Mobile phone screen dead pixel detection method and system and electronic equipment

Abstract

The invention provides a mobile phone screen dead pixel detection method and system and electronic equipment, and relates to the technical field of visual inspection, and the method comprises the following steps: carrying out dead pixel detection importance evaluation on a mobile phone screen; based on the protection layer detection weight, driving the dead pixel detection first channel to carry out dead pixel detection on the protection layer; based on the touch layer detection weight, dead pixel detection is carried out on the touch layer according to the dead pixel detection second channel; carrying out dead pixel detection on the display layer according to a dead pixel detection third channel based on the display layer detection weight; inputting the protective layer dead pixel detection result, the touch layer dead pixel detection result and the display layer dead pixel detection result into a screen quality adaptation detection channel; and when the screen quality adaptation inspection result is qualified, generating a screen assembly instruction. According to the method and the device, the technical problem of relatively low accuracy of screen dead pixel detection due to lack of detection of a multi-layer structure of the screen in the prior art can be solved, and the accuracy of screen dead pixel detection is improved through layered detection.

Description

Technical Field

[0001] The present application relates to the field of visual detection technology, and in particular to a method, system and electronic device for detecting bad pixels on a mobile phone screen. Background Art

[0002] Bad pixels usually refer to abnormal pixels on the screen, including bright spots (pixels that emit abnormal light), dark spots (pixels that do not emit light), and dead spots (completely unresponsive). At present, the method of detecting bad pixels on mobile phone screens uses high-resolution cameras or optical sensors to capture screen display images to detect bad pixels on the display layer, such as dead spots, bright spots, and light leakage. However, most methods only focus on one layer of the screen, or take the same detection measures for different layers, which is prone to false detection or missed detection, and it is difficult to fully cover all possible types of screen defects, thus affecting the detection accuracy.

[0003] In summary, the prior art has a technical problem that, because detection is usually performed on a single layer, the multi-layer structure of the screen cannot be covered, resulting in some bad pixels being missed, further affecting the accuracy of bad pixel detection on the mobile phone screen. Summary of the invention

[0004] The purpose of this application is to provide a method, system and electronic device for detecting bad pixels on a mobile phone screen, so as to solve the technical problem in the prior art that the detection is usually performed on a single layer and fails to cover the multi-layer structure of the screen, resulting in some bad pixels being missed, further affecting the accuracy of bad pixel detection on the mobile phone screen.

[0005] In view of the above problems, the present application provides a method, system and electronic device for detecting bad pixels on a mobile phone screen.

[0006] In a first aspect, the present application provides a method for detecting bad pixels on a mobile phone screen, which is implemented by a mobile phone screen bad pixel detection system, wherein the method for detecting bad pixels on a mobile phone screen comprises: evaluating the importance of bad pixel detection according to a target mobile phone screen, obtaining a protective layer detection severity, a touch layer detection severity and a display layer detection severity, wherein the target mobile phone screen comprises a protective layer, a touch layer and a display layer; based on the protective layer detection severity, driving a first bad pixel detection channel to perform bad pixel detection on the protective layer, and obtaining a protective layer bad pixel detection result; based on the touch .... Measure the weight, perform bad pixel detection on the touch layer according to the second bad pixel detection channel, and obtain the bad pixel detection result of the touch layer; based on the display layer detection weight, perform bad pixel detection on the display layer according to the third bad pixel detection channel, and obtain the bad pixel detection result of the display layer; input the bad pixel detection result of the protective layer, the bad pixel detection result of the touch layer, and the bad pixel detection result of the display layer into the screen quality adaptation inspection channel to obtain the screen quality adaptation inspection result; when the screen quality adaptation inspection result is qualified, generate a screen assembly instruction, and execute the assembly of the target mobile phone screen according to the screen assembly instruction.

[0007] In a second aspect, the present application further provides a mobile phone screen bad pixel detection system, which is used to execute a mobile phone screen bad pixel detection method as described in the first aspect, wherein the mobile phone screen bad pixel detection system includes: an importance evaluation module, the importance evaluation module is used to perform bad pixel detection importance evaluation according to the target mobile phone screen, and obtain the protection layer detection severity, touch layer detection severity and display layer detection severity, and the target mobile phone screen includes a protection layer, a touch layer and a display layer; a first detection module, the first detection module is used to drive the bad pixel detection first channel to perform bad pixel detection on the protection layer based on the protection layer detection severity, and obtain the protection layer bad pixel detection result; a second detection module, the second detection module is used to detect the bad pixel based on the touch layer detection severity, Perform bad pixel detection on the touch layer according to the second bad pixel detection channel to obtain a bad pixel detection result of the touch layer; a third detection module, the third detection module is used to perform bad pixel detection on the display layer according to the third bad pixel detection channel based on the display layer detection severity to obtain a bad pixel detection result of the display layer; an adaptation verification module, the adaptation verification module is used to input the bad pixel detection result of the protection layer, the bad pixel detection result of the touch layer and the bad pixel detection result of the display layer into the screen quality adaptation inspection channel to obtain a screen quality adaptation inspection result; an assembly instruction generation module, the assembly instruction generation module is used to generate a screen assembly instruction when the screen quality adaptation inspection result is qualified, and execute the assembly of the target mobile phone screen according to the screen assembly instruction.

[0008] In a third aspect, the present application also provides an electronic device comprising: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the steps of a method for detecting bad pixels on a mobile phone screen as described in any one of the first aspects above.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages: By evaluating the importance of bad pixel detection according to the target mobile phone screen, the protection layer detection severity, touch layer detection severity and display layer detection severity are obtained, and the target mobile phone screen includes a protection layer, a touch layer and a display layer; based on the protection layer detection severity, the first bad pixel detection channel is driven to perform bad pixel detection on the protection layer to obtain the protection layer bad pixel detection result; based on the touch layer detection severity, the touch layer is performed with a bad pixel detection according to a second bad pixel detection channel to obtain the touch layer bad pixel detection result; based on the display layer detection severity, the display layer is performed with a bad pixel detection according to a third bad pixel detection channel to obtain the display layer bad pixel detection result; the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result are input into the screen quality adaptation inspection channel to obtain the screen quality adaptation inspection result; when the screen quality adaptation inspection result is qualified, a screen assembly instruction is generated, and the assembly of the target mobile phone screen is performed according to the screen assembly instruction. That is to say, by evaluating the importance of testing the three levels of the target mobile phone screen, the detection severity of each level is determined, and according to the detection severity of each level, a dedicated detection channel is used to detect the bad pixels of the corresponding level. The bad pixel detection results of the three levels are input into a comprehensive inspection channel to obtain the inspection results. When the inspection results are qualified, screen assembly instructions are generated to assemble the target mobile phone screen, thereby improving the accuracy of bad pixel detection and thus improving the overall quality of the mobile phone screen.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 A flowchart of a method for detecting bad pixels on a mobile phone screen is provided for this application; Figure 2 A schematic diagram of the structure of a mobile phone screen bad pixel detection system for this application; Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of the present application.

