Tablet computer screen quality detection method and system

Abstract

The present invention relates to the field of quality management, and discloses a tablet computer screen quality detection method and system, comprising: a detection image acquisition module, used for acquiring a first surface image of a target screen in a power-off state and acquiring a second surface image of the target screen in a power-on state; an image analysis module, used for judging whether the target screen has defects; an image marking module, marking defect positions on a first standard image and a second standard image with defects according to the judgment result of the image analysis module; a quality assessment module, obtaining a quality coefficient according to the number of defects of the first standard image and the second standard image and the total number of defects after the first standard image and the second standard image are overlapped, and judging the quality of the target screen; a quality prediction module, obtaining a quality trend coefficient according to the number of target screens with defects in a set period of time and a curve of the quality coefficient changing with time, and predicting the quality trend of the target screen after comparing with a preset interval.

Description

Technical Field

[0001] The present invention relates to the field of quality management, and in particular to a tablet computer screen quality detection method and system. Background Art

[0002] Tablet computers are portable smart devices, and their screen quality directly affects their overall service life and user experience. Therefore, tablet computer manufacturers need to conduct quality inspections on finished products after assembly to ensure that the quality of the tablet computer screens is qualified.

[0003] However, existing screen quality inspections mostly use manual visual observation of appearance defects. The general method is to first observe with the naked eye whether there are obvious scratches, cracks or damage, and then check whether there are bright spots and dark spots by turning on the screen and setting it to pure black and pure white. Then, check the integrity of each color by setting the screen to red, green and blue. Although most of the appearance quality problems can be found in the above process, manual observation on large-scale production lines will obviously cause misjudgment and missed detection, and the efficiency of the above detection method is also low. Once a quality problem occurs, batch problems are likely to occur. If feedback is given at this time, a large number of problematic products have accumulated, which is obviously not conducive to quality control feedback on the production line. Summary of the invention

[0004] The purpose of the present invention is to provide a tablet computer screen quality detection method and system to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A tablet computer screen quality detection system, comprising:

[0007] The detection image acquisition module includes a first image acquisition unit and a second image acquisition unit, wherein the first image acquisition unit is used to acquire a first surface image of the target screen in a power-off state, and the second image acquisition unit is used to acquire a second surface image of the target screen in a power-on state;

[0008] An image analysis module, used for processing the first surface image and the second surface image to form a first standard image and a second standard image, and comparing the first standard image and the second standard image with corresponding preset comparison images to determine whether the target screen has defects;

[0009] An image marking module, which marks defect locations on the first standard image and the second standard image where defects exist, according to the judgment result of the image analysis module;

[0010] The quality assessment module obtains the number of defects of the first standard image and the second standard image and the total number of defects after the first standard image and the second standard image overlap according to the defect position information on the first standard image and the second standard image, obtains the quality coefficient after processing, and judges the quality of the target screen according to the quality coefficient;

[0011] The quality prediction module obtains the number of target screens with defects within a set period of time and the curve of the quality coefficient changing with time, calculates the quality trend coefficient, and predicts the quality trend of the target screen based on the comparison between the quality trend coefficient and the preset interval;

[0012] The scheduling module is used to schedule the production line accordingly according to the predicted quality trend of the quality prediction module.

[0013] As a further technical solution, the process of obtaining the quality coefficient is:

[0014] By formula: Calculate the quality factor R i ;

[0015] Among them, y i is the defect concentration index, is the number of defects on the first standard image, is the number of defects on the second standard image, is the total number of defects after the first standard image and the second standard image overlap, and α, β, and γ are preset coefficients.

[0016] As a further technical solution, the process of judging the quality of the target screen according to the quality coefficient is as follows:

[0017] The quality factor R i and the preset quality threshold R th Compare;

[0018] If R i ≥R th , then the current target screen quality is judged to be unqualified; if R i <R th , then the current target screen quality is judged to be qualified.

[0019] As a further technical solution, the process of obtaining the quality trend coefficient is:

[0020] By formula:

[0021]

[0022] Get the quality trend coefficient K I ;

[0023] Among them, μ i is the conversion coefficient, tk ,t k+1 are the start time and end time of the preset period, R i (t) is the curve of quality coefficient changing with time, R i0 (t) is the predicted change curve of the quality coefficient over time, m is the number of preset time periods, n is the number of target screens with unqualified quality in each preset time period, ΔR i is the reference value obtained based on historical data, ρ i1 , i2 is the weight coefficient.

[0024] As a further technical solution, the defect concentration index is obtained by:

[0025] A third image is formed by overlapping the first standard image and the second standard image, and the third image is placed in a coordinate system so that a center point of the third image coincides with a coordinate origin;

[0026] like The first mode is adopted for acquisition;

[0027] like The second mode is used for acquisition;

[0028] Where b is the preset number of standard defects, which is formulated based on industry standards and historical data.

