Panel screen quality monitoring system based on image analysis processing

Abstract

The invention relates to the technical field of tablet personal computer monitoring, and discloses a tablet screen quality monitoring system based on image analysis processing, which comprises the following modules: an image acquisition module, which comprises an industrial camera and an annular light source, the annular light source is used for irradiating the flat screen in a screen-off state and performing image shooting on the irradiated flat screen in the screen-off state through the industrial camera; and the image preprocessing module is used for dividing the shot and acquired flat screen image into a plurality of pixel areas, and performing difference graph calculation on the pixel area image acquired in the bright screen state and the pixel area image acquired in the dark screen state to obtain difference graph data. Through differential graph calculation, a brightness balance coefficient and a nonlinear suppression factor are introduced, and dynamic compensation of ambient light and adaptive enhancement of signals are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of tablet computer monitoring, and specifically to a tablet screen quality monitoring system based on image analysis and processing. Background Art

[0002] As a key display component of modern electronic devices, tablet screens are widely used in various products such as mobile phones, tablet computers, and monitors. They are mainly made of various technologies such as liquid crystals and organic light-emitting diodes, and present image and text information by controlling the light emission or light transmission of pixel points. Their quality directly affects the user's visual experience and the overall performance of the device.

[0003] For tablet screen monitoring, most of the existing technologies use image acquisition devices to obtain tablet screen images, and then use a variety of algorithms to process and analyze the images to detect whether there are defects on the screen. These technologies can, to a certain extent, identify some obvious defects, such as large-sized dead pixels and severe scratches.

[0004] However, some of the existing technologies do not timely consider the influence of ambient light interference on image acquisition during monitoring, resulting in the acquired image data being unable to accurately reflect the true state of the screen. When analyzing and processing the images, it is difficult to effectively distinguish the active light-emitting characteristics and passive refractive light-emitting characteristics of the tablet screen, which causes the defect characteristics to be easily masked by the interference signals brought by the ambient light when detecting some minor defects, thus resulting in missed detections.

[0005] Therefore, the present invention proposes a tablet screen quality monitoring system based on image analysis and processing to solve the deficiencies in the existing technologies. Summary of the Invention

[0006] The purpose of the present invention is to provide a tablet screen quality monitoring system based on image analysis and processing to solve the problem that the defect characteristics are easily masked by the interference signals brought by the ambient light in the existing technologies, thus resulting in missed detections.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A tablet screen quality monitoring system based on image analysis and processing includes the following modules:

[0009] Image acquisition module: It includes an industrial camera and a ring light source. The industrial camera is used to take images of the tablet screen in the lit state, and the ring light source is used to irradiate the tablet screen in the off state, and the industrial camera is used to take images of the irradiated tablet screen in the off state.

[0010] Image preprocessing module: Divide the captured flat screen image into multiple pixel regions, and obtain the differential map data by calculating the differential map of the pixel region image collected in the lit screen state and the pixel region image collected in the off screen state.

[0011] Defect monitoring module: According to the differential map data obtained by the image preprocessing module, perform defect density detection, regional continuity detection and dynamic response detection on the flat screen, and decide whether the flat screen is qualified according to the detection data;

[0012] Learning module: Collect and record various data detected by the defect detection module, and record the current monitoring environment to facilitate subsequent adjustment of the detection data threshold.

[0013] Preferably, the differential map data is obtained through the formula:

[0014]

[0015] Where, is the differential map data of the pixel region image at the coordinate , is the pixel brightness value of the pixel region image in the lit screen state at the coordinate , is the pixel brightness value of the pixel region image in the off screen state at the coordinate , is the brightness balance coefficient, is the non-linear suppression factor.

[0016] Preferably, the defect density detection is used to evaluate the overall defect density of the screen, through the formula:

[0017]

[0018] Where, is the defect density index, the number of defective pixels is the number of coordinates of defective pixels detected in the divided pixel regions, is the total number of pixels of the screen.

[0019] Preferably, the regional continuity detection is used to limit the maximum connected defect region of scratch type, through the formula:

[0020]

[0021] Where, is the regional continuity detection index, is the number of pixels of a single defect region, is the total area of the screen.