[0013] Explanation of the accompanying drawings: importance evaluation module 11, first detection module 12, second detection module 13, third detection module 14, adaptation verification module 15, assembly instruction generation module 16, bus 300, receiver 301, processor 302, transmitter 303, memory 304, bus interface 305. DETAILED DESCRIPTION

[0014] This application provides a method, system and electronic device for detecting bad pixels on mobile phone screens, which solves the technical problem in the prior art that some bad pixels are missed due to the failure to cover the multi-layer structure of the screen because the detection is usually carried out on a single layer, which further affects the accuracy of the detection of bad pixels on the mobile phone screen. By evaluating the importance of detection of the three layers of the target mobile phone screen, the detection weight of each layer is determined, and according to the detection weight of each layer, a special detection channel is used to detect the bad pixels of the corresponding layer, and the bad pixel detection results of the three layers are input into the comprehensive inspection channel to obtain the inspection results. When the inspection results are qualified, the screen assembly instructions are generated to assemble the target mobile phone screen, which improves the accuracy of bad pixel detection and thus improves the overall quality of the mobile phone screen.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For example, please refer to the attached Figure 1The present application provides a method for detecting bad pixels on a mobile phone screen, wherein the method is applied to a mobile phone screen bad pixel detection system, and the method specifically comprises the following steps: S100: evaluating the importance of bad pixel detection according to a target mobile phone screen to obtain a protection layer detection severity, a touch layer detection severity and a display layer detection severity, wherein the target mobile phone screen includes a protection layer, a touch layer and a display layer.

[0017] Specifically, the target mobile phone screen includes a protective layer, a touch layer and a display layer, wherein the protective layer refers to the outermost layer of the screen, which is responsible for resisting external physical damage (such as scratches, impacts) and protecting internal components from physical damage or contamination; the touch layer is located under the protective layer, sensing the capacitive signal of the user's touch and converting it into input instructions; the display layer is located under the touch layer, used to present visual information (images, text, etc.), and is usually composed of an LCD or OLED screen. The importance of bad pixel detection on the three layers of the target mobile phone screen is usually evaluated based on the specific functions of each layer and the impact of defects on the overall quality.

[0018] According to the degree of damage caused by bad pixels to the main functions of the screen (such as display effect or touch sensitivity), the probability of bad pixels, the cost of repair or replacement, etc., the importance of bad pixel detection is evaluated to obtain the detection severity of the interlayer. The detection severity of the protective layer refers to the degree of attention paid to the defects of the protective layer when performing bad pixel detection. The higher the detection severity of the protective layer, the more image data of the protective layer is collected; the detection severity of the touch layer refers to the degree of attention paid to the defects of the touch layer when performing bad pixel detection. The greater the detection severity of the touch layer, the more touch operations are performed; the detection severity of the display layer refers to the degree of attention paid to the defects of the display layer when performing bad pixel detection. The greater the detection severity of the display layer, the more data is collected for the image of the display layer. Through the evaluation of the importance of bad pixel detection, the overall quality of the mobile phone screen can be accurately evaluated, and targeted bad pixel detection can be carried out according to the importance of different levels, which will help improve the accuracy and efficiency of detection.

[0019] S200: Based on the detection severity of the protection layer, driving a first bad pixel detection channel to perform bad pixel detection on the protection layer to obtain a bad pixel detection result of the protection layer.

[0020] Further, S200 of the present application includes: the first channel of bad pixel detection includes a visual camera, an image filter and a protective layer bad pixel detection model; based on the detection severity of the protective layer, the visual camera is driven to capture an image of the protective layer to obtain a protective layer captured image; the protective layer captured image is input into the image filter to obtain a protective layer enhanced image; the protective layer enhanced image is input into the protective layer bad pixel detection model to generate the protective layer bad pixel detection result.

[0021] Specifically, the first channel of bad pixel detection includes a visual camera, an image filter and a protective layer bad pixel detection model. The visual camera is a high-resolution image acquisition device used to obtain image data of the mobile phone screen; the image filter is an image processing tool designed to process the collected original image to improve the image quality and details to make it more suitable for subsequent analysis; the protective layer bad pixel detection model is a model trained based on machine learning or image analysis algorithms, which is used to analyze the processed images and identify bad pixels (such as scratches, cracks, etc.) therein.

[0022] The degree of image acquisition of the protective layer is determined based on the detection severity of the protective layer. If the detection severity of the protective layer is high, the visual camera will collect more image data to ensure full coverage of all potential bad pixel areas to capture more details. Visual cameras usually use high-resolution CCD or CMOS sensors, combined with appropriate light sources and shooting angles to ensure that the details of the protective layer surface can be clearly captured. The protective layer acquisition image refers to the image data collected by the visual camera, which reflects the status of the protective layer, including possible defects, scratches, cracks and other information on the protective layer.

[0023] After the image is collected, it is input to the image filter for processing. The image filter is used to remove noise or unnecessary interference in the image and enhance the bad pixel features. Select a suitable image filter, such as a non-local mean filter. Remove noise by comparing the similarities of different areas of the image while retaining more detailed information. De-noising and smoothing effects are achieved by calculating the similarity between different areas in the image and weighting the average pixel value according to the similarity. Adaptive search window technology is used to dynamically adjust the window size according to the characteristics of the image area to obtain multiple pixel blocks. For each pixel block, a similarity measure with other pixel blocks is calculated, such as the Euclidean distance. According to the similarity, the current pixel point is compared with other pixel points in its neighborhood, and different weights are assigned and weighted averaged. Similar pixel blocks have greater weights and non-similar pixel blocks have smaller weights.

[0024] The weighted average result is used as the filtering result of the current pixel to achieve the denoising effect. Since there may be many tiny scratches, particles or noise on the protective layer, by using non-local mean filtering, the details in the image (such as scratches and cracks) are retained and the noise is effectively removed. For example, the standard deviation of the noise in the original image is 15. After NLM filtering, the standard deviation of the noise is reduced to 5, and the details of scratches and cracks are more prominent, which is convenient for subsequent detection. After being processed by the non-local mean filter, the obtained protective layer enhanced image will be clearer and less noisy than the original image, while retaining the important bad pixel features in the original image.

[0025] Collect a large amount of sample data containing images of protective layers, covering normal areas and bad pixel areas. Each image will be marked with the location or type of bad pixels. De-noise the image, such as using mean filtering, bilateral filtering, or Gaussian filtering to remove background noise and retain bad pixel features. Perform operations such as cropping, scaling, or rotating the image to enhance the generalization ability of the model. Convert the image to a format suitable for deep learning model input (such as 224x224 pixel RGB image). Divide the processed image set into a training set and a validation set. Construct a convolutional neural network, including multiple convolutional layers, pooling layers, and fully connected layers. The convolutional layer is used to extract image features, the pooling layer is used to reduce dimensionality and reduce the computational burden, and the fully connected layer is used for classification.

[0026] The convolutional neural network is trained through the training set, and the network weights are optimized using the back propagation algorithm so that it can accurately identify bad pixels in the image. During the training process, a loss function (such as cross entropy loss) is used to measure the error between the model output and the actual label, and the network parameters are updated through an optimization algorithm (such as the Adam optimizer). According to the cross entropy loss function, the cross entropy loss is calculated and then summed to obtain the overall loss value. The trained model is evaluated through the validation set to check its performance on unseen data, including accuracy, precision, recall, etc. The trained model is used as a protective layer bad pixel detection model to detect bad pixels (such as scratches, cracks, etc.) on the protective layer of the mobile phone screen. By inputting an image and analyzing its features, the output image shows whether there are bad pixels and the location or type of the bad pixels.