[0029] As a further technical solution, the first mode is:

[0030] Mark the overlapping defect locations and get the number of overlapping defects g j ,

[0031] By formula: Calculate the defect concentration index y i ;

[0032] Where n is the number of unqualified target screens within a preset period of time.

[0033] As a further technical solution, the second mode is:

[0034] Mark the overlapping defect locations, and connect the outermost key marks around the coordinate origin in sequence to form a closed figure as the key defect area, and the rest of the area as the non-key defect area;

[0035] The area range is delineated with a preset radius d with each key mark as the center, and the key marks that touch the area range are regarded as the concentration area;

[0036] By formula: Calculate the defect concentration index yi ;

[0037] Where n is the number of unqualified target screens in a preset period, δ1 and δ2 are preset proportional coefficients, and S jh is the area of ​​the key defect area, S z is the total area of ​​the third image, S cj is the total area of ​​the concentration area, p j is the number of concentrated areas, ε1 and ε2 are influence coefficients.

[0038] As a further technical solution, the working process of the scheduling module is:

[0039] The mass trend coefficient K I With the preset interval [K I0 , K I1 ] for comparison;

[0040] If K I ∈[K I0 , K I1 ], the production line quality is judged to be normal and the production line continues to operate;

[0041] like The production line quality is judged to be abnormal and the production line is stopped.

[0042] A tablet computer screen quality detection method is applicable to the above-mentioned tablet computer screen quality detection system.

[0043] Beneficial effects of the present invention:

[0044] (1) Through the above technical solution, it is possible to count the target screens with defects within a preset period of time, exclude the target screens without defects through a pre-elimination method, and then analyze the target screens with defects. Obviously, the more defects there are on the screen and the higher the overlap of defects in the images obtained in the power-on state and the power-off state, the higher the probability that the defects on the target screen are real defects, thereby improving the accuracy of the quality detection of the target screen; and by analyzing the number of defects on the first standard image, the number of defects on the second standard image, and the total number of defects after the first standard image and the second standard image overlap, it is possible to accurately judge the quality of the target screen and reduce the probability of misjudgment. In addition, a defect concentration index is introduced. Obviously, the higher the defect concentration, the more likely the target screen is to have quality problems in the same area. At this time, the production line station corresponding to the defect position can be checked in a focused manner, and improvements can be made in a timely manner to improve the overall quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1 As shown, the present invention is a tablet computer screen quality detection system, comprising:

[0049] The detection image acquisition module includes a first image acquisition unit and a second image acquisition unit, wherein the first image acquisition unit is used to acquire a first surface image of the target screen in a power-off state, and the second image acquisition unit is used to acquire a second surface image of the target screen in a power-on state;

[0050] An image analysis module, configured to process the first surface image and the second surface image to form a first standard image and a second standard image, and compare the first standard image and the second standard image with corresponding preset comparison images to determine whether the target screen has defects;

[0051] An image marking module, which marks defect locations on the first standard image and the second standard image where defects exist, according to the judgment result of the image analysis module;

[0052] The quality assessment module obtains the number of defects of the first standard image and the second standard image and the total number of defects after the first standard image and the second standard image overlap according to the defect position information on the first standard image and the second standard image, obtains the quality coefficient after processing, and judges the quality of the target screen according to the quality coefficient;

[0053] The quality prediction module obtains the number of target screens with defects within a set period of time and the curve of the quality coefficient changing with time, calculates the quality trend coefficient, and predicts the quality trend of the target screen based on the comparison between the quality trend coefficient and the preset interval;

[0054] The scheduling module is used to schedule the production line accordingly according to the predicted quality trend of the quality prediction module.

[0055] In this embodiment, the first surface image of the target screen in the power-off state and the second surface image in the power-on state are respectively acquired by the detection image acquisition module, and the first surface image and the second surface image are processed by the image analysis module to form the first standard picture and the second standard picture, and then the first standard picture and the second standard picture are compared with the corresponding preset comparison picture to determine whether there are defects in the target screen. If there are no defects, it is determined that the target screen has good quality. Otherwise, the image marking module marks the defect positions on the first standard picture and the second standard picture with defects according to the judgment result of the image analysis module, and then the defect position information on the first standard picture and the second standard picture is used to obtain the respective defect numbers of the first standard picture and the second standard picture and the total defect number after the first standard picture and the second standard picture overlap, and the quality is obtained after processing. The quality coefficient is obtained, and the quality of the target screen is judged according to the quality coefficient. Then, the number of target screens with defects in a set period of time and the change curve of the quality coefficient over time are obtained through the quality prediction module, and the quality trend coefficient is calculated. The quality trend of the target screen is predicted by comparing the quality trend coefficient with the preset interval; finally, the scheduling module performs corresponding scheduling of the production line according to the predicted quality trend of the quality prediction module; through the above technical solution, it is possible to count the target screens with defects in a preset period of time, exclude the target screens without defects by the pre-elimination method, and then analyze the target screens with defects. Obviously, the more defects there are on the screen, and the higher the overlap of defects in the images obtained in the power-on state and the power-off state, the higher the probability that the defects on the target screen are real defects, thereby improving the accuracy of the quality detection of the target screen;

[0056] It should be noted that the first surface image and the second surface image are taken by a professional industrial camera, wherein the second surface image is an image obtained when the target screen is white or black, so as to facilitate the discovery of bright spot or dark spot defects, which will not be elaborated here.

