A TFT testing device and testing method thereof

Abstract

The present invention discloses a TFT testing device and a testing method thereof. A TFT screen testing station is formed on the surface of an operating table. The above-mentioned fixture corresponds to the TFT screen testing station. A partition is formed between the test cabinet and the upper display device cabinet. Three industrial cameras are hoisted on the partition, wherein one industrial camera is located in the middle, and the industrial camera located in the middle is vertically projected on the lower TFT screen testing station. The other two industrial cameras are arranged on both sides of the industrial camera in the middle, and the two industrial cameras on both sides are obliquely projected on the lower TFT screen testing station. The three industrial cameras are arranged on the same horizontal plane and distributed in a triangular shape. The three industrial cameras are respectively located at a vertex position of the triangle, and each industrial camera is equipped with an LED fill light. Compared with the traditional test TFT screen using a single camera, the use of three camera solutions in the test process increases the processing of sampling pictures to ensure the accuracy of subsequent parameter comparison.

Description

Technical Field

[0001] The present invention relates to the technical field of TFT testing systems, in particular to a TFT testing device, and more particularly to a mechanical structure of the device and a testing method thereof. Background Art

[0002] With the development of science and technology, the operation interface of household electrical appliances is becoming more and more humanized and intelligent, from the traditional button interface to the monotonous LED and LCD interface, and now the colorful and touch-operated interface is widely used in household appliances. This is also accompanied by the problem of how to effectively monitor such interfaces during the production process. Among them, the screen test of TFT material is more difficult, and the test indicators are typical.

[0003] At present, the test items of TFT screen include: brightness, resolution, contrast, pixels, pixel pitch, etc. Among them, the pixel test has extremely high requirements for image sampling. The traditional test method is to use a high-precision camera to take pictures directly above the screen for sampling. However, due to issues such as focal length and light shooting angle, the reliable sampling range is very limited. Summary of the invention

[0004] The purpose of the present invention is to provide a TFT testing device, which adopts a solution of shooting with three industrial cameras at the same time, eliminates the influence of the light angle when shooting in the same area, and reduces the influence of the system error on the test by calculating three sets of data when measuring the same pixel point at the same time, thereby solving the problems existing in the prior art.

[0005] In order to achieve the above-mentioned purpose, the technical scheme of the present invention is: a TFT testing device, the device includes a device frame, the lower part of the device frame is an electric control cabinet, the upper part of the device frame forms a display device cabinet, the middle part of the device frame forms a test cabinet, the bottom surface of the test cabinet is provided with an operating table, a set of fixtures are provided at the operating table, a TFT screen test station is formed on the surface of the operating table, the above-mentioned fixtures correspond to the TFT screen test station, a partition is formed between the test cabinet and the upper display device cabinet, three industrial cameras are hoisted at the partition, among which one industrial camera is located in the middle, and the industrial camera located in the middle is vertically projected on the lower TFT screen test station, and the other two industrial cameras are arranged on both sides of the industrial camera in the middle, and the two industrial cameras on both sides are obliquely projected on the lower TFT screen test station, the three industrial cameras are arranged on the same horizontal plane, and are distributed in a triangular shape on the same horizontal plane, the three industrial cameras are respectively located at a vertex position of the triangle, and each industrial camera is equipped with an LED fill light.

[0006] The present invention also discloses a testing method for a TFT testing device, wherein the testing method is:

[0007] S1: Place the TFT screen under test on the TFT screen test station. The fixture fixes the TFT screen under test on the screen test station. Three industrial cameras take photos and samples of the TFT screen below at the same time, and transmit the three sampled images to the computer host.

[0008] S2: The computer host processes three pictures of the TFT screen into one picture to reduce the inaccuracy of physical data caused by system errors. According to the processing of the data algorithm, the influence of the light angle during shooting is eliminated. At the same time, when measuring the same pixel point, the influence of the system error on the test is reduced by calculating three sets of data. The data algorithm is completed in the fuzzy reasoning of the fuzzy controller. This algorithm can be divided into three parts: input variables, fuzzy controller, and control object.

