JPH0399376A - Device for inspecting picture quality - Google Patents

Abstract

PURPOSE:To improve the quality of a display screen by allowing a tested device to output a test pattern, discriminating the 1st and 2nd images photographed by a camera and detecting the positional shift and inclination on the display screen of the tested display device. CONSTITUTION:Plural positions of respective sides are scanned at equal pitches based upon the center lines F1, F2 of a frame memory 5B and the coordinates Xi, Yi to be changed from '0' to '1' of the digital signal of the 1st image data stored in the memory 5B are detected to find out positional coordinates. Then a lighting 8 is turned off by the control of a controller 3A, a test pattern P is outputted on a display screen 1A and the 2nd image data D2 obtained by photographing the test pattern P by a camera 2 is inputted. A straight line the best fitted to the detected coordinates Xii, Yii is calculated by least square and the 2nd ID data D12 is formed to find out the inclinations phi3, phi4 of respec tive sides.

Description

[Detailed Description of the Invention] [Summary] An image quality inspection device that photographs a display device under test with a camera and checks the image quality of the display device under test using the photographed image data. The purpose of this is to improve reliability of testing and reduce testing man-hours by determining pass/fail through identification. an identification data creation unit that creates second identification data in which the test pattern is identified using the image data of the test pattern, and compares the second identification data with the first identification data as a reference; An image processing device is configured to be formed by a CPU that controls the calculation of the positional deviation and tilt of a display screen in the device and makes a quality determination, and an interface unit that outputs a signal based on the quality determination.

[Industrial Application Field] The present invention relates to an image quality inspection device that photographs a display device under test using a camera and checks the image quality of the display device under test using the photographed image data.

In recent years, with the development of information technology, the spread of computers has been remarkable, and even ordinary homes are now equipped with personal computers, and cathode ray tube displays (CRTs) are used in these personal computers.

The production volume of display devices such as liquid crystal displays (LCDs) and plasma displays (PDPs) is increasing.

Additionally, manufacturers of these display devices conduct tests to inspect the image quality of each completed display device during the manufacturing process.

Therefore, since the test for inspecting image quality will require a huge number of man-hours as the production volume increases, it is desired that the test be streamlined.

(Prior art) Conventionally, this has been carried out as shown in the conventional explanatory diagram in Fig. 8. In Fig. 8, (a) is a perspective view, (b1) is a front view of the display screen, and (b2) is a test FIG. 3 is an explanatory diagram of pattern detection.

As shown in FIG. 8(a), the display device under test 1 to be tested is loaded on the base plate 11, transported by the driven conveyor 10, and positioned at a predetermined location. For example, a cross-shaped test pattern PA shown in (b1) is output on the display screen 1A of the device 1, and the test pattern PA is visually checked from the direction of the arrow.

In this case, the cross-shaped test pattern PA is formed within the maximum display area IB shown by the dotted line on the display screen 1A formed inside the frame 4.

Therefore, as shown in (b2), the test pattern P is set with the reference horizontal level provided by the chassis (not shown) installed inside the display device under test 1 as the reference position B.
Passage or failure was determined based on the horizontality or verticality of A.

Normally, such a pass/fail judgment is made based on the predetermined length of the test pattern PA! (approximately 225 mobs), measure the left and right heights in the horizontal direction and H2 from the reference position B, and when the difference exceeds 3 mm, or the inclination θ exceeds approximately 0.8°. At the time, it was determined to be defective.

Therefore, the horizontality or verticality of the test pattern PA is visually checked, and if it is below a predetermined tolerance value, it is determined to be a good product, and if it exceeds the tolerance value, it is determined to be a defective product.

[Problem to be solved by the invention]

However, when checking by visual inspection, the inspection standards fluctuate due to eye fatigue, making it difficult to accurately measure the actual slope θ, and furthermore, judgments may vary depending on the inspection personnel performing the check. become.

Furthermore, since there is a limit to the speed of such an inspection, it requires a huge amount of man-hours and a large number of inspection personnel.

Therefore, there were problems in that the quality varied and a large number of inspection personnel were required.

Therefore, an object of the present invention is to improve test reliability and reduce test man-hours by determining pass/fail by identifying photographed image data.

[Means to solve the problem]

FIG. 1 is a diagram explaining the principle of the present invention.

As shown in FIG. 1, the frame 4 is
The test pattern P is identified using the first identification data D11 that identifies the test pattern P and the second image data D2 that identifies the test pattern P.
The identification data creation unit 5 creates the identification data 012 of The image processing device 3 is configured to include a CPU 6 that controls the calculation of the positional deviation σ and the inclination Φ and makes a quality determination, and an interface unit 7 that outputs a signal S based on the quality determination.