[0022] Preferably, the dynamic response detection is used to monitor the trend of screen performance decay, through the formula:

[0023]

[0024] Among them, is the dynamic response detection index, K is the number of consecutive detections, is the average value of the differential graph data in historical detections.

[0025] Preferably, the is used to eliminate environmental light interference and is obtained through the following formula:

[0026]

[0027] Preferably, the non-linear suppression factor is used to prevent distortion in overexposed areas. According to different types of screens, the parameters are also different accordingly, as follows:

[0028] For LCD screens, = 0.02

[0029] For OLED screens, = 0.03.

[0030] Preferably, the defect density index has a threshold value when facing LCD screens, and a threshold value when facing OLED screens. The regional continuity index has a threshold value and the dynamic response index has a threshold value

[0031] Advantages of the present invention:

[0032] 1. Through differential graph calculation, the present invention introduces a brightness balance coefficient and a non-linear suppression factor, achieving dynamic compensation for environmental light and adaptive enhancement of signals, enabling the monitoring system to maintain stable detection performance in complex lighting environments, and solving the problem in the prior art that defect features are easily masked by interference signals brought by environmental light, resulting in missed detections.

[0033] 2. The defect monitoring module of the present invention performs defect density detection, regional continuity detection, and dynamic response detection based on differential graph data, calculates corresponding indicators through specific formulas and compares them with threshold values set for different screen materials, can comprehensively evaluate the overall quality of the screen, and determine whether the screen is qualified, achieving the effect of accurately evaluating the quality of flat panel screens.

[0034] 3. By collecting and recording various data detected by the defect detection module and monitoring the environment, the present invention facilitates subsequent adjustment of the detection data threshold, enabling the monitoring system to continuously optimize according to the actual situation, adapt to different production environments and screen types, and achieving the effects of improving the adaptability and accuracy of the monitoring system.

[0035] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0037] Figure 1 It is a schematic diagram of the module framework of a flat panel screen quality monitoring system based on image analysis and processing according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0039] Please refer to Figure 1 As shown, the present invention is a flat panel screen quality monitoring system based on image analysis and processing, which is characterized by including the following modules:

[0040] Image acquisition module: It includes an industrial camera and a ring light source. The industrial camera is used to take images of the flat panel screen in the lit state, and the ring light source is used to irradiate the flat panel screen in the off state and take images of the irradiated flat panel screen in the off state through the industrial camera.

[0041] Image preprocessing module: The captured flat panel screen images are divided into multiple pixel regions, and differential map data is obtained by calculating the difference between the pixel region images collected in the lit state and the pixel region images collected in the off state.

[0042] Defect monitoring module: Based on the differential map data obtained by the image preprocessing module, defect density detection, regional continuity detection, and dynamic response detection are performed on the flat panel screen, and it is decided whether the flat panel screen is qualified according to the detection data.

[0043] Learning module: Collect and record various types of data detected by the defect detection module, and record the current monitoring environment to facilitate subsequent adjustment of the detection data threshold.

[0044] In this embodiment, after the flat panel screen enters the monitoring area, first make the flat panel in the lit screen state, and then use an industrial camera to capture the lit flat panel screen to obtain the flat panel screen image in the lit screen state. Subsequently, make the flat panel in the off-screen state, and then after irradiating the off-screen flat panel screen with an annular light source, use the industrial camera to capture the off-screen and irradiated flat panel screen again to obtain the flat panel screen image in the off-screen state.

[0045] Subsequently, the image preprocessing module divides the flat panel screen image in the lit screen state and the flat panel screen image in the off-screen state into regions, so that each flat panel screen image in each state is divided into multiple pixel regions. At this time, the image preprocessing module calculates the difference map for each divided pixel region and obtains the difference map data. Subsequently, according to the obtained difference map data, compare it with the average difference map data to determine whether there are defects in the divided pixel regions.

[0046] When there are defects in the pixel region, further defect monitoring needs to be carried out on the pixel region, including defect density detection, regional continuity detection, and dynamic response detection. Defect density detection is mainly used to detect the screen as a whole from a macroscopic perspective to judge the overall defects of the flat panel screen, while regional continuity detection is mainly used to detect scratch-like defects existing on the flat panel screen to judge the integrity of the flat panel screen. The final dynamic response detection is mainly used to detect whether there are problems with the backlight module and performance material attenuation of the screen.