[0009] s1 Input variable: given a fixed value. Since the parameters of the three cameras are exactly the same, that is, the number of pixels in the sampled images is the same, and in order to improve the test efficiency, the area of ​​the sampled images is extracted according to the size difference of the TFT screen, that is, the algorithm is only processed for the TFT screen area in the sampled image; since the technology of cutting out an area from the image is relatively mature, it will not be repeated here;

[0010] S2 fuzzy controller:

[0011] a) Fuzzification:

[0012] Since the test of TFT screen includes brightness, resolution, contrast, pixels, pixel pitch and other items, when processing pictures, the picture content is processed digitally. The coordinates of each pixel are determined, that is, (x, y). After the RGB (red, green, blue) color analysis of the pixel is performed, (r, g, b) is obtained. Assuming that there are n pixels in a picture, n*[(x, y, r), (x, y, g), (x, y, b)], n groups of three-dimensional matrices can be obtained.

[0013] At the same time, after RGB operation, we can get color (Hue), saturation (Sat), brightness (Lum) and other data, and we can get n*[(x,y,h), (x,y,s), (x,y,l)], n groups of three-dimensional matrices.

[0014] Finally, the pixel points are reduced in order according to (x, y), that is, the multi-dimensional nonlinear discrete data is reduced to a discrete two-dimensional array, and n (i, j) can be obtained. The fuzzy processing of the given quantity is completed.

[0015] b) Fuzzy reasoning:

[0016] After the three sampled images are blurred, three sets of two-dimensional arrays are generated. Since the number of pixels in the sampled images is the same, the three sets of data positions correspond one to one. Each set of points is fitted, where the curve obtained by fitting the image sampled by the camera directly above is set as the standard curve, and the other two sets are processed for error according to the standard quantity, and finally the standard curve is corrected.

[0017] c) Defuzzification:

[0018] The data of fuzzy inference fitting is discretized into n (i, j) again, and then the inverse operation is performed to generate n*[(x,y,r), (x,y,g), (x,y,b)], n*[(x,y,h), (x,y,s), (x,y,l)], and the RGB data is substituted into the picture taken directly above for correction.

[0019] d) Knowledge Base:

[0020] The knowledge base includes color RGB synthesis and decomposition operations; matrix operations; differential operations; integral operations; and fitting operations.

[0021] s3 controls the object. At this time, the processed image is the controlled object of the entire algorithm.

[0022] S3: The host computer uploads the final judgment result data to the display.

[0023] The present invention discloses a TFT testing device. Compared with the traditional method of testing a TFT screen using a single camera, the method uses three cameras in the testing process to increase the processing of sampled images, increase the reliability of the TFT screen parameter test, improve the test efficiency of the TFT screen test, increase the range of camera sampling, expand the types of products tested at the same test station, and reduce the cost of rebuilding the testing device to ensure the accuracy of subsequent parameter comparison. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of the TFT testing device structure.

[0025] Figure 2 This is a three-dimensional diagram of the TFT test device.

[0026] Figure 3 This is a flow chart of the testing method of the TFT testing device.

[0027] Figure 4 This is the flow chart of the fuzzy controller. DETAILED DESCRIPTION

[0028] The present invention will be further described below with reference to the accompanying drawings.

[0029] The present invention discloses a TFT testing device, which is different from the prior art in that: the device comprises a device frame 1, the lower part of the device frame 1 is an electric control cabinet 2, the upper part of the device frame 1 forms a display device cabinet 3, the middle part of the device frame 1 forms a test cabinet 4, the bottom surface of the test cabinet 4 is provided with an operating table 5, a set of fixtures 6 are provided at the operating table 5, a TFT screen test station 7 is formed on the surface of the operating table 5, the fixtures 6 correspond to the TFT screen test station 7, a partition 8 is formed between the test cabinet 4 and the upper display device cabinet 3, and a There are three industrial cameras 9, one of which is located in the middle and is vertically projected onto the lower TFT screen test station 7. The other two industrial cameras 9 are arranged on both sides of the middle industrial camera 9, and the two industrial cameras 9 on both sides are obliquely projected onto the lower TFT screen test station 7. The three industrial cameras 9 are arranged on the same horizontal plane and are distributed in a triangular shape on the same horizontal plane. The three industrial cameras 9 are respectively located at a vertex of the triangle, and each industrial camera 9 is equipped with an LED fill light.

