ADC 70 to generate a digital representation of the image incident on array 80.

One way of calibrating the apparatus according to the present invention will now be described with reference to Figure 3. At step 200, CPU 50, under the control of computer program code stored in RAM 40 and ROS 30, requests type information identifying display 90. At step 210, the information may be manually entered by the operator via a keyboard or keypad (not shown) for example. Alternatively, at step 20, the type information may be obtained automatically from, for example, a bar code fixed to display 90. Central processing unit 50 uses the type information to retrieve display specifications corresponding to display 90 from a data base pre-stored in large capacity store 20.

Referring to Figure 4, the display specifications include the geometry and pitch of the perforations in the shadowmask of the CRT in display under test 90. The geometry and pitch of the perforations define the distance X between any one of the pixels, Red, Green or Blue, on the screen and its nearest neighbour of the same colour.

Referring back to Figure 3, at step 220, CPU 50 configures display adaptor 100 to generate video signals for producing a test pattern in the form of a green block on the screen of display 90 as a function of the received display specifications. At step 230, CCD array is placed against the screen to detect the test pattern. At step 230, the test pattern is detected by CCD array 80. At step 240, the output of CCD array 80 is digitised by ADC 70. CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60.

By analysing the output data thus stored in RAM 40 at step 250, CPU 50 determines which of the elements 85 of CCD array 80 correspond to the centroids of adjacent illuminated pixels along a pixel row of the display screen. At step 260, CPU 50 calculates geometrically the actual distance between the illuminated pixels from the shadowmask type and pitch in the display specification. By dividing the actual distance by the number of CCD elements along a row of the array between those corresponding to the adjacent illuminated pixels, CPU 50 can calibrate the apparatus for measuring distances in terms of CCD elements. It will be appreciated that in other embodiments of the present invention, the apparatus may be calibrated in this manner from a test image including either a red block or a blue block.

If convergence measurements are required, at step 270, CPU 50 predicts, in terms of CCD elements, the positions of the red and blue pixels as a function of the shadowmask type and pitch and the detected positions of the green pixels. CPU 50 stores the resulting map of pixel positions and colours into RAM. The calibration routine then terminates at #2. If convergence measurements are not required, the calibration routine terminates at #1.

Referring to Figure 5, if there is no rotational error between CCD array 80 and the screen, the centroids of each row of pixels correspond to a row of elements of the CCD array 80. However, if CCD array 80 is skewed relative to the screen, the centroids of one row of pixels correspond to elements of different rows of CCD array 80. For small angles a (typically less than 10 degrees) of skew, the error can be corrected by CPU 50 from the known dimensions of CCD array 80. If the angle of skew is too large for correction, CPU 50 instructs the operator to rotate the array to reduce the error.

Referring now to Figure 6, to measure the convergence of the display under test 90, at step 300, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of a white horizontal line on the screen of display under test 90. As before during calibration, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 310, CPU 50 determines the vertical axis centroids of the Red, Green and Blue components of the detected horizontal line. At step 320, CPU 50 compares the vertical axis centroids determined at step 310 with the pixel map stored in RAM 40 and stores the differences in RAM 40.The differences indicate the vertical mis-convergence of display under test 90.

At step 340, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of a white vertical line on the screen of display under test 90. Again, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 350, CPU 50 determines the horizontal axis centroids of the Red, Green and Blue components of the detected horizontal line. At step 360, CPU 50 compares the horizontal axis centroids determined at step 310 with the pixel map stored in RAM 40 and stores the differences in RAM 40. The differences indicate the horizontal mis-convergence of display under test 90. The vertical and horizontal misconvergences are displayed at step 370.

Referring now to Figure 7, to measure the vertical inner character contrast ratio of the display under test, at step 400, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of the character "e" in white on the screen of display under test 90. As before during calibration, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 410, CPU 50 determines the vertical luminence profile of the detected character as a function of the digitised output from ADC 70 stored in RAM 40. At step 420, CPU 50 calculates the vertical contrast ratio from the maximum and minimum values in the vertical luminence profile.

To measure the horizontal inner character contrast ratio, at step 430, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of the character "m" in white on the screen of display under test 90. Again, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 440, CPU 50 determines the horizontal luminence profile of the detected character as a function of the digitised output from ADC 70 stored in RAM 40. At step 450, CPU 50 calculates the horizontal contrast ratio from the maximum ind minimum glues in the horizontal luminence profile.The vertical and horizontal inner character contrast ratios are displayed at step 370.

It will be appreciated that, if display under test 90 were replaced by a monochrome display, the vertical and horizontal inner character contrast ratios can be measured in accordance with the present invention from the same characters, "m" and "e" displayed in the colour of the screen phosphor. It will also be appreciated that, the above-mentioned vertical and horizontal inner character contrast ratio measurements can be repeated at different points on the screen to produce averaged values.

It will further be appreciated that, in other embodiment of the present invention, characters with strokes similar to "m" and "e" may be used to effect the above-mentioned measurements.

Referring to Figure 8, to measure the vertical and horizontal line width of the display under test, at step 500, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of a horizontal line in white on the screen of display under test 90. As during calibration, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 510, CPU 50 determines the vertical luminence profile of the detected line as a function of the digitised output from ADC 70 stored in RAM 40. At step 520, CPU 50 determines, from the vertical luminence profile, the horizontal line width in terms of CCD elements at a predefined percentage of the peak luminence.