[0047] After the defect detection module finishes defect density detection, regional continuity detection, and dynamic response detection on the flat panel screen, collect the obtained defect density index, regional continuity index, dynamic response detection index, and monitored environmental temperature to facilitate subsequent adjustment of the overall monitoring threshold.

[0048] The difference map data is obtained through the formula:

[0049]

[0050] Where, is the difference map data of the pixel region image at the coordinate , is the pixel brightness value of the pixel region image in the lit screen state at the coordinate , is the pixel brightness value of the pixel region image in the off-screen state at the coordinate , is the brightness balance coefficient, is a non-linear suppression factor.

[0051] In the process of industrial production monitoring of flat panel screens, the lit state of the flat panel screen can mainly reflect the active light-emitting characteristics of the screen, such as the pixel driving, backlight uniformity and other aspects of the performance of the flat panel screen, while the off state of the flat panel screen will expose the surface physical characteristics, such as some scratches on the flat panel surface and some coating defects generated during production. By performing image processing on the flat panel screen in the lit state and the off state and combining system algorithms, the light-emitting characteristics and surface defects of the screen can be identified.

[0052] Some of the existing screen brightness detection technologies are single-modal detections. In the lit state, by making the entire flat panel screen turn white, a lack of light emission at a certain location on the flat panel screen can be detected, thereby judging the internal circuit defects of the flat panel screen. By scanning the flat panel screen in the off state, the defects of the flat panel screen itself can be detected.

[0053] In this example, an industrial camera is also used to take pictures of the flat panel screen, and the collected flat panel screen images are divided into multiple pixel regions, and each pixel region consists of multiple coordinates The coordinates of the pixel region The pixel brightness value at is In the lit state where the flat panel screen emits light by itself, the pixel brightness value of the coordinate is recorded as ;

[0054] In daily life, there is often an ambient light in the natural environment and the living environment. In the off state, the flat panel screen is irradiated by the ambient light. According to the physical theorem, when the ambient light irradiates the surface of the flat panel screen, a refracted light will be formed on the flat panel screen. At this time, the flat panel screen is in a passive light-emitting state, and the refracted light generated by the flat panel screen will cover up the defects existing in the flat panel screen itself, making it difficult for users to find the defects of the flat panel screen.

[0055] However, natural light also exists in most industrial detections. Therefore, when scanning and monitoring the flat panel screen in the off state, these natural lights will have a certain impact on the detection data.

[0056] Therefore, in order to consider the user experience, in the process of flat panel screen production monitoring, this invention also simulates the daily use scenario and monitors the flat panel screen in the off state under ambient light irradiation. In the off state, the brightness value of the coordinates of the pixel regions irradiated by the ambient light is recorded as The brightness difference value between the self-emission of the flat panel screen and the refracted light emission caused by the ambient light is ;

[0057] Brightness balance coefficient The main function is to reduce the brightness difference value of the pixel area coordinates. The brightness balance coefficient is mainly obtained through the following formula;

[0058]

[0059] In the differential graph data calculation formula, the numerator term:

[0060]

[0061] Its purpose is to separate the active light-emitting characteristics of the flat panel screen in the lit state from the passive refractive light-emitting characteristics of the flat panel screen in the off state. In the off state, light is emitted onto the flat panel screen by an annular light source. When the off flat panel screen is irradiated by the light emitted by the annular light source, corresponding refracted light will be generated.

[0062] The normal pixel area image in the lit state is approximately equal to that in the off state after compensation by the brightness balance coefficient while the pixel area image with defects will show a difference; For example, for a certain screen, in the off state

[0063] and in the lit state with a brightness balance coefficient then after compensation by the brightness balance coefficient Therefore, the brightness difference value obtained by performing a differential calculation between the compensated by the brightness balance coefficientand the value in the lit state will highlight the defective area of the pixel area image.

[0064]

[0065] And in the differential graph data calculation formula, the denominator term:

[0065]

[0066] Its purpose is to prevent signal over-saturation from occurring between the value in the lit state of the pixel area and the value in the off state Specifically, in the numerator term there will be an error in the brightness balance coefficient and the brightness difference value of the low-defect pixel area will be too large, so the denominator term needs to be introduced to suppress the signal amplitude in the high-brightness area. The 1 in the denominator term mainly serves to ensure that the denominator is always greater than 1 to avoid division-by-zero errors, and the non-linear suppression factor It is mainly used for control to generate the high brightness difference.