[0027] The protection layer enhanced image processed by the graphic filter is input into the trained protection layer bad pixel detection model. The features in the image are gradually extracted through multiple convolutional layers and pooling layers, and the extracted features are input into the fully connected layer for classification. According to the input image, the bad pixel detection result is output. If a bad pixel is detected, the location (such as the coordinate box) and type of the bad pixel are output; if no bad pixel is detected, no bad pixel is output. The amount of collected image data is dynamically adjusted through the protection layer detection weight to ensure that the high-weight area obtains sufficient image coverage, thereby effectively improving the comprehensiveness of bad pixel detection. The collected image is processed using an image filter to effectively remove noise and highlight the bad pixel features. The enhanced image is analyzed by the trained bad pixel detection model, which can automatically identify bad pixels, reduce manual intervention, and improve detection efficiency and accuracy.

[0028] S300: Based on the detection severity of the touch layer, perform bad pixel detection on the touch layer according to a second bad pixel detection channel to obtain a bad pixel detection result of the touch layer.

[0029] Further, S300 of the present application includes: the second channel of bad pixel detection includes a touch robot and a touch layer bad pixel recognition model; based on the touch layer detection severity, the touch robot performs multiple touch operations on the touch layer to obtain multiple groups of touch response data; the multiple groups of touch response data are input into the touch layer bad pixel recognition model to obtain multiple groups of touch layer bad pixel recognition results; the multiple groups of touch layer bad pixel recognition results are integrated to generate the touch layer bad pixel detection result.

[0030] Specifically, the second bad pixel detection channel is a channel specifically used to detect bad pixels in the touch layer, including a touch robot and a touch layer bad pixel recognition model. A touch robot is an automated device used to simulate human touch operations. It can usually accurately perform a series of operations on the touch screen (such as clicking, sliding, pressing, etc.) and collect touch response data. The touch layer bad pixel recognition model is similar to the protective layer bad pixel detection model. Both are used to detect and identify bad pixels. The difference is that the touch layer bad pixel recognition model is applied to the touch layer, and the bad pixel recognition result is obtained based on the touch response data.

[0031] According to the touch layer detection severity of the target mobile phone screen, it is determined how many touch operations are needed. The greater the touch layer detection severity, the more potential problems there are in the touch layer or the more accurate the detection is required. Therefore, the touch robot needs to perform more touch operations. According to the touch layer detection severity, the touch robot performs a series of touch operations, including clicking, sliding, long pressing, etc. Each touch operation will record touch response data to reflect the reaction of the touch layer. Touch response data refers to the response data collected by the touch robot when performing touch operations, including information such as the location, strength, and response time of the touch operation.

[0032] The acquired multiple sets of touch response data are input into the trained touch layer bad pixel recognition model. The model analyzes the response data and identifies whether there are bad pixels (such as dead pixels, failure pixels, etc.) on the touch layer. The touch layer bad pixel recognition model is a model based on machine learning or deep learning, which is specially used to detect and identify bad pixels (such as dead pixels, drift pixels, etc.) on the touch layer of mobile phone screens. By analyzing the touch response data, it determines whether there are defects in the touch layer and identifies the type or location of these defects.

[0033] The training process of the touch layer bad pixel recognition model is similar to that of the protective layer bad pixel detection model. For the simplicity of the manual, the training process is not repeated. The touch layer bad pixel recognition model processes each set of touch response data and finally generates a bad pixel recognition result for each set of data. Each recognition result indicates whether there is a bad pixel at the touch position. If there is a bad pixel, the coordinates or specific type of the bad pixel are usually returned. Multiple sets of touch layer bad pixel recognition results are fused, usually using a data fusion algorithm, such as voting method, weighted average method, maximum probability method, etc. The purpose is to improve the reliability of bad pixel detection by integrating multiple sets of data. All recognition results are integrated. If a certain area is identified as a bad pixel in multiple operations, then this area is more likely to have a real bad pixel. The result of each touch operation is a bad pixel or normal. After multiple touch operations, the voting method is used, and the point with the most votes is the bad pixel.

[0034] The final bad pixel detection results are generated through the fused data, including the location, type (dead pixel, drift point, etc.) and possible impact range of each bad pixel. Based on the fused information, the final bad pixel detection report is formed, which may include detailed data such as specific bad pixel coordinates, type, area, etc. for subsequent processing or screen repair. By fusing the results of multiple touch operations, it is possible to effectively eliminate misjudgments caused by occasional errors or deficiencies in a single touch operation. For example, if a location is displayed as a bad pixel only in one operation, but is displayed as normal in other multiple touch operations, the location will be identified as normal.

[0035] Through multiple touch operations and data fusion, the error caused by a single operation can be reduced, and the accuracy of bad pixel detection can be significantly improved. Even if there is an error in a touch response data, other data can make up for it, and finally a more reliable bad pixel detection result can be obtained. The setting of the touch layer detection weight allows the detection to be flexibly adjusted to meet the detection needs of different screens.

[0036] S400: Based on the detection severity of the display layer, perform bad pixel detection on the display layer according to a third bad pixel detection channel to obtain a bad pixel detection result of the display layer.

[0037] Specifically, the parameters of the third channel of bad pixel detection (including the camera device, the image enhancer and the K mode display bad pixel recognition models) are adjusted according to the display layer detection severity to achieve accurate detection of bad pixels in the display layer. According to the display layer detection severity, the camera device is used to collect images of the display layer according to K predetermined display modes to obtain K mode display acquisition images. The greater the display layer detection severity, the more data is collected for the display layer. The K mode display acquisition images include acquisition images corresponding to the K predetermined display modes.

[0038] The K mode display acquisition images are input into the image intensifier, and each image is enhanced by the image enhancement function. The contrast of the image is improved, the brightness difference in the image is amplified, and the bad pixels (such as dead pixels or bright spots) become more obvious. K enhanced images are obtained from the image intensifier, corresponding to different display modes. These enhanced images are input into K mode display bad pixel recognition models. Each model specifically corresponds to a display mode. The model will determine whether there are bad pixels, as well as the location and type of the bad pixels based on the features in the image. By using K different mode display bad pixel recognition models respectively, detection can be performed in different display modes to obtain K mode display bad pixel recognition results, which include information such as the location, type and degree of the bad pixels.

[0039] The K modes display bad pixel recognition results are fused to obtain a unified bad pixel detection result, including bad pixel location, bad pixel type, bad pixel number, final judgment, etc. The bad pixel location is the bad pixel coordinate after fusion, usually the pixel coordinate position, or the bounding box containing multiple bad pixel areas; bad pixel types include bright spots, dead pixels, color blocks, etc.; the number of bad pixels is the total number of bad pixels detected in the entire display layer; the final judgment refers to whether the screen is considered to have serious defects. If there are bad pixels, the detailed location and type of the bad pixels are output; if there are no bad pixels, the output is normal. By using multiple display modes, different types of bad pixels can be detected, which improves the accuracy of detection.

[0040] S500: Inputting the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into a screen quality adaptation inspection channel to obtain a screen quality adaptation inspection result.