[0057] The process of obtaining the quality coefficient is:

[0058] By formula: Calculate the quality factor R i ;

[0059] Among them, y i is the defect concentration index, is the number of defects on the first standard image, is the number of defects on the second standard image, is the total number of defects after the first standard image and the second standard image overlap, and α, β, and γ are preset coefficients.

[0060] The process of judging the target screen quality based on the quality coefficient is as follows:

[0061] The quality factor R i and the preset quality threshold R th Compare;

[0062] If R i ≥R th , then the current target screen quality is judged to be unqualified; if R i <R th , then the current target screen quality is judged to be qualified.

[0063] In this embodiment, by formula: Calculate the quality factor R i ; Then the quality coefficient R i and the preset quality threshold R th Compare; if R i ≥R th , then the current target screen quality is judged to be unqualified; if R i <R th , then the current target screen is judged to be of qualified quality. By analyzing the number of defects on the first standard image, the number of defects on the second standard image, and the total number of defects after the first standard image and the second standard image overlap, the target screen can be accurately judged for quality, reducing the probability of misjudgment, and introducing a defect concentration index. Obviously, the higher the defect concentration, the more likely the target screen is to have quality problems in the same area. At this time, the production line station corresponding to the defect position can be checked in detail, and improvements can be made in time to improve the overall quality.

[0064] The process of obtaining the quality trend coefficient is:

[0065] By formula:

[0066]

[0067] Get the quality trend coefficient K I ;

[0068] Among them, μ i is the conversion coefficient, t k ,t k+1 are the start time and end time of the preset period, R i (t) is the curve of quality coefficient changing with time, R i0 (t) is the predicted change curve of the quality coefficient over time, m is the number of preset time periods, n is the number of target screens with unqualified quality in each preset time period, ΔR i is the reference value obtained based on historical data, ρ i1 , i2 is the weight coefficient.

[0069] The working process of the scheduling module is as follows:

[0070] The mass trend coefficient K I With the preset interval [K I0 , K I1 ] for comparison;

[0071] If K I ∈[K I0 , K I1 ], the production line quality is judged to be normal and the production line continues to operate;

[0072] like The production line quality is judged to be abnormal and the production line is stopped.

[0073] In this embodiment, the formula Calculate the mass trend coefficient K I , R i (t) is the curve of quality coefficient changing with time, R i0 (t) is the predicted change curve of the quality coefficient over time, m is the number of preset time periods, n is the number of target screens with unqualified quality in each preset time period, ΔR i is the reference value obtained based on historical data, ρ i1 , i2 is the weight coefficient, is the average of the total number of defects after the first standard image and the second standard image overlap; Obviously, when the production line is running normally, if the number of target screens with unqualified quality is greater within the preset period, it means that the quality trend is worse and batch quality problems are more likely to occur. According to the formula, it can be directly seen that |R i0 (t)-R i (t)|The larger the deviation, the more unstable the quality and the greater the volatility. Therefore, the higher the probability of quality problems in the next preset period. The larger the difference is, the greater the deviation between the number of defects after the first standard image and the second standard image overlap and the mean is, and the more unstable it is; the establishment of a model based on the two parameters can more accurately predict the quality of subsequent product quotas, so as to make production line scheduling measures in advance, reduce losses and improve product quality.

[0074] The process of obtaining the defect concentration index is as follows:

[0075] A third image is formed by overlapping the first standard image and the second standard image, and the third image is placed in a coordinate system so that a center point of the third image coincides with a coordinate origin;

[0076] like The first mode is adopted for acquisition;

[0077] The first mode is:

[0078] Mark the overlapping defect locations and get the number of overlapping defects g j ,

[0079] By formula: Calculate the defect concentration index y i ;

[0080] Where n is the number of unqualified target screens within a preset period of time.

[0081] like The second mode is used for acquisition;

[0082] Where b is the preset number of standard defects, which is formulated based on industry standards and historical data.