[Effect]

That is, the frame 4 is created using the first image data D1 obtained by photographing the frame 4 of the display device under test 1 on which the test pattern P is displayed.
The identification data creation unit 5 creates first identification data D11 that identifies the test pattern P and second identification data 012 that identifies the test pattern P based on the second image data D2 obtained by photographing the test pattern P. The second identification data 012 is compared with the data D11 as a reference, the positional deviation σ and the slope Φ of the display screen 1A in the display device under test 1 are calculated under the control of the CPU, and the quality is determined. The determination is notified by the output signal S from the interface section 7.

Therefore, compared to conventional visual checks, more accurate judgments can be made, and the checking work can be performed without requiring human labor, thereby reducing the number of testing man-hours.

〔Example〕

The present invention will be explained in detail below with reference to FIGS. 2 to 7.

FIG. 2 is a configuration diagram of an embodiment according to the present invention.

3 is an explanatory diagram of the display screen of the present invention, (a) is a front view, (b) is a side view, FIG. 4 is a flowchart of the present invention, FIG. 5 is a density histogram, and FIG. (a) (b
) are explanatory diagrams for identifying frames, (a1) (a2) (
b) and (c) show diagrams for explaining the identification of test patterns. The same reference numerals indicate the same objects throughout the figures.

As shown in FIG. 2, the display screen 1 of the display device under test 1
A camera 2 for photographing A is provided, and the first and second image data DI and D2 photographed by the camera 2 are sent to an image processing device 3.

Furthermore, a hood 9 is provided between the display device under test 1 and the camera 2 to block external light, while the image processing device 3 uses a light 8 provided in the optical system 2A of the camera 2. A controller 3A that causes blinking, an identification data creation unit 5 that identifies the first and second image data DI and D2, and an interface unit 7 that outputs a signal S that notifies pass/fail determination are connected to a bus 6^ of the CPU 6. It is made up of things that are done.

Furthermore, as shown in FIGS. 3(a) and 3(b), the display screen 1A is formed inside the frame 4 of the tapered part of the outer frame 4A indicated by diagonal lines, and a rectangular shape is formed on the display screen 1A during inspection. A test pattern P is output.

The size of the test pattern P in this case is formed to be somewhat smaller than the maximum display area IB having the height CL width C2 shown by the dotted line.

Therefore, the controller 3A controls the frame 4 to be photographed by simply turning on the illumination 8, and the test pattern P to be photographed by turning off the illumination 8 and further outputting the test pattern P to the display screen 1A. The first and second image data DI and D2 of the frame 4 and the test pattern P are captured.

Further, the identification data creation unit 5 to which the first and second image data 01.02 are inputted is configured to input the first and second image data D.
i, a camera interface 5A that performs A/D conversion of D2, and digitized first and second image data 0.1. D
2, a ROM 5C that stores the identification program, and digitized image data D.
According to the instructions of the identification program, the first image data D1 of frame 4 and the second image data D of test pattern P are
2, and a RAM 5D that stores first and second identification data D11.012 created from the first and second image data 01.02.
It is composed of.

Therefore, the image quality of the display screen 1A is checked in the order shown in FIG.

First, the illumination 8 is turned on under the control of the controller 3A, and the frame 4 is photographed by the camera 2 to obtain first image data D.
In this case, the illumination 8 is arranged so that the light irradiated by the illumination 8 is small at a flat portion of the outer frame 4 and becomes large at a tapered portion of the frame 4.

Therefore, as shown in FIG. 5, the camera interface 5A counts the number of pixels corresponding to each density of 0 to 255 gradations in the first image data DI, and creates a density histogram.

In this case, since the total number of pixels in frame 4 is known in advance,
The area indicated by the diagonal line E is calculated by integrating the area equal to the total number of pixels in frame 4 from the brightest side of the density histogram.The density t at this time is determined as the slice level, and the density t is determined as the slice level. By converting the first image data D1
is converted into a digital signal, and after conversion, the frame memory 5B
is stored.

Next, as shown in FIG. 6(a), the frame memory 5
1 on each side at equal pitch intervals based on the center lines Fl and F2 of B.
0 to 20 locations are scanned and the first image data D1 is converted into a digital signal and stored in the frame memory 5B.
The coordinate Xi changes from 0' to 1'.

Detect Yi and find the position coordinates.

Identification data D based on the coordinates Xi and Yi detected in this way
11 in the RAM 5D, and the respective coordinates Xi, Yi
The straight lines that best fit are shown below (1), (2)
The first identification data D1 is calculated by the least squares method of the formula
1 is formed as shown in (b).

However, assuming that the straight line to be sought is represented by Y = AX + 8, the slope of each side calculated in this way with respect to the straight line is 01 ~ θ
4 is calculated as shown in FIG. 6(b), and further, (3)
The average slope θ on each side can be calculated using the formula.