[0041] Specifically, the screen quality adaptation inspection channel is an evaluation channel that integrates multiple bad pixel detection results, which is used to inspect the quality adaptability of mobile phone screens, including a screen quality adaptation evaluator and a screen quality adaptation inspector. The screen quality adaptation evaluator is used to evaluate the quality of each layer of the screen and calculate the adaptation coefficient. The screen quality adaptation inspector determines whether the screen is qualified by setting a quality adaptation evaluation threshold. The protection layer bad pixel detection results, touch layer bad pixel detection results and display layer bad pixel detection results are input into the screen quality adaptation evaluator of the screen quality adaptation inspection channel to evaluate the screen quality, and the adaptation coefficient of each layer, namely, the protection layer quality adaptation coefficient, the touch layer quality adaptation coefficient and the display layer quality adaptation coefficient, is obtained by weighted calculation based on the protection layer quality adaptation coefficient, the touch layer quality adaptation coefficient and the display layer quality adaptation coefficient and the corresponding weights to obtain the screen quality adaptation evaluation coefficient. Different weights can be assigned to each layer of bad pixels according to the degree of their impact on the screen quality.

[0042] Input the screen quality adaptation evaluation coefficient into the screen quality adaptation tester for verification to determine whether the screen quality meets the standard. The final test result is generated by comparing the input screen quality adaptation evaluation coefficient with the set threshold. The set threshold is used to determine whether the screen quality is qualified, usually the lower limit of the screen quality adaptation evaluation coefficient. If the screen quality adaptation evaluation coefficient is greater than or equal to the screen quality adaptation evaluation threshold, the screen quality adaptation test result is qualified, otherwise, it is unqualified. By comprehensively evaluating the bad pixel detection results of each layer, screens with more serious problems can be efficiently screened out, improving the efficiency of screening, and combining the bad pixel detection results of multiple levels to make the screen quality evaluation more comprehensive and accurate, avoiding the single-level bad pixel detection from missing potential problems at other levels.

[0043] S600: When the screen quality adaptation inspection result is qualified, a screen assembly instruction is generated, and the assembly of the target mobile phone screen is performed according to the screen assembly instruction.

[0044] Specifically, when the screen quality adaptation test result is qualified, that is, the screen quality adaptation evaluation coefficient is greater than or equal to the screen quality adaptation evaluation threshold, it means that all levels of the screen meet the quality standards and there are no bad pixels or defects that affect use. According to the screen quality adaptation test results, a screen assembly instruction is generated, which can be a digital signal or a physical instruction, instructing the robot or manual operator on the production line to assemble the screen in a certain order. According to the assembly instructions, the equipment on the production line (such as robots, automated robotic arms, etc.) will begin the actual assembly of the mobile phone screen. At this time, the various components of the screen - protective layer, touch layer, display layer, etc. - will be combined according to the predetermined process to complete the screen assembly process. By automatically generating and executing the screen assembly instructions and passing the screen quality adaptation test, only qualified screens will enter the assembly link, avoiding unqualified screens from being mistakenly assembled into the final product, reducing rework and scrap rates.

[0045] Further, S400 of the present application includes: the third channel of bad pixel detection includes a camera device, an image enhancer and K pattern display bad pixel recognition models corresponding to K predetermined display modes, wherein K is a positive integer greater than 1; based on the display layer detection weight and the camera device, the display layer is imaged according to the K predetermined display modes to obtain K pattern display captured images; the K pattern display captured images are input into the image enhancer to obtain K pattern display enhanced images; the K pattern display enhanced images are input into the K pattern display bad pixel recognition models to output K pattern display bad pixel recognition results; data fusion is performed according to the K pattern display bad pixel recognition results to generate the display layer bad pixel detection result.

[0046] Specifically, the third channel of bad pixel detection is used for bad pixel detection of the display layer, and includes three main components: a camera device, an image enhancer, and K pattern display bad pixel recognition models corresponding to K predetermined display modes. The camera device is a device for image acquisition, usually a high-resolution camera, which is used to capture the image data of the screen, can capture the details on the screen, and can capture the image of the display layer in different display modes. The image enhancer processes the image acquired from the camera device to enhance the clarity and contrast of the image, thereby making the bad pixels more obvious.

[0047] K predetermined display modes refer to a variety of display modes (such as pure color display mode, grayscale mode, grid mode, etc.) used to detect bad pixels in the display layer. Each display mode helps identify different types of display problems (such as color difference, dead pixels, bright spots, etc.) through different color or grayscale distributions. K represents the number of predetermined modes, and K is a positive integer greater than 1. Usually 5 to 10 different modes are selected. Common display modes usually include but are not limited to: pure color display mode (such as black, white, red, green, blue, a single color tone helps detect color difference and color block bad pixels), grayscale display mode (displays a gradient from black to white, helps detect bright spots, dead pixels and color difference, etc.), grid display mode (displays a uniform grid on the screen, which helps detect the brightness and distribution of pixels), color gradient mode (displays a gradient from one color to another, which can reveal abnormal color changes on the screen).

[0048] K-mode display bad pixel recognition models use trained deep learning models (such as convolutional neural networks) for bad pixel detection for different display modes. Each model is trained for a display mode, can identify bad pixels in that mode, and output the recognition results.

[0049] According to the detection severity of the display layer, the target display layer is captured multiple times by the camera device, and each capture corresponds to a different display mode. The number of image captures depends on the detection severity of the display layer. If the severity is high, it means that the detection demand for bad pixels in the display layer is higher, and the number of images captured by the camera device will be more. For example, a high-resolution camera (such as a camera with 1080p or 4K resolution) is used to capture the details of the screen, especially to capture bad pixels in different display modes. According to the detection severity of the display layer, the camera device captures images of the display layer through K predetermined display modes to obtain K mode display capture images. The K mode display capture images are input into the image enhancer, and the image enhancer processes the images according to the predefined enhancement algorithm (image enhancement function) to make the bad pixels on the screen more prominent. Each input image is enhanced by the image enhancement function to obtain K mode display enhanced images. After being processed by the image enhancement function, the details in the original image are enhanced, especially the changes in brightness and contrast are more obvious. The enhanced image will be used for subsequent bad pixel detection.

[0050] The K-mode display enhanced images are input into the corresponding K-mode display bad pixel recognition models for bad pixel recognition. The models recognize the presence of bad pixels based on the image features in each display mode (such as color block changes, grayscale changes, brightness changes, etc.). The K-mode display bad pixel recognition model refers to the bad pixel detection model used in each mode, which is usually a convolutional neural network or other deep learning algorithm, specially trained for a specific display mode. For example, in black mode, the convolutional neural network detects that there are multiple black bright spots in a certain area on the screen and marks them as bad pixels; in grid mode, the model identifies the failure area at the intersection of the grid lines and marks them as bad pixels.

[0051] K modes display bad pixel recognition models collect images corresponding to each display mode, and each model learns how to distinguish normal pixels from bad pixels based on the input images. The specific training process is similar to the above-mentioned protective layer bad pixel detection model, but there are slight differences. Each model will analyze the image independently and output the recognition results, including the bad pixel location, bad pixel type, bad pixel degree, and existing mode. K modes display bad pixel recognition results are: the output results obtained after model processing, usually the judgment of whether there is a bad pixel or the location and type of the bad pixel. Each model will output the bad pixel detection results in its corresponding mode based on the input enhanced image. By using K different modes to display the bad pixel recognition model, detection can be performed in different display modes, which can cover more types of bad pixel types. In different modes, the display method and expression of bad pixels are different. Using multiple specialized models for analysis can improve the comprehensiveness and accuracy of bad pixel detection.

