Mobile phone screen quality monitoring system based on image recognition

Abstract

The invention discloses a mobile phone screen quality monitoring system based on image recognition, which belongs to the technical field of screen detection, and comprises a quality detection platform for constructing a detection scene to detect a mobile phone screen; the image acquisition module is used for acquiring an image of a to-be-detected mobile phone screen; the analysis module comprises an appearance analysis module and a performance analysis module and is used for processing the acquired image so as to detect the appearance quality and the performance quality of the mobile phone screen in combination with the image information; and the display module is used for displaying the analysis result to testers. The mobile phone screen defect can be automatically judged through the image processing technology, manual judgment is not needed, the omission ratio can be effectively reduced, meanwhile, the detection efficiency is guaranteed, the defect type can be automatically judged after the defect is detected, and therefore the overall detection efficiency can be greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of screen detection, and in particular relates to a mobile phone screen quality monitoring system based on image recognition. Background Art

[0002] Mobile phone screens are typically inspected for quality before leaving the factory, and are only released after meeting factory standards. Traditional methods for inspecting mobile phone screen quality rely on manual visual inspection or simple instrumentation. Manual inspection methods have significant drawbacks. Inspectors easily become fatigued after working for long periods of time, which leads to decreased concentration and a higher rate of missed detection of minor defects. Furthermore, subjective judgment standards vary among inspectors, making it difficult to ensure the consistency and accuracy of inspection results. Furthermore, manual inspection is inefficient and cannot meet the rapid inspection requirements of large-scale production and recycling scenarios. Simple instrument-assisted inspections, such as those that only detect basic screen indicators such as brightness and color, cannot fully cover the complex types of defects that may occur on the screen, making it difficult to provide a comprehensive and accurate assessment of screen quality. Summary of the Invention

[0003] The purpose of the present invention is to provide a mobile phone screen quality monitoring system based on image recognition to solve the problems faced in the above-mentioned background technology.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] A mobile phone screen quality monitoring system based on image recognition, the quality monitoring system comprising:

[0006] A quality inspection platform, which is used to build inspection scenarios and inspect mobile phone screens;

[0007] An image acquisition module, configured to acquire an image of the screen of the mobile phone to be detected;

[0008] An analysis module, comprising an appearance analysis module and a performance analysis module. The appearance analysis module analyzes and processes the acquired image to detect the appearance quality. The performance analysis module analyzes and processes the acquired image to detect the performance quality.

[0009] The display module is used to display and count the analysis results to the testers.

[0010] Furthermore, the working method of the appearance analysis module is:

[0011] Place the mobile phone screen on a uniform backlight panel and use a high-precision camera to shoot from multiple angles to obtain images of the mobile phone screen;

[0012] The images at each angle are denoised and corrected, and compared with the preset standard images at each angle. If the two images are inconsistent, it is determined that there are defects in the appearance.

[0013] Furthermore, the appearance analysis module is also used to perform defect detection on the mobile phone screen that is judged to have defects in appearance, and the detection method is:

[0014] The Canny algorithm is used to extract the outline of the defective area of ​​the mobile phone screen from the defective image, and the area S, perimeter W and length L of the outline are calculated. , the defect is considered to be a crack or scratch, otherwise the defect is considered to be a spot or stain;

[0015] At the same time, the defect area appears by image acquisition The number of times, thereby determining the density f, when , the defect is determined to be a crack, otherwise the defect is determined to be a scratch;

[0016] Obtaining the shape feature values ​​of the defect area through edge recognition technology , the larger the shape eigenvalue, the more regular the shape, and in the HSV space, obtain the color eigenvalue of the defect area ,The larger the color feature value, the richer the color of the defect area. When , the defect is judged to be a stain, otherwise it is judged to be a spot;

[0017] in, 、 as well as are their respective defect judgment thresholds.