At step 530, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of a vertical line in white on the screen of display under test 90. As during calibration, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output ofbADC 70 into RAM 40 via bus architecture 60. At step 540, CPU 50 determines the horizontal luminence profile of the detected line as a function of the digitised output from ADC 70 stored in RAM 40. At step 550, CPU 50 determines, from the horizontal luminence profile, the vertical line width in terms of CCD elements at a predefined percentage of the peak luminence.At step 370, CPU 50 converts the vertical and horizontal line width measurements from CCD elements into appropriate metric units for display to the operator.

Referring now to Figure 9, to measure the spot size of the display under test 90, at step 600, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of a white spot on the screen of display under test 90. As during calibration, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 610, CPU 50 determines the horizontal luminence profile of the detected spot as a function of the digitised output from ADC 70 stored in RAM 40. At step 620, CPU 50 determines from the horizontal luminence profile the spot width X in terms of CCD elements at a predefined percentage of the peak luminence.At step 630, CPU 50 determines the vertical luminence profile of the detected spot as a function of the digitised output from ADC 70 stored in RAM 40. At step 640, CPU 50 determines, from the vertical luminence profile, the spot height Y in terms of CCD elements at a predefined percentage of the peak luminence. At step 650, CPU 50 calculates the overall spot size by applying Pythagoras theorem to X and Y. At step 660, CPU 50 converts the calculated overall spot size from CCD elements into appropriate metric units for display to the operator.

Referring now to Figure 10, to measure the spot shape of display under test 90, at step 700, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of a white spot on the screen of display under test 90. As during calibration, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 710, CPU 50 determines from the digitised output of ADC 70 stored in RAM 40, a luminence contour of the detected spot in terms of CCD elements at a predefined percentage of the peak luminence.At step 720, CPU 50 determines the major and minor axes of the luminence contour and calculates the ellipticity of the luminence contour as a function of the major and minor axes. At step 730, CPU 50 converts the calculated numerical values from CCD elements into appropriate metric units for display to the operator. At step 740, CPU 50 compiles a pictorial representation of the spot for display to the operator.

Referring now to Figure 11, to measure the spot profile of display under test 90, at step 800, CPU 50 configures display adaptor 100 to generate video signals for generating a test pattern in the form of a white spot on the screen of display under test 90. As during calibration, CCD array 80 is placed against the screen to detect the test pattern, the test pattern is detected by CCD array 80, the output of CCD array 80 is digitised by ADC 70, and CPU 50 reads the output of ADC 70 into RAM 40 via bus architecture 60. At step 810, CPU 50 determines from the digitised output of ADC 70 stored in RAM 40, a luminence contour of the detected spot in terms of CCD elements at a predefined percentage of the peak luminence.Depending on a selection made by the operator at step 820, CPU 50 either, at step 840, generates a representation of the spot profile in the form of a contour map as a function of the luminence contour for display to the operator, or, at step 830, generates a pseudo three dimensional representation of the spot profile as function of the luminence contour for display to the operator.

In the preferred embodiment of the present invention hereinbefore described CPU 50, RAM 40, and ROS 30 are part of a data processor such as a personal computer and are connected to the CCD array handset 80 via a flexible communication cable. However, in other embodiments of the present invention, CCD array handset 80 is in the form of a remote device incorporating CPU 50, RAM 40 and ROS 30. In such embodiments of the present invention, the remote device is provided with a port for connection to a host data processor to permit display specifications to be loaded as required into RAM 40 from a database stored in a large capacity store of the host data processor. The remote device is provided with a keypad permit the operator to select measurement functions and a display panel for displaying resulting measurements.

Claims

1. Image analysis apparatus comprising: a two dimensional array of photodetectors (80) for receiving light from pixels of a display screen (90); a memory (40) for storing intensity values generated by the photodetectors (80) in response to incident light from the pixels; and a processor (50) adapted to determine one or more performance parameters of the display screen (90) as a function of the intensity values stored in the memory (40) and of the distance between the centres of adjacent photodetectors in the array (80).

2. Apparatus as claimed in claim 1, wherein the processor (50) is adapted to determine the inner character contrast ratio of characters displayed by the display.

3. Apparatus as claimed in claim 1, wherein the processor (50) is adapted to determine the line width generated by the display.

4. Apparatus as claimed in claim 1, wherein the processor (50) is adapted to determine the spot size generated by the display.

5. Apparatus as claimed in claim 1, wherein the processor (50) is adapted to determine the spot shape produced by the display.

6. Apparatus as claimed in claim 1, wherein the processor (50) is adapted to determine the spot profile produced by the display.

7. Apparatus as claimed in claim 1, wherein the processor (50) is adapted to determine the convergence of the display.

8. Apparatus as claimed in any preceding claim, wherein the array of photodetectors comprising a charge-coupled device.

9. Apparatus as claimed in any preceding claim, comprising a handset encasing the processor, the array of photodetectors, and the memory.

10. Apparatus substantially as hereinbefore described with reference to the accompanying drawings.

11. A method of measuring a performance parameter of an image displayed on a display screen, the method comprising: superimposing a two dimensional array of photodetectors on the screen to receive light from pixels of the screen; storing intensity values generated by the photodetectors in response to incident light from the pixels; and determining the performance parameter as a function of the intensity values stored in the memory and of the distance between the centres of adjacent photodetectors in the array.