[0052] The bad pixel recognition results from different modes are integrated to generate a final display layer bad pixel detection result. The goal of data fusion is to improve the accuracy of detection by integrating the detection results of multiple modes, avoid missing possible bad pixels, and solve the false positive or false negative problems that may be generated by identification in different modes. The bad pixel recognition results output by the K-mode display bad pixel recognition model are used as input. Each of these results contains the bad pixel detection situation in the mode, such as the location of the bad pixel, the type of bad pixel (such as bright spot, dead spot, etc.), and the situation where the bad pixel may not be detected in some modes. According to the specific application requirements, select a suitable fusion method, such as weighted voting method (combined by voting, the position with the most votes is considered to be the final bad pixel), merge the bad pixel recognition results of each mode, and generate a unified bad pixel detection result.

[0053] If a bad pixel is detected at the same position in different modes, then this position will be considered to have a bad pixel. For example, if both the black mode and the red mode have a bad pixel at the same position, the final detection result will consider that there is a bad pixel at this position. If multiple modes detect a bad pixel, the final result will consider that there is a bad pixel, even if some modes fail to detect it alone. After completing the data fusion, the final display layer bad pixel detection result is generated, including the bad pixel location, bad pixel type, number of bad pixels, final judgment, etc. For example, in the black mode, bright spots are easier to identify; in the red mode, areas with uneven colors or pixel damage are easier to detect. By fusing the detection results of multiple modes, some bad pixels can be avoided from being ignored. Through data fusion, the location of the bad pixel can be accurately located to avoid false alarms or missed alarms due to the low sensitivity of some modes to bad pixels.

[0054] Furthermore, the present application also includes the following steps: the image enhancer includes an image enhancement function, and the image enhancement function is: Among them, I enh The mode corresponding to characterization I displays the pixel value of the enhanced image, the I characterization mode displays the pixel value of the acquired image, the ALV characterization mode displays the average brightness value of the acquired image, the SDB characterization mode displays the standard deviation of the pixel brightness of the acquired image, and the CEF characterizes the contrast enhancement factor.

[0055] Specifically, an image enhancer is a device or algorithm used to improve image quality by adjusting image brightness, contrast, clarity and other parameters to make key details (such as bad pixels) in the image more obvious. The image enhancement function is the core algorithm of the image enhancer, which is used to process the original image and generate an enhanced image. The specific form of the image enhancement function is: Among them, I represents the pixel value of the acquired image in the display mode; ALV represents the average brightness value of the acquired image in the display mode, which is the average brightness of all pixels in the image and is used to reflect the overall brightness level of the image; SDB represents the standard deviation of the pixel brightness of the acquired image in the display mode. The standard deviation measures the dispersion of the brightness in the image and can reflect the contrast and detail richness of the image; CEF represents the contrast enhancement factor, which is a coefficient used to enhance the contrast of the image, thereby making the details in the image more prominent; I enh Represents the pixel value of the enhanced image. After being processed by the enhancement function, each pixel value of the image will be adjusted to make the details of the image clearer.

[0056] Get K mode display acquisition images, which contain detailed information of the screen in different display modes. Calculate the average brightness value (ALV) of the original image, which is the average brightness value of all pixels in the image, usually calculated by the ratio of the sum of all pixel brightness values ​​to the total number of pixels. Calculate the standard deviation (SDB) of the pixel brightness of the original image. The standard deviation measures the distribution of the brightness of the image pixels. It is usually calculated by calculating the square of the difference between the brightness value of each pixel and the average brightness value, adding all the square values ​​and dividing by the total number of pixels to get a value, and then taking the square root of this value. The contrast enhancement coefficient is a constant value set according to the degree of enhancement required. It is usually adjusted through experiments to enhance the image contrast and make the details more prominent. For example, suppose the pixel values ​​of an image are distributed between 0 and 255, representing the grayscale level from black to white, the average brightness value of the image is 100, the standard deviation of the pixel brightness is 20, and the contrast enhancement factor is set to 1.5. If the original brightness value of a pixel is 120, then the new brightness value of the pixel after enhancement is 1.5, indicating that the brightness value of the pixel becomes 1.5 after enhancement. After proper scaling and adjustment, the details in the image are more prominent.

[0057] By using the contrast enhancement coefficient, the contrast of the image is significantly improved, making the details (such as bad pixels) more prominent; through the processing of the image enhancement function, the brightness difference in the image will be amplified, helping the detection model to more easily identify bad pixels, especially for those bad pixels with low brightness or weak contrast, image enhancement can make these defects more obvious; by improving the clarity and contrast of the image, the model can better identify bad pixels and reduce misjudgments and missed judgments.

[0058] Further, S500 of the present application includes: the screen quality adaptability inspection channel includes a screen quality adaptability evaluator and a screen quality adaptability inspector; the protective layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result are input into the screen quality adaptability evaluator to obtain a screen quality adaptability evaluation coefficient; the screen quality adaptability evaluation coefficient is input into the screen quality adaptability inspector to output the screen quality adaptability inspection result, wherein the screen quality adaptability inspector includes a screen quality inspection operator, and the screen quality inspection operator includes: if the screen quality adaptability evaluation coefficient is greater than or equal to a screen quality adaptability evaluation threshold, the screen quality adaptability inspection result is qualified, and if the screen quality adaptability evaluation coefficient is less than the screen quality adaptability evaluation threshold, the screen quality adaptability inspection result is unqualified.

[0059] Specifically, the screen quality adaptation test channel is a channel used to comprehensively evaluate the quality of mobile phone screens and perform adaptability tests based on the test results. It includes a screen quality adaptation evaluator and a screen quality adaptation tester, which can evaluate and verify the quality of the screen to ensure that the screen meets the quality standards. The screen quality adaptation evaluator is a tool for evaluating the overall quality of mobile phone screens. According to the bad pixel detection results of different layers (protective layer, touch layer, display layer), a comprehensive screen quality adaptation evaluation coefficient is calculated. It includes two sub-models, namely the screen quality multidimensional evaluation model and the screen quality adaptation calculation model. The screen quality multidimensional evaluation model is responsible for quality evaluation of the three key layers of the mobile phone screen, including the protective layer quality evaluation model, the touch layer quality evaluation model and the display layer quality evaluation model. It evaluates the quality of the protective layer, the touch layer and the display layer respectively, and outputs the corresponding quality evaluation coefficients. The screen quality adaptation calculation model calculates the screen quality adaptation evaluation coefficient based on the quality evaluation coefficient output by the screen quality multidimensional evaluation model, including three weights, namely the protective layer quality adaptation weight, the touch layer quality adaptation weight and the display layer quality adaptation weight.

[0060] The protection layer bad pixel detection results, the touch layer bad pixel detection results and the display layer bad pixel detection results are input into the model corresponding to each layer in the screen quality multidimensional evaluation model in the screen quality adaptability evaluator for quality evaluation, and the protection layer quality adaptation coefficient, the touch layer quality adaptation coefficient and the display layer quality adaptation coefficient are obtained, and they are input into the screen quality adaptation calculation model for weighted calculation, that is, the quality adaptation coefficient of each layer is multiplied by its corresponding weight, and the three values ​​are added to obtain the screen quality adaptation evaluation coefficient.