[0018] Furthermore, the working method of the performance analysis module is:

[0019] When it is determined that there is no defect in the appearance, the performance deviation value of the mobile phone screen in the continuous charging state and the non-continuous charging state is obtained. 、 ;

[0020] By formula Get the performance status value ;

[0021] when When the screen is faulty, it is judged that there is a defect in the performance of the mobile phone screen;

[0022] in, as well as is the weight coefficient, The performance defect judgment threshold.

[0023] Furthermore, the performance deviation value as well as The acquisition method is:

[0024] Obtain various detection indicators of the mobile phone screen in the non-continuous charging state , through the formula The performance deviation value under non-continuous charging state is obtained, where n is the total number of detection index items. is the ideal value of the i-th detection indicator under non-continuous charging state, is the weight of the i-th detection index in the non-continuous charging state, ;

[0025] Under continuous charging state, obtain various detection indicators of the mobile phone screen , through the formula Obtain the performance deviation value under continuous charging state , is the ideal value of the i-th detection indicator under continuous charging state, is the weight of the i-th detection indicator under continuous charging state.

[0026] Furthermore, the working method of the performance analysis module also includes:

[0027] When it is determined that there is no defect in the performance of the mobile phone screen, the performance status value of the mobile phone screen is obtained m times continuously , thus obtaining the stability coefficient ,when When the phone is detected, it is also judged that there are defects in the performance of the mobile phone screen;

[0028] in, is the performance status value of the mobile phone screen obtained for the jth time, , is the stability coefficient judgment threshold.

[0029] Furthermore, the display module works as follows:

[0030] When it is judged that there is a defect in the appearance, the appearance defect will be displayed and an alarm will be issued, and the specific defect type will be displayed at the same time;

[0031] When it is judged that there is a performance defect, the performance defect will be displayed and an alarm will be issued. At the same time, the detection items with obvious abnormal detection indicators will be displayed;

[0032] The number of appearance defects, the number of performance defects and the proportion of each defect type in the entire inspection batch are counted to judge the condition of the mobile phone screens in the entire batch.

[0033] Beneficial effects of the present invention:

[0034] The present invention can automatically judge the defects of mobile phone screens through image processing technology, which does not require manual judgment and can effectively reduce the missed detection rate while ensuring detection efficiency. Moreover, after the defects are detected, the defect type can be automatically judged, which can greatly improve the overall efficiency of detection.

[0035] The present invention can perform a comprehensive analysis based on multiple detection indicators of the mobile phone screen under continuous charging conditions and non-continuous charging conditions to more accurately detect the performance of the mobile phone screen and improve the accuracy of the detection.

[0036] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 This is a system structure diagram of the present invention. DETAILED DESCRIPTION

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0040] In one embodiment, a mobile phone screen quality monitoring system based on image recognition is disclosed. Figure 1 As shown, the quality monitoring system includes:

[0041] Quality testing platform, which is used to build testing scenarios and test mobile phone screens;

[0042] An image acquisition module is used to acquire an image of the mobile phone screen to be detected;

[0043] An analysis module, including an appearance analysis module and a performance analysis module. The appearance analysis module analyzes and processes the acquired images to detect the appearance quality. The performance analysis module analyzes and processes the acquired images to detect the performance quality.

[0044] Display module: The display module is used to display and count the analysis results to the testers.

[0045] Through the above technical solution, the present application first constructs a detection scene of a mobile phone screen to facilitate the detection of the mobile phone screen, including the appearance detection and performance detection of the mobile phone screen, and then obtains the image of the mobile phone screen to be detected through the image acquisition module and uploads it to the analysis module. After the analysis module processes the image information, the appearance of the mobile phone screen is detected by the appearance analysis module to determine whether there are defects in the appearance of the mobile phone screen. At the same time, the performance quality of the mobile phone screen is detected according to the performance analysis module to determine whether there are defects in the performance of the mobile phone, and finally displayed by the display module. In this way, it does not require manual detection, and can automatically detect the appearance defects of the mobile phone screen according to the image recognition technology, greatly reducing the occurrence of missed detection, and can automatically identify the type of defects, and does not require manual judgment, which can effectively improve the efficiency of detection; at the same time, when detecting the performance of the mobile phone screen, a comprehensive analysis can be performed based on the performance of the mobile phone screen in the continuous charging state and the non-continuous charging state, so that the performance of the mobile phone screen can be detected more accurately, which can effectively improve the detection accuracy.