[0030] In a specific implementation, test cabinet doors 10 that open upwards are provided on both sides of the test cabinet 4 .

[0031] In the specific implementation, a computer host 11 is provided in the electric control cabinet 2, a display 12 is provided in the display device cabinet 3, and a printer is provided in the display device cabinet 3 on the display side. The display and the printer are both connected to the computer host, and the above three industrial cameras are all connected to the computer host.

[0032] In the specific implementation, each industrial camera is provided with a corresponding LED fill light, and three LED lamp beads are distributed in a fan shape in the LED fill light 13. The light spreading range of the three LED lamp beads can cover the lower TFT screen test station 7 to the maximum extent.

[0033] A test method for a TFT test device, which is different from the prior art in that: the test method is

[0034] S1: Place the TFT screen under test on the TFT screen test station. The fixture fixes the TFT screen under test on the screen test station. Three industrial cameras take photos and samples of the TFT screen below at the same time, and transmit the three sampled images to the computer host.

[0035] S2: The computer host processes three pictures of the TFT screen into one picture to reduce the inaccuracy of physical data caused by system errors. According to the processing of the data algorithm, the influence of the light angle during shooting is eliminated. At the same time, when measuring the same pixel point, the influence of the system error on the test is reduced by calculating three sets of data. The data algorithm is completed in the fuzzy reasoning of the fuzzy controller. This algorithm can be divided into three parts: input variables, fuzzy controller, and control object.

[0036] s1 input variable: given a fixed value. Since the parameters of the three cameras are exactly the same, that is, the number of pixels in the sampled images is the same, and in order to improve the test efficiency, the area of ​​the sampled images is extracted according to the size difference of the TFT screen, that is, the algorithm is only processed for the TFT screen area in the sampled images;

[0037] S2 fuzzy controller:

[0038] a) Fuzzification:

[0039] Since the test of TFT screen includes brightness, resolution, contrast, pixels, pixel pitch and other items, when processing pictures, the picture content is processed digitally. The coordinates of each pixel are determined, that is, (x, y). After the RGB (red, green, blue) color analysis of the pixel is performed, (r, g, b) is obtained. Assuming that there are n pixels in a picture, n*[(x, y, r), (x, y, g), (x, y, b)], n groups of three-dimensional matrices can be obtained.

[0040] At the same time, after RGB operation, we can get color (Hue), saturation (Sat), brightness (Lum) and other data, and we can get n*[(x,y,h), (x,y,s), (x,y,l)], n groups of three-dimensional matrices.

[0041] Finally, the pixel points are reduced in order according to (x, y), that is, the multi-dimensional nonlinear discrete data is reduced to a discrete two-dimensional array, and n (i, j) can be obtained. The fuzzy processing of the given quantity is completed.

[0042] b) Fuzzy reasoning:

[0043] After the three sampled images are blurred, three sets of two-dimensional arrays are generated. Since the number of pixels in the sampled images is the same, the three sets of data positions correspond one to one. Each set of points is fitted, where the curve obtained by fitting the image sampled by the camera directly above is set as the standard curve, and the other two sets are processed for error according to the standard quantity, and finally the standard curve is corrected.

[0044] c) Defuzzification:

[0045] The data of fuzzy inference fitting is discretized into n (i, j) again, and then the inverse operation is performed to generate n*[(x,y,r), (x,y,g), (x,y,b)], n*[(x,y,h), (x,y,s), (x,y,l)], and the RGB data is substituted into the picture taken directly above for correction.

[0046] d) Knowledge Base:

[0047] The knowledge base includes color RGB synthesis and decomposition operations; matrix operations; differential operations; integral operations; and fitting operations.

[0048] s3 controls the object. At this time, the processed image is the controlled object of the entire algorithm.

[0049] S3: The host computer uploads the final judgment result data to the display.

[0050] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to the above descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.