[0061] The screen quality adaptation evaluation coefficient is input into the screen quality adaptation verifier to make the final quality qualification judgment. The screen quality adaptation verifier includes a screen quality inspection operator, which is used to judge whether the screen quality meets the standard, and compares the input screen quality adaptation evaluation coefficient with the set threshold to generate the final inspection result. The screen quality adaptation evaluation threshold is a pre-set threshold used to judge whether the screen quality adaptation evaluation coefficient is qualified as the minimum standard.

[0062] If the screen quality adaptation evaluation coefficient is greater than or equal to the threshold, the screen is judged to be qualified; if the screen quality adaptation evaluation coefficient is less than the threshold, the screen is judged to be unqualified. Through the combination of the screen quality adaptation evaluator and the screen quality adaptation tester, an objective and standardized evaluation of screen quality can be achieved, and screens that do not meet the quality standards can be quickly screened out and eliminated, thereby improving the efficiency of the production line and reducing the scrap rate.

[0063] Further, the present application also includes the following steps: the screen quality adaptation evaluator includes a screen quality multidimensional evaluation model and a screen quality adaptation calculation model, wherein the screen quality multidimensional evaluation model includes a protection layer quality evaluation model, a touch layer quality evaluation model, and a display layer quality evaluation model, and the screen quality adaptation calculation model includes a protection layer quality adaptation weight, a touch layer quality adaptation weight, and a display layer quality adaptation weight; the protection layer bad pixel detection result, the touch layer bad pixel detection result, and the display layer bad pixel detection result are input into the screen quality multidimensional evaluation model to obtain a protection layer quality evaluation coefficient, a touch layer quality evaluation coefficient, and a display layer quality evaluation coefficient; according to the protection layer quality evaluation coefficient, the touch layer quality evaluation coefficient, and the display layer quality evaluation coefficient, a subtraction operation is performed on the protection layer quality expectation coefficient, the touch layer quality expectation coefficient, and the display layer quality expectation coefficient to obtain a protection layer quality adaptation coefficient, a touch layer quality adaptation coefficient, and a display layer quality adaptation coefficient; the protection layer quality adaptation coefficient, the touch layer quality adaptation coefficient, and the display layer quality adaptation coefficient are input into the screen quality adaptation calculation model to obtain the screen quality adaptation evaluation coefficient.

[0064] Specifically, the screen quality adaptation evaluator is used to evaluate the quality of each layer of the screen and calculate the adaptation coefficient, including a screen quality multidimensional evaluation model and a screen quality adaptation calculation model. The screen quality multidimensional evaluation model is a quality evaluation model that comprehensively considers all components of the mobile phone screen (such as the protective layer, touch layer, and display layer), independently evaluates the quality of each layer, and obtains the quality evaluation coefficient of each layer through evaluations in multiple dimensions. The screen quality multidimensional evaluation model includes a protective layer quality evaluation model, a touch layer quality evaluation model, and a display layer quality evaluation model, which are used to evaluate the quality of the mobile phone screen protective layer, touch layer, and display layer respectively. The screen quality adaptation calculation model is a model used to calculate the overall screen quality adaptability based on the quality evaluation results of each layer, including the importance weight of each layer (protective layer, touch layer, display layer) in the screen quality adaptation calculation. Different layers may have different effects on the overall screen quality, so when calculating the adaptation coefficient, different weights need to be given according to the quality evaluation results of each layer.

[0065] The bad pixel detection results of the protective layer, the bad pixel detection results of the touch layer, and the bad pixel detection results of the display layer are input into the multi-dimensional screen quality evaluation model. In other words, the bad pixel detection results of the protective layer are input into the protective layer quality evaluation model, the bad pixel detection results of the touch layer are input into the touch layer quality evaluation model, and the bad pixel detection results of the display layer are input into the display layer quality evaluation model. The multi-dimensional screen quality evaluation model will evaluate the quality of each layer based on the bad pixel detection results of each layer.

[0066] Taking the protective layer as an example, the protective layer quality assessment model calculates the quality assessment coefficient of the protective layer based on the input bad pixel detection results. If the number of bad pixels detected is large, the quality assessment coefficient will be low. The protective layer quality assessment model will preset a threshold. The number of bad pixels exceeding the threshold will significantly reduce the quality coefficient; bad pixels located in the center of the screen have a greater impact than bad pixels at the edge; different types of bad pixels (such as bright spots and dead pixels) have different effects on quality assessment, and the corresponding weights are set according to the bad pixel type. The protective layer quality assessment model performs weighted calculations based on the above three factors and finally obtains the protective layer quality assessment coefficient. After the calculation is completed, the protective layer quality assessment model outputs a quality assessment coefficient, which represents the overall quality level of the protective layer, usually between 0 (worst quality) and 100 (best quality). The protective layer quality expectation coefficient is preset as the quality value of the screen layer under ideal conditions. The protective layer quality adaptation coefficient is obtained by subtracting the protective layer quality expectation coefficient from the protective layer quality assessment coefficient.

[0067] Similarly, the above steps are also performed for the touch layer and the display layer to obtain the corresponding touch layer quality evaluation coefficient and display layer quality evaluation coefficient. The touch layer quality expectation coefficient and the display layer quality expectation coefficient under the ideal state are pre-set, and the touch layer quality adaptation coefficient and the display layer quality adaptation coefficient are obtained by subtraction operation. The quality adaptation coefficient of each layer is input into the screen quality adaptation calculation model according to its corresponding weight, and weighted calculation is performed to obtain the final screen quality adaptation evaluation coefficient. In other words, the screen quality adaptation evaluation coefficient = protective layer quality adaptation coefficient × protective layer quality adaptation weight + touch layer quality adaptation coefficient × touch layer quality adaptation weight + display layer quality adaptation coefficient × display layer quality adaptation weight. By combining the bad pixel detection results of each layer and performing multi-dimensional quality evaluation, the overall quality of the mobile phone screen can be fully reflected, not only considering the bad pixels of a single layer, but also comprehensively considering the impact of different layers on the screen quality. By introducing weighted calculation, a reasonable evaluation can be performed according to the quality impact weight of each layer. Weighted calculation ensures that more important layers (such as the display layer) can occupy a larger proportion in the final result, thereby improving the accuracy of the inspection results.

[0068] In summary, the method for detecting bad pixels on a mobile phone screen provided by the present application has the following technical effects: By evaluating the importance of bad pixel detection according to the target mobile phone screen, the protection layer detection severity, touch layer detection severity and display layer detection severity are obtained, and the target mobile phone screen includes a protection layer, a touch layer and a display layer; based on the protection layer detection severity, the first bad pixel detection channel is driven to perform bad pixel detection on the protection layer to obtain the protection layer bad pixel detection result; based on the touch layer detection severity, the touch layer is performed with a bad pixel detection according to a second bad pixel detection channel to obtain the touch layer bad pixel detection result; based on the display layer detection severity, the display layer is performed with a bad pixel detection according to a third bad pixel detection channel to obtain the display layer bad pixel detection result; the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result are input into the screen quality adaptation inspection channel to obtain the screen quality adaptation inspection result; when the screen quality adaptation inspection result is qualified, a screen assembly instruction is generated, and the assembly of the target mobile phone screen is performed according to the screen assembly instruction. That is to say, by evaluating the importance of testing the three levels of the target mobile phone screen, the detection severity of each level is determined, and according to the detection severity of each level, a dedicated detection channel is used to detect the bad pixels of the corresponding level. The bad pixel detection results of the three levels are input into a comprehensive inspection channel to obtain the inspection results. When the inspection results are qualified, screen assembly instructions are generated to assemble the target mobile phone screen, thereby improving the accuracy of bad pixel detection and thus improving the overall quality of the mobile phone screen.