[0046] The working method of the appearance analysis module is as follows: the mobile phone screen is placed on a uniform backlight plate and a high-precision camera is used to shoot images from multiple angles to obtain images of the mobile phone screen;

[0047] The images at each angle are denoised and corrected, and compared with the preset standard images at each angle. If the two images are inconsistent, it is determined that there are defects in the appearance.

[0048] The visual analysis module is also used to perform defect detection on mobile phone screens that are judged to have defects in appearance. The detection method is as follows:

[0049] The Canny algorithm is used to extract the outline of the defective area of ​​the mobile phone screen from the defective image, and the area S, perimeter W and length L of the outline are calculated. , the defect is considered to be a crack or scratch, otherwise the defect is considered to be a spot or stain;

[0050] At the same time, the defect area appears by image acquisition The number of times, thereby determining the density f, when , the defect is determined to be a crack, otherwise the defect is determined to be a scratch;

[0051] Obtaining the shape feature values ​​of the defect area through edge recognition technology , the larger the shape eigenvalue, the more regular the shape, and in the HSV space, obtain the color eigenvalue of the defect area ,The larger the color feature value, the richer the color of the defect area. When , the defect is judged to be a stain, otherwise it is judged to be a spot;

[0052] in, 、 as well as They are respective defect judgment thresholds, which can be formulated based on empirical data.

[0053] The above scheme provides a specific method for the appearance analysis module to judge the appearance defects of the mobile phone screen. First, the mobile phone screen is placed on a uniform backlight plate, and a high-precision camera is used to shoot at multiple angles to obtain multiple images of the mobile phone screen. In this way, the image information of the mobile phone screen can be fully covered. Then, the images at each angle are subjected to denoising, correction and other processing operations to obtain the processed image, so as to realize the acquisition of the image of the corresponding angle of the standard intact mobile phone screen as the preset standard image, and then the images obtained at each angle are compared with each preset standard image respectively. If they are completely consistent, it means that the appearance of the mobile phone screen is intact. When the two images are inconsistent, it means that the appearance of the mobile phone screen has defects, and it is judged that the appearance has defects. In this way, the screen with appearance defects and the intact screen can be quickly and automatically distinguished. It does not require manual distinction, has high accuracy, and can improve detection efficiency. When it is judged that there are defects in the appearance, the Canny algorithm is used to extract the contour of the defective area of ​​the mobile phone screen from the defective image, and the area S, perimeter W and length L of the contour are calculated. If , the defect is considered to be a crack or scratch, otherwise it is considered to be a spot or stain; since cracks or scratches are generally characterized by being elongated, with a small area but a large perimeter, while spots or stains are characterized by being irregular in shape and relatively large in area, the contour of the defect area is obtained, and the contour area S, perimeter W, and length L are calculated. , indicating that the perimeter is large and the length is long, which is more likely to be a crack or scratch. Otherwise, it is more likely to be a spot or stain. Therefore, when it is judged to be a crack or scratch defect, the image processing technology is used to continue to obtain the defect area. The number of times, thereby determining the density f, when , then the defect is determined to be a crack, otherwise it is determined to be a scratch. Since scratches are generally single, while screen cracks are divergent, if the number of scratches or cracks in the defect area is large, then the density of the area must be high, and then it is judged that the defect is likely to be a crack. Otherwise, the density is low, indicating that it is likely to be a scratch. This can quickly distinguish between scratches and cracks. Similarly, when judging spot defects and stain defects, the shape feature value of the defect area is obtained through edge recognition technology. , the larger the shape eigenvalue, the more regular the shape, and in the HSV space, obtain the color eigenvalue of the defect area ,The larger the color feature value, the richer the color of the defect area. When , the defect is judged to be a stain, otherwise it is judged to be a spot; Formula It is expressed as the ratio of shape feature value to color feature value. The larger the ratio, the more regular the shape and the single color. The smaller the ratio, the more irregular the shape and the more diverse the color. Generally speaking, spots are usually single color or close to the screen background color, while stains are multi-colored. Similarly, spots are generally regular, while stains are irregular. Therefore, when If the defect is a stain, it is judged as a stain; otherwise, it is judged as a spot. This allows for quick differentiation between spot and stain defects. In this way, image processing technology can automatically identify defects on mobile phone screens, eliminating the need for manual judgment and effectively reducing missed detection rates while ensuring detection efficiency. Furthermore, after a defect is detected, the defect type can be automatically determined, significantly improving overall detection efficiency.