θ=1/4 (θ1+θ2+θ3+θ4) (3) Furthermore, by connecting the corners formed by each side with diagonal lines, the center G can be found from the intersection point.

All such calculations are performed under the control of the CPU 6.
This is performed by causing the hardware processor 5E to execute according to instructions from the arithmetic program stored in the M5G.

Furthermore, next, the lighting 8 is controlled by the controller 3A.
is turned off, the test pattern P is output on the display screen 1A, and second image data D2 obtained by photographing the test pattern P is captured by the camera 2.

The camera interface 5A that has taken in the second image data D2 performs A/D conversion in the same manner as in the case of the first image data 01 described above, and stores it in the frame memory 5B.

Next, the frame memory 5B is converted into a digital signal.
The second image data D2 stored in the frame memory 5
It is rotated about the center point G of B by the previously detected inclination θ, and the state shown in (a1) of FIG. 7 is changed to the state shown in (a2).

This rotation causes the tilt of the display device 1 under test during photographing,
Alternatively, the tilt of the camera 2 itself is corrected to match the attitude of the frame 4 with the center lines Fl and F2 in the frame memory 5B.

Therefore, as shown in FIG. 7(b), a test pattern is created by scanning at a predetermined pitch, similar to the case where the first image data D1 is detected and identified using the four sides of the memory area of the frame memory 5B as a reference. The coordinates of the second image data D2 based on P are detected.

The straight line that best fits the coordinates Xii and Yii detected in this way is calculated by the least squares method using the above-mentioned equations (1) and (2), and the second identification data 012 shown in (c) is formed. And the slope Φ1 on each side. Φ2. Φ3
, find Φ4.

Finally, the corners of each of the four sides are calculated, the corners are connected by diagonals, and the center G1 is calculated from the intersection of the diagonals.

Therefore, the positional deviation is determined based on whether the positional deviation amount σ between the center G of the frame memory 5B and the center G1 calculated by identifying the second image data D2 is within a predetermined standard. Calculate the slope Φ of the average value of slope Φ1 to Φ4 by performing the same as the above-mentioned equation (3),
The slope is determined based on whether the slope Φ is within a predetermined standard.

Therefore, if either the positional deviation amount σ or the slope Φ exceeds a predetermined standard and it is determined to be defective, the interface section 7 outputs a signal S.

Further, if either the positional deviation amount σ or the inclination Φ is within a predetermined standard, it is determined that the product is good, and the next display device 1 to be tested is checked by repeating the check.

〔Effect of the invention〕

As described above, according to the present invention, the display screen of the display device under test is changed by causing the display device under test to output a test pattern and identifying the first and second images captured by the camera. It is possible to check the quality of the display screen by detecting positional deviations and inclinations in the display screen.

Therefore, compared to conventional visual checks, it is possible to improve quality by performing highly reliable inspections, and to reduce the number of man-hours required for checking, which has great practical effects.

[Brief explanation of drawings]

FIG. 1 is a diagram explaining the principle of the present invention. FIG. 2 is a configuration diagram of an embodiment according to the present invention. FIG. 3 is an explanatory diagram of the display screen of the present invention, in which (a) is a front view and (b) is a side view. FIG. 4 is a flow chart diagram of the present invention. Figure 5 is a density histogram. (a) and (b) of FIG. 6 are explanatory diagrams for identifying frames. (a1) (a2) (b) (c) of Fig. 7 are explanatory diagrams for identifying the test pattern. Fig. 8 is an explanatory diagram of the conventional method. (a) is a perspective view.
(b2) shows a front view of the display screen, and (b2) shows an explanatory diagram of test pattern detection. In the figure, 1 is a display device under test, and 2 is a camera. 3 is an image processing device, 4 is a frame. 5 is an identification data creation unit, and 6 is a CPU. 7 is the interface section, and LA is the display screen. Diagram for explaining the principle of the present invention Fig. 1 P test pattern Fig. 1 △ display screen (b) 1 of the present invention! Explanatory diagram of the water surface Flowchart diagram of the present invention Density histogram A diagram

Claims (1)

[Claims] Display device under test (1) that displays a test pattern (P)
), the test pattern (P) and the display device under test (1
), a camera (2) that photographs the frame (4), first image data (D1) that photographs the frame (4), and second image data (D2) that photographs the test pattern (P). an image processing device (3) for identifying the display device under test (3);
An image quality inspection device for checking the image quality in step 1), wherein the first identification data (D11) identifying the frame (4) using the first image data (D1) and the second image an identification data creation unit (5) that creates second identification data (D12) that identifies the test pattern (P) based on the data (D2); The second identification data (D12) are compared, and the positional deviation (σ) and tilt (σ) of the display screen (1A) in the display device under test (1) are compared.
The CPU (
6) and an interface unit (7) that outputs a signal (S) based on the quality determination, forming the image processing device (3).