[0069] Embodiment 2, based on the same inventive concept as the method for detecting bad pixels on a mobile phone screen in the aforementioned embodiment 1, the present application also provides a system for detecting bad pixels on a mobile phone screen, see the attached Figure 2 , the mobile phone screen bad pixel detection system comprises: An importance evaluation module 11, the importance evaluation module 11 is used to perform bad pixel detection importance evaluation according to the target mobile phone screen, and obtain the protection layer detection severity, the touch layer detection severity and the display layer detection severity, and the target mobile phone screen includes a protection layer, a touch layer and a display layer; a first detection module 12, the first detection module 12 is used to drive the first bad pixel detection channel to perform bad pixel detection on the protection layer based on the protection layer detection severity, and obtain the protection layer bad pixel detection result; a second detection module 13, the second detection module 13 is used to perform bad pixel detection on the touch layer according to the second bad pixel detection channel based on the touch layer detection severity, and obtain the touch layer bad pixel detection result; A third detection module 14, the third detection module 14 is used to perform bad pixel detection on the display layer based on the detection severity of the display layer according to the third bad pixel detection channel to obtain a display layer bad pixel detection result; an adaptation verification module 15, the adaptation verification module 15 is used to input the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into the screen quality adaptation inspection channel to obtain a screen quality adaptation inspection result; an assembly instruction generation module 16, the assembly instruction generation module 16 is used to generate a screen assembly instruction when the screen quality adaptation inspection result is qualified, and execute the assembly of the target mobile phone screen according to the screen assembly instruction.

[0070] Furthermore, the first detection module 12 in the mobile phone screen bad pixel detection system is also used for: the first bad pixel detection channel includes a visual camera, an image filter and a protective layer bad pixel detection model; based on the protective layer detection severity, the visual camera is driven to capture an image of the protective layer to obtain a protective layer captured image; the protective layer captured image is input into the image filter to obtain a protective layer enhanced image; the protective layer enhanced image is input into the protective layer bad pixel detection model to generate the protective layer bad pixel detection result.

[0071] Furthermore, the second detection module 13 in the mobile phone screen bad pixel detection system is also used for: the second bad pixel detection channel includes a touch robot and a touch layer bad pixel recognition model; based on the touch layer detection severity, the touch robot performs multiple touch operations on the touch layer to obtain multiple groups of touch response data; the multiple groups of touch response data are input into the touch layer bad pixel recognition model to obtain multiple groups of touch layer bad pixel recognition results; and the multiple groups of touch layer bad pixel recognition results are integrated to generate the touch layer bad pixel detection result.

[0072] Furthermore, the third detection module 14 in the mobile phone screen bad pixel detection system is also used for: the bad pixel detection third channel includes a camera device, an image enhancer and K pattern display bad pixel recognition models corresponding to K predetermined display modes, wherein K is a positive integer greater than 1; based on the display layer detection weight and the camera device, the display layer is imaged according to the K predetermined display modes to obtain K pattern display captured images; the K pattern display captured images are input into the image enhancer to obtain K pattern display enhanced images; the K pattern display enhanced images are input into the K pattern display bad pixel recognition models to output K pattern display bad pixel recognition results; data fusion is performed according to the K pattern display bad pixel recognition results to generate the display layer bad pixel detection result.

[0073] Furthermore, the third detection module 14 in the mobile phone screen bad pixel detection system is also used for: the image enhancer includes an image enhancement function, and the image enhancement function is: Among them, I enh The mode corresponding to characterization I displays the pixel value of the enhanced image, the I characterization mode displays the pixel value of the acquired image, the ALV characterization mode displays the average brightness value of the acquired image, the SDB characterization mode displays the standard deviation of the pixel brightness of the acquired image, and the CEF characterizes the contrast enhancement factor.

[0074] Furthermore, the adaptation verification module 15 in the mobile phone screen bad pixel detection system is also used for: the screen quality adaptation verification channel includes a screen quality adaptation evaluator and a screen quality adaptation verifier; the protective layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result are input into the screen quality adaptation evaluator to obtain a screen quality adaptation evaluation coefficient; the screen quality adaptation evaluation coefficient is input into the screen quality adaptation verifier to output the screen quality adaptation verification result, wherein the screen quality adaptation verifier includes a screen quality verification operator, and the screen quality verification operator includes: if the screen quality adaptation evaluation coefficient is greater than or equal to a screen quality adaptation evaluation threshold, the screen quality adaptation verification result is qualified, and if the screen quality adaptation evaluation coefficient is less than the screen quality adaptation evaluation threshold, the screen quality adaptation verification result is unqualified.

[0075] Furthermore, the adaptation verification module 15 in the mobile phone screen bad pixel detection system is also used for: the screen quality adaptation evaluator includes a screen quality multidimensional evaluation model and a screen quality adaptation calculation model, wherein the screen quality multidimensional evaluation model includes a protection layer quality evaluation model, a touch layer quality evaluation model, and a display layer quality evaluation model, and the screen quality adaptation calculation model includes a protection layer quality adaptation weight, a touch layer quality adaptation weight, and a display layer quality adaptation weight; the protection layer bad pixel detection result, the touch layer bad pixel detection result, and the display layer bad pixel detection result are input into the screen quality multidimensional evaluation model. A model is provided to obtain a protection layer quality assessment coefficient, a touch layer quality assessment coefficient and a display layer quality assessment coefficient; according to the protection layer quality assessment coefficient, the touch layer quality assessment coefficient and the display layer quality assessment coefficient, a subtraction operation is performed on the protection layer quality expectation coefficient, the touch layer quality expectation coefficient and the display layer quality expectation coefficient to obtain a protection layer quality adaptation coefficient, a touch layer quality adaptation coefficient and a display layer quality adaptation coefficient; the protection layer quality adaptation coefficient, the touch layer quality adaptation coefficient and the display layer quality adaptation coefficient are input into the screen quality adaptation calculation model to obtain the screen quality adaptation evaluation coefficient.

[0076] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The method and specific examples for detecting bad pixels on a mobile phone screen in Embodiment 1 are also applicable to a system for detecting bad pixels on a mobile phone screen in this embodiment. Through the detailed description of the method for detecting bad pixels on a mobile phone screen, those skilled in the art can clearly understand the system for detecting bad pixels on a mobile phone screen in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0077] Embodiment three, based on the inventive concept of a method for detecting bad pixels on a mobile phone screen in the aforementioned embodiment one, the present application also provides an electronic device, comprising: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the steps of a method for detecting bad pixels on a mobile phone screen described in any one of the aforementioned embodiments one.

[0078] Attached Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application. Figure 3In the embodiment, the bus architecture is represented by bus 300, which may include any number of interconnected buses and bridges, and bus 300 connects various circuits including one or more processors represented by processor 302 and memory represented by memory 304. Bus 300 may also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, namely a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 may be used to store data used by processor 302 when performing operations.