[0054] The working method of the performance analysis module is: when it is judged that there is no defect in the appearance, the performance deviation value of the mobile phone screen in the continuous charging state and the non-continuous charging state is obtained. 、 ;

[0055] By formula Get the performance status value ;

[0056] when When the screen is faulty, it is judged that there is a defect in the performance of the mobile phone screen;

[0057] in, as well as is the weight coefficient, is the performance defect judgment threshold, and the performance deviation value as well as The acquisition method is: in the non-continuous charging state, obtain the various detection indicators of the mobile phone screen , through the formula The performance deviation value under non-continuous charging state is obtained, where n is the total number of detection index items. is the ideal value of the i-th detection indicator under non-continuous charging state, is the weight of the i-th detection index in the non-continuous charging state, ;

[0058] Under continuous charging state, obtain various detection indicators of the mobile phone screen , through the formula Obtain the performance deviation value under continuous charging state , is the ideal value of the i-th detection indicator under continuous charging state, is the weight of the i-th detection indicator under continuous charging state.

[0059] Through the above technical solution, this embodiment provides a specific working method of the performance analysis module. First, when it is determined that there is no defect in the appearance, various detection indicators of the mobile phone screen are obtained under continuous charging state. , through the formula Obtain the performance deviation value under continuous charging state , is the ideal value of the i-th detection indicator under continuous charging state, is the weight of the i-th detection index in the continuous charging state; then, in the non-continuous charging state, obtain the various detection indicators of the mobile phone screen , through the formula The performance deviation value under non-continuous charging state is obtained, where n is the total number of detection index items. is the ideal value of the i-th detection indicator under non-continuous charging state, is the weight of the i-th detection index in the non-continuous charging state. In the above, the ideal value of each detection item can be determined according to the detection standard, and the weight coefficient of each item can be determined artificially based on experience; in the continuous charging state and the non-continuous charging state of the mobile phone screen, multiple simulation tests are performed on the mobile phone screen, and the image in the model test is obtained through image technology, and the image is analyzed to obtain various detection indicators. For example, a full white screen is displayed and the entire screen is photographed, and the average brightness level of each area is calculated using an image processing algorithm to obtain a brightness uniformity detection index; by quickly switching colors (such as alternating between black and white), the grayscale change speed during the transition is recorded to determine the response time detection index, and a specific test pattern (such as a thin line or dot matrix) displayed on the screen is photographed by a high-resolution camera, and then the detail clarity in the image is analyzed to obtain a pixel density detection index; then each detection index is compared with the corresponding ideal value. If the difference is smaller, it means that the performance index of the detection item is better. If the difference is larger, the performance index of the detection item is worse. If it exceeds a certain value, the detection item is judged to be abnormal. Finally, the formula as well as Comprehensive analysis to obtain performance deviation value 、 , after assigning the corresponding weights to the two, through the formula Get the overall performance value of the mobile phone screen , we can see that the smaller the value, the smaller the performance of the mobile phone screen, so when When , it is judged that there is a defect in the performance of the mobile phone screen, where The performance defect judgment threshold is determined based on empirical data. This method allows for a more accurate performance test of the phone screen based on a comprehensive analysis of various test indicators under both continuous and discontinuous charging conditions, thereby improving detection accuracy.