[0083] The second mode is:

[0084] Mark the overlapping defect locations, and connect the outermost key marks around the coordinate origin in sequence to form a closed figure as the key defect area, and the rest of the area as the non-key defect area;

[0085] The area range is delineated with a preset radius d with each key mark as the center, and the key marks that touch the area range are regarded as the concentration area;

[0086] By formula: Calculate the defect concentration index y i ;

[0087] Where n is the number of unqualified target screens in a preset period, δ1 and δ2 are preset proportional coefficients, and S jh is the area of ​​the key defect area, S z is the total area of ​​the third image, S cj is the total area of ​​the concentration area, p j is the number of concentrated areas, ε1 and ε2 are influence coefficients.

[0088] In this embodiment, the defect concentration index y is obtained by two modes respectively. i ,when ≤b, then the first mode is used to obtain, that is, through the formula: Calculate the defect concentration index y i ;when Then the second mode is used to obtain it, that is, through the formula: Calculate the defect concentration index y iThrough the two modes, targeted analysis of defect concentration can be performed on target screens with different defect levels to improve the actual quality coefficient of each target screen and improve the accuracy of quality judgment.

[0089] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A tablet computer screen quality detection system, characterized in that: include: The detection image acquisition module includes a first image acquisition unit and a second image acquisition unit, wherein the first image acquisition unit is used to acquire a first surface image of the target screen in a power-off state, and the second image acquisition unit is used to acquire a second surface image of the target screen in a power-on state; An image analysis module, configured to process the first surface image and the second surface image to form a first standard image and a second standard image, and compare the first standard image and the second standard image with corresponding preset comparison images to determine whether the target screen has defects; An image marking module, which marks defect locations on the first standard image and the second standard image where defects exist, according to the judgment result of the image analysis module; The quality assessment module obtains the number of defects of the first standard image and the second standard image and the total number of defects after the first standard image and the second standard image overlap according to the defect position information on the first standard image and the second standard image, obtains the quality coefficient after processing, and judges the quality of the target screen according to the quality coefficient; The quality prediction module obtains the number of target screens with defects within a set period of time and the curve of the quality coefficient changing with time, calculates the quality trend coefficient, and predicts the quality trend of the target screen based on the comparison between the quality trend coefficient and the preset interval; The scheduling module is used to schedule the production line accordingly according to the quality trend predicted by the quality prediction module; The process of obtaining the quality coefficient is: By formula: Calculate the quality factor ; in, is the defect concentration index, is the number of defects on the first standard image, is the number of defects on the second standard image, is the total number of defects after the first standard image and the second standard image overlap, , , is the preset coefficient; The process of judging the target screen quality based on the quality coefficient is as follows: The quality factor With preset quality threshold Compare; like ≥ , then the current target screen quality is judged to be unqualified; if < , then the current target screen quality is judged to be qualified.

2. The tablet computer screen quality detection system according to claim 1, characterized in that: The process of obtaining the quality trend coefficient is: By formula: ； Get the quality trend coefficient ; in, is the conversion coefficient, , The start and end time of the preset time period respectively. is the curve of quality coefficient changing with time, is the predicted change curve of quality coefficient over time, is the number of preset time periods, n is the number of target screens with unqualified quality in each preset time period, is a reference value obtained based on historical data, , is the weight coefficient.

3. The tablet computer screen quality detection system according to claim 1, characterized in that: The process of obtaining the defect concentration index is as follows: A third image is formed by overlapping the first standard image and the second standard image, and the third image is placed in a coordinate system so that a center point of the third image coincides with a coordinate origin; like ≤b, the first mode is used for acquisition; like > b, the second mode is used for acquisition; Where b is the preset number of standard defects, which is formulated based on industry standards and historical data.

4. The tablet computer screen quality detection system according to claim 3, characterized in that: The first mode is: Mark the overlapping defect locations and get the number of overlapping defects , through the formula: Calculate the defect concentration index ; In the formula, The number of unqualified target screens within a preset period of time.

5. The tablet computer screen quality detection system according to claim 3, characterized in that: The second mode is: Mark the overlapping defect locations, and connect the outermost key marks around the coordinate origin in sequence to form a closed figure as the key defect area, and the rest of the area as the non-key defect area; The area range is delineated with a preset radius d with each key mark as the center, and the key marks that touch the area range are regarded as the concentration area; By formula: Calculate the defect concentration index ; In the formula, The number of unqualified target screens within a preset period of time. , is the preset scale factor, is the area of ​​the key defect area, is the total area of ​​the third image, is the total area of ​​the concentration area, is the number of concentrated areas, , is the influence coefficient.

6. The tablet computer screen quality detection system according to claim 1, characterized in that: The working process of the scheduling module is as follows: The mass trend factor With preset interval Make a comparison; like ∈ , the production line quality is judged to be normal and the production line continues to operate; like ∉ , the production line quality is judged to be abnormal and the production line is stopped.

7. A tablet computer screen quality detection method, characterized in that: The detection method is applicable to the tablet computer screen quality detection system described in any one of claims 1 to 6.

Images