[0067] Specifically, when the brightness difference value is too large, the denominator will increase, so the non-linear suppression factor is required to suppress the brightness difference output value. According to the different types of flat panel screens, the value of the non-linear suppression factor will also be different. When the flat panel screen uses LCD material, the value of the non-linear suppression factor is 0.02, while when the flat panel screen uses OLED material, the value of the non-linear suppression factor is 0.03.

[0068] When the image of the pixel region with defects passes through the differential map calculation to obtain the differential map data After that, the obtained differential map data is compared with the differential map data of the image of the normal pixel region to determine that there are defects in the pixel region image.

[0069] By performing differential calculation on the flat panel screen in the lit state and the flat panel screen in the off state, after eliminating the influence of ambient light, the brightness difference value generated by the pixel points in a single pixel region is the actual defect existing in the flat panel screen, and the internal defects of the flat panel screen can be captured according to the obtained differential map data.

[0070] Defect density detection is used to evaluate the overall defect density of the screen. Through the formula:

[0071]

[0072] Among them, is the defect density index, the number of defective pixels is the number of coordinates of defective pixels detected in the divided pixel region, is the total number of pixels of the screen.

[0073] In this embodiment, for the overall defect density of the screen, first, the number of coordinates of defective pixels detected in all defective pixel regions is counted, and then the proportion of defective pixels in the total number of pixels of the screen is calculated to obtain the defect density index;

[0074] More specifically, after obtaining the number of defective pixel coordinates, it reflects the density of the defective pixel regions of the entire flat panel screen, thereby reflecting whether the overall internal circuit of the flat panel screen meets the standards.

[0075] And screens made of different materials have corresponding threshold indicators. When the flat panel screen uses LCD material, the defect density threshold , the adjustment value is set to according to historical experience. When the flat panel screen uses an OLED material, the defect density threshold , and the adjustment value is set according to historical experience as .

[0076] For example, a flat panel screen made of LCD material has a total screen pixel of 5 million, so the allowable defective pixels of this flat panel screen are . When the defective pixels exceed , then this screen is unqualified;

[0077] Region continuity detection is used to limit the maximum connected defect region of scratches. Through the formula:

[0078]

[0079] where, is the region continuity detection index, is the number of pixels in a single defect region, is the total screen area.

[0080] In this embodiment, the main purpose is to monitor the maximum connected defect area region to avoid visual differences caused by concentrated defects. Since the human eye is more sensitive to local continuous defects, it is necessary to strictly restrict the continuous defect area existing on the flat panel screen;

[0081] The specific continuous defect area is that the pixel coordinates of multiple defective pixels are adjacent and connected and distributed in the pixel region, and these adjacent and connected pixel coordinates constitute surface defect features such as scratches and pixel dead points on the screen surface. Compared with the single-modal scanning detection, the present invention appears to be more detailed and microscopic.

[0082] For flat panels made of LCD and OLED materials, the screen region continuity threshold , and the adjustment value is set according to historical experience as .

[0083] For example: If the maximum continuous defect threshold of a certain screen is 0.005%, that is, the maximum allowable continuous defective pixels are 250, then a scratch of 15×15 = 225 pixels will be accepted, while a scratch of 16×16 = 256 pixels will not be accepted, and thus it is determined that the flat panel screen is unqualified.

[0084] The dynamic response detection is used to monitor the screen performance decay trend. Through the formula:

[0085]

[0086] where, is the dynamic response detection index, is the number of consecutive detections, Is the average value of the differential graph data in historical detection.

[0087] In this embodiment, its main purpose is to identify potential risks for the flat panel screen before the defect density index has not reached the defect density threshold and the regional continuity index has not reached the threshold ;

[0088] Specifically, through the formula:

[0089]

[0090] When continuously monitoring multiple screens, if the presented dynamic response index gradually rises and exceeds the set dynamic response threshold , it means that the performance of the flat panel screen gradually decays, and there are defects in the manufacturing process, resulting in the occurrence of unqualified situations for multiple flat panel screens. The preset dynamic response thresholds for LCD material screens and OLED material screens , and the adjustment value is set to according to historical experience.