[0060] The working method of the performance analysis module also includes: when it is determined that there is no defect in the performance of the mobile phone screen, obtaining the performance status value of the mobile phone screen for m consecutive times , thus obtaining the stability coefficient ,when When the phone is detected, it is also judged that there are defects in the performance of the mobile phone screen;

[0061] in, is the performance status value of the mobile phone screen obtained for the jth time, , is the stability coefficient judgment threshold.

[0062] Through the above technical solution, this embodiment provides another working method of the performance analysis module, which is to obtain the performance status value of the mobile phone screen m times in a row when it is judged that there is no defect in the performance of the mobile phone screen. , thus obtaining the stability coefficient ,when When the phone screen is detected, it is also judged that there are defects in the performance of the phone screen; is the performance status value of the mobile phone screen obtained for the jth time, The stability coefficient judgment threshold is determined based on empirical data. The performance status values ​​of the mobile phone screen are collected multiple times for comprehensive analysis to obtain the overall performance fluctuation. Generally speaking, a mobile phone screen with good performance has a small difference in performance status values ​​during each test. The smaller the set fluctuation, the greater the stability coefficient value. The larger the value, the greater the fluctuation of the performance status value obtained in each test, and the more unstable the performance of the mobile phone screen. In this way, it can not only further test the performance of the mobile phone screen, but also judge the potential problems of the mobile phone screen based on the fluctuation, so as to further improve the accuracy of the test and ensure the quality of the mobile phone screen.

[0063] The working method of the display module is as follows: when it is determined that there is a defect in the appearance, the appearance defect is displayed and an alarm is issued, and the specific defect type is displayed at the same time;

[0064] When it is judged that there is a performance defect, the performance defect will be displayed and an alarm will be issued. At the same time, the detection items with obvious abnormal detection indicators will be displayed;

[0065] The number of appearance defects, the number of performance defects and the proportion of each defect type in the entire inspection batch are counted to judge the condition of the mobile phone screens in the entire batch.

[0066] Through the above technical solution, this embodiment provides a specific working method of the display module. When it is judged that there are defects in the appearance, the appearance defects are displayed and an alarm reminder is issued, and the specific defect type is displayed at the same time, so that the tester can distinguish the defective mobile phone screen and quickly know the defect type, which is convenient for subsequent processing; at the same time, when it is judged that there are defects in the performance, the performance defects are displayed and an alarm reminder is issued, and the detection items with obviously abnormal detection indicators are displayed. In this way, the detection items with obviously abnormal detection indicators can be displayed, which is convenient for management personnel to know which detection indicator is likely to be abnormal when the mobile phone screen performance has defects, which is convenient for management; at the same time, the number of appearance defects, the number of performance defects and the proportion of various defect types in the entire detection batch are counted, so as to judge the condition of the mobile phone screens of the entire batch. If the number of defects is high or the proportion of defect types is high, it means that the quality of the mobile phone screen during overall production is low. In this way, subsequent production can be adjusted and controlled in time according to the cause of the defect to ensure production quality.

[0067] It should be noted that, in order to facilitate analysis and calculation, all parameters in the above calculation formula are dimensionless operations performed after the units are selected.