Claims

1. A testing method for a TFT testing device, characterized in that: The device comprises a device frame (1), the lower part of the device frame (1) is an electric control cabinet (2), the upper part of the device frame (1) forms a display device cabinet (3), the middle part of the device frame (1) forms a test cabinet (4), the bottom surface of the test cabinet (4) is provided with an operating table (5), a set of fixtures (6) are provided at the operating table (5), a TFT screen test station (7) is formed on the surface of the operating table (5), the fixture (6) corresponds to the TFT screen test station (7), a partition (8) is formed between the test cabinet (4) and the upper display device cabinet (3), three industrial cameras (9) are hoisted at the partition (8), wherein one industrial camera (9) is located in the middle The industrial camera (9) located in the middle is vertically projected onto the TFT screen test station (7) at the bottom, the other two industrial cameras (9) are arranged on both sides of the industrial camera (9) in the middle, the two industrial cameras (9) on both sides are obliquely projected onto the TFT screen test station (7) at the bottom, the three industrial cameras (9) are arranged on the same horizontal plane, and are distributed in a triangular shape on the same horizontal plane, the three industrial cameras (9) are respectively located at a vertex position of the triangular shape, each industrial camera (9) is equipped with an LED fill light, a computer host (11) is arranged in the electric control cabinet (2), and the three industrial cameras are all connected to the computer host; Both sides of the test cabinet (4) are provided with test cabinet doors (10) that open upwards; The test method is S1: Place the TFT screen under test on the TFT screen test station. The fixture fixes the TFT screen under test on the screen test station. Three industrial cameras take photos and samples of the TFT screen below at the same time, and transmit the three sampled images to the computer host. S2: The computer host processes three pictures of the TFT screen into one picture to reduce the inaccuracy of physical data caused by system errors. According to the processing of the data algorithm, the influence of the light angle during shooting is eliminated. At the same time, when measuring the same pixel point, the influence of the system error on the test is reduced by calculating three sets of data. The data algorithm is completed in the fuzzy reasoning of the fuzzy controller. This algorithm is divided into three parts: input variables, fuzzy controller, and control object. s1 input variable: given a fixed value. Since the parameters of the three cameras are exactly the same, that is, the number of pixels in the sampled images is the same, and in order to improve the test efficiency, the area of ​​the sampled images is extracted according to the size difference of the TFT screen, that is, the algorithm is only processed for the TFT screen area in the sampled images; S2 fuzzy controller: a) Fuzzification: Since the test of TFT screen includes brightness, resolution, contrast, pixels, pixel pitch, etc., when processing pictures, the picture content is processed digitally, and the coordinates of each pixel are determined, that is, (x, y). Then the RGB (red, green, blue) color analysis of the pixel is performed to obtain (r, g, b); assuming that there are n pixels in a picture, n*[(x, y, r), (x, y, g), (x, y, b)], n sets of three-dimensional matrices can be obtained; At the same time, after RGB operation, we can get the color Hue, saturation Sat, and brightness Lum data, and we can get n*[(x, y, h)、(x, y, s)、(x, y, l)], n sets of three-dimensional matrices; Finally, the pixel points are reduced in order according to (x, y), that is, the multi-dimensional nonlinear discrete data is reduced to a discrete two-dimensional array, and n (i, j) can be obtained, completing the fuzzy processing of the given quantity; b) Fuzzy reasoning: After the three sampled images are blurred, three groups of two-dimensional arrays are generated. Since the number of pixels in the sampled images is the same, that is, the positions of the three groups of data correspond one to one, each group of points is fitted. Among them, the curve obtained by fitting the image sampled by the camera directly above is set as the standard curve. The other two groups are processed for errors according to the standard quantities, and finally the standard curve is corrected; c) Defuzzification: The fuzzy inference fitting data is discretized into n (i, j) again, and then the inverse operation is performed to generate n*[(x, y, r), (x, y, g), (x, y, b)], n*[(x, y, h), (x, y, s), (x, y, l)], and the RGB data is substituted into the picture taken directly above for correction; d) Knowledge Base: The knowledge base includes color RGB synthesis and decomposition operations; matrix operations; differential operations; integral operations; fitting operations; s3 control object, the processed image is the controlled object of the entire algorithm; S3: The host computer uploads the final judgment result data to the display.

Images