[0091] When judging whether a flat panel screen is qualified, the key is to compare whether the defect density index , the regional continuity index , and the dynamic response index exceed any of the set defect density thresholds , the regional continuity threshold , and the dynamic response threshold . When any one of them exceeds the set threshold, the flat panel screen is determined to be unqualified. When multiple continuously monitored dynamic response indexes all exceed the set dynamic response threshold , it means that the performance of this batch of flat panel screens gradually decays, which also means that the processing technology needs to be adjusted accordingly.

[0092] After detecting each flat panel screen, the detected data is collected and recorded together with the detection environment temperature at that time to facilitate subsequent adjustment of the threshold. When the defect density index and the regional continuity index of the continuously detected flat panel screens do not reach the set defect density threshold and the regional continuity threshold , while the dynamic response index exceeds the dynamic response threshold , the defect density threshold needs to be correspondingly reduced.With the regional continuity threshold ;

[0093] When the defect density index during the detection of the flat panel screen and the regional continuity index both exceed the set defect density threshold and the regional continuity threshold , and the dynamic response index does not exceed the dynamic response threshold , then it is necessary to appropriately increase the defect density threshold and the regional continuity threshold .

[0094] The above content is only an example and explanation of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution. As long as they do not deviate from the concept of the invention or exceed the scope defined by this claim book, they should all fall within the protection scope of the present invention.

Claims

1. A flat panel screen quality monitoring system based on image analysis and processing, characterized in that, It includes the following modules: Image acquisition module: It includes an industrial camera and a ring light source. The industrial camera is used to take images of the tablet screen in the lit screen state, and the ring light source is used to irradiate the tablet screen in the off-screen state, and the industrial camera is used to take images of the irradiated tablet screen in the off-screen state; Image preprocessing module: Divide the captured tablet screen image into multiple pixel regions, and obtain differential map data by calculating the differential map of the pixel region image collected in the lit screen state and the pixel region image collected in the off-screen state. Defect monitoring module: According to the differential map data generated by the image preprocessing module, perform defect density detection, regional continuity detection and dynamic response detection on the tablet screen, and determine whether the tablet screen is qualified based on the detection data; Learning module: Collect and record various data detected by the defect detection module, and record the current monitoring environment to facilitate subsequent adjustment of the detection data threshold.

2. The flat panel screen quality monitoring system based on image analysis and processing according to claim 1, characterized in that, The differential map data is obtained through the formula: Among them, is the differential map data of the pixel region image at the coordinate . is the pixel brightness value of the pixel region image at the coordinate in the screen-on state, is the pixel brightness value of the pixel region image at the coordinate in the screen-off state, is the brightness balance coefficient, is the non-linear suppression factor.

3. The flat panel screen quality monitoring system based on image analysis and processing according to claim 1, wherein, The defect density detection is used to evaluate the overall defect density of the screen, through the formula: Among them, is the defect density index, and the number of defective pixels is the number of coordinates of defective pixels detected within the divided pixel region. is the total number of pixels on the screen.

4. A flat screen quality monitoring system based on image analysis and processing according to claim 3, characterized in that, The regional continuity detection is used to limit the maximum connected defect area of scratch-like defects, through the formula: Among them, is the regional continuity detection index, is the number of pixels in a single defect area, is the total screen area.

5. The flat screen quality monitoring system based on image analysis and processing according to claim 4, characterized in that The dynamic response detection is used to monitor the decay trend of the screen performance, through the formula: Among them, is a dynamic response detection index, is the number of consecutive detections, is the average value of the differential graph data in historical detections.

6. A flat screen quality monitoring system based on image analysis and processing according to claim 2, characterized in that, The said used to eliminate the environmental light interference and obtained by the following formula:

7. The flat panel screen quality monitoring system based on image analysis and processing according to claim 2, characterized in that, The non-linear suppression factor is used to prevent distortion in overexposed areas. Depending on different types of screens, the parameters are also different accordingly, as follows: LCD screen, =0.02 OLED screen = 0.03 8. An on - image - analysis - processing - based flat - screen quality monitoring system according to claim 5, characterized in that, The defect density index Threshold when facing the LCD screen , Threshold when facing the OLED screen , The regional continuity index Threshold , The dynamic response index Threshold .

Images