[0068] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A mobile phone screen quality monitoring system based on image recognition, characterized in that: The quality monitoring system includes: A quality inspection platform, which is used to build inspection scenarios and inspect mobile phone screens; An image acquisition module, configured to acquire an image of the screen of the mobile phone to be detected; An analysis module, comprising an appearance analysis module and a performance analysis module. The appearance analysis module analyzes and processes the acquired image to detect the appearance quality. The performance analysis module analyzes and processes the acquired image to detect the performance quality. The display module is used to display and count the analysis results to the testers.

2. The mobile phone screen quality monitoring system based on image recognition according to claim 1, characterized in that: The working method of the appearance analysis module is: Place the mobile phone screen on a uniform backlight panel and use a high-precision camera to shoot from multiple angles to obtain images of the mobile phone screen; The images at each angle are denoised and corrected, and compared with the preset standard images at each angle. If the two images are inconsistent, it is determined that there are defects in the appearance.

3. The mobile phone screen quality monitoring system based on image recognition according to claim 2, characterized in that: The appearance analysis module is also used to perform defect detection on mobile phone screens that are judged to have appearance defects, and the detection method is: The Canny algorithm is used to extract the outline of the defective area of ​​the mobile phone screen from the defective image, and the area S, perimeter W and length L of the outline are calculated. , the defect is considered to be a crack or scratch, otherwise the defect is considered to be a spot or stain; At the same time, the defect area appears by image acquisition The number of times, thereby determining the density f, when , the defect is determined to be a crack, otherwise the defect is determined to be a scratch; Obtaining the shape feature values ​​of the defect area through edge recognition technology , the larger the shape eigenvalue, the more regular the shape, and in the HSV space, obtain the color eigenvalue of the defect area ,The larger the color feature value, the richer the color of the defect area. When , the defect is judged to be a stain, otherwise it is judged to be a spot; in, 、 as well as are their respective defect judgment thresholds.

4. The mobile phone screen quality monitoring system based on image recognition according to claim 1, characterized in that: The working method of the performance analysis module is: When it is determined that there is no defect in the appearance, the performance deviation value of the mobile phone screen in the continuous charging state and the non-continuous charging state is obtained. 、 ; By formula Get the performance status value ; when When the screen is faulty, it is judged that there is a defect in the performance of the mobile phone screen; in, as well as is the weight coefficient, The performance defect judgment threshold.

5. The mobile phone screen quality monitoring system based on image recognition according to claim 4, characterized in that: The performance deviation value as well as The acquisition method is: Obtain various detection indicators of the mobile phone screen in the non-continuous charging state , through the formula The performance deviation value under non-continuous charging state is obtained, where n is the total number of detection index items. is the ideal value of the i-th detection indicator under non-continuous charging state, is the weight of the i-th detection index in the non-continuous charging state, ; Under continuous charging state, obtain various detection indicators of the mobile phone screen , through the formula Obtain the performance deviation value under continuous charging state , is the ideal value of the i-th detection indicator under continuous charging state, is the weight of the i-th detection indicator under continuous charging state.

6. The mobile phone screen quality monitoring system based on image recognition according to claim 4, characterized in that: The working method of the performance analysis module also includes: When it is determined that there is no defect in the performance of the mobile phone screen, the performance status value of the mobile phone screen is obtained m times continuously , thus obtaining the stability coefficient ,when When the phone is detected, it is also judged that there are defects in the performance of the mobile phone screen; in, is the performance status value of the mobile phone screen obtained for the jth time, , is the stability coefficient judgment threshold.

7. The mobile phone screen quality monitoring system based on image recognition according to claim 1, characterized in that: The working method of the display module is: When it is judged that there is a defect in the appearance, the appearance defect will be displayed and an alarm will be issued, and the specific defect type will be displayed at the same time; When it is judged that there is a performance defect, the performance defect will be displayed and an alarm will be issued. At the same time, the detection items with obvious abnormal detection indicators will be displayed; The number of appearance defects, the number of performance defects and the proportion of each defect type in the entire inspection batch are counted to judge the condition of the mobile phone screens in the entire batch.

Images