[0079] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0080] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for detecting bad pixels on a mobile phone screen, characterized in that: include: Perform bad pixel detection importance evaluation on a target mobile phone screen to obtain a protection layer detection severity, a touch layer detection severity, and a display layer detection severity, wherein the target mobile phone screen includes a protection layer, a touch layer, and a display layer; Based on the detection severity of the protection layer, driving a first bad pixel detection channel to perform bad pixel detection on the protection layer to obtain a bad pixel detection result of the protection layer; Based on the detection severity of the touch layer, performing bad pixel detection on the touch layer according to a second bad pixel detection channel to obtain a bad pixel detection result of the touch layer; Based on the detection severity of the display layer, performing bad pixel detection on the display layer according to a third bad pixel detection channel to obtain a bad pixel detection result of the display layer; Inputting the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into a screen quality adaptation inspection channel to obtain a screen quality adaptation inspection result; When the screen quality adaptation inspection result is qualified, a screen assembly instruction is generated, and the assembly of the target mobile phone screen is performed according to the screen assembly instruction.

2. A method for detecting bad pixels on a mobile phone screen as claimed in claim 1, characterized in that: Based on the detection severity of the protection layer, driving the first bad pixel detection channel to perform bad pixel detection on the protection layer to obtain a bad pixel detection result of the protection layer, including: The first bad pixel detection channel includes a visual camera, an image filter and a protective layer bad pixel detection model; Based on the detection weight of the protective layer, driving the visual camera to collect an image of the protective layer to obtain a protective layer collection image; Inputting the protection layer acquisition image into the image filter to obtain a protection layer enhanced image; The protection layer enhanced image is input into the protection layer bad pixel detection model to generate the protection layer bad pixel detection result.

3. A method for detecting bad pixels on a mobile phone screen as claimed in claim 1, characterized in that: Based on the touch layer detection severity, performing bad pixel detection on the touch layer according to a bad pixel detection second channel to obtain a touch layer bad pixel detection result, including: The second bad pixel detection channel includes a touch robot and a touch layer bad pixel recognition model; Based on the detection weight of the touch layer, the touch robot performs multiple touch operations on the touch layer to obtain multiple sets of touch response data; Inputting the multiple sets of touch response data into the touch layer bad pixel recognition model to obtain multiple sets of touch layer bad pixel recognition results; The multiple groups of touch layer bad pixel recognition results are integrated to generate the touch layer bad pixel detection result.

4. A method for detecting bad pixels on a mobile phone screen as claimed in claim 1, characterized in that: Based on the detection severity of the display layer, performing bad pixel detection on the display layer according to the bad pixel detection third channel to obtain a bad pixel detection result of the display layer, including: The third bad pixel detection channel includes a camera device, an image intensifier, and K pattern display bad pixel recognition models corresponding to K predetermined display modes, wherein K is a positive integer greater than 1; Based on the detection weight of the display layer and the camera device, the display layer is imaged according to the K predetermined display modes to obtain K mode display acquisition images; Inputting the K pattern display acquisition images into the image intensifier to obtain K pattern display enhanced images; Inputting the K pattern display enhanced images into the K pattern display bad pixel recognition models, and outputting K pattern display bad pixel recognition results; Data fusion is performed based on the K pattern display bad pixel identification results to generate the display layer bad pixel detection result.

5. A method for detecting bad pixels on a mobile phone screen as claimed in claim 4, characterized in that: The image enhancer includes an image enhancement function, which is: ； Among them, I enh The mode corresponding to characterization I displays the pixel value of the enhanced image, the I characterization mode displays the pixel value of the acquired image, the ALV characterization mode displays the average brightness value of the acquired image, the SDB characterization mode displays the standard deviation of the pixel brightness of the acquired image, and the CEF characterizes the contrast enhancement factor.

6. A method for detecting bad pixels on a mobile phone screen as claimed in claim 1, characterized in that: Inputting the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into the screen quality adaptation inspection channel to obtain the screen quality adaptation inspection result, including: The screen quality adaptation inspection channel includes a screen quality adaptation evaluator and a screen quality adaptation inspector; Inputting the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into the screen quality adaptation evaluator to obtain a screen quality adaptation evaluation coefficient; The screen quality adaptation evaluation coefficient is input into the screen quality adaptation verifier, and the screen quality adaptation verification result is output, wherein the screen quality adaptation verifier includes a screen quality verification operator, and the screen quality verification operator includes: if the screen quality adaptation evaluation coefficient is greater than or equal to a screen quality adaptation evaluation threshold, the screen quality adaptation verification result is qualified; if the screen quality adaptation evaluation coefficient is less than the screen quality adaptation evaluation threshold, the screen quality adaptation verification result is unqualified.

7. A method for detecting bad pixels on a mobile phone screen as claimed in claim 6, characterized in that: Inputting the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into the screen quality adaptation evaluator to obtain a screen quality adaptation evaluation coefficient, including: The screen quality adaptation evaluator includes a screen quality multidimensional evaluation model and a screen quality adaptation calculation model, wherein the screen quality multidimensional evaluation model includes a protection layer quality evaluation model, a touch layer quality evaluation model, and a display layer quality evaluation model, and the screen quality adaptation calculation model includes a protection layer quality adaptation weight, a touch layer quality adaptation weight, and a display layer quality adaptation weight; Inputting the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into the screen quality multi-dimensional evaluation model to obtain a protection layer quality evaluation coefficient, a touch layer quality evaluation coefficient and a display layer quality evaluation coefficient; According to the protection layer quality assessment coefficient, the touch layer quality assessment coefficient and the display layer quality assessment coefficient, a subtraction operation is performed on the protection layer quality expectation coefficient, the touch layer quality expectation coefficient and the display layer quality expectation coefficient to obtain a protection layer quality adaptation coefficient, a touch layer quality adaptation coefficient and a display layer quality adaptation coefficient; The protection layer quality adaptation coefficient, the touch layer quality adaptation coefficient and the display layer quality adaptation coefficient are input into the screen quality adaptation calculation model to obtain the screen quality adaptation evaluation coefficient.

8. A mobile phone screen bad pixel detection system, characterized in that: The method for detecting bad pixels on a mobile phone screen according to any one of claims 1 to 7 is used to implement the method, wherein the mobile phone screen bad pixel detection system comprises: An importance evaluation module, the importance evaluation module is used to perform bad pixel detection importance evaluation according to a target mobile phone screen, and obtain a protection layer detection severity, a touch layer detection severity, and a display layer detection severity, wherein the target mobile phone screen includes a protection layer, a touch layer, and a display layer; A first detection module, the first detection module is used to drive a first bad pixel detection channel to perform bad pixel detection on the protection layer based on the detection severity of the protection layer, and obtain a bad pixel detection result of the protection layer; A second detection module, the second detection module is used to perform bad pixel detection on the touch layer according to a second bad pixel detection channel based on the touch layer detection severity, and obtain a bad pixel detection result of the touch layer; A third detection module, the third detection module is used to perform bad pixel detection on the display layer according to a third bad pixel detection channel based on the detection severity of the display layer, and obtain a bad pixel detection result of the display layer; An adaptation verification module, the adaptation verification module is used to input the protection layer bad pixel detection result, the touch layer bad pixel detection result and the display layer bad pixel detection result into a screen quality adaptation inspection channel to obtain a screen quality adaptation inspection result; An assembly instruction generation module is used to generate a screen assembly instruction when the screen quality adaptation inspection result is qualified, and to assemble the target mobile phone screen according to the screen assembly instruction.

9. An electronic device, comprising: at least one processor; a memory communicatively coupled to the at least one processor; Wherein, the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of a method for detecting bad pixels on a mobile phone screen as described in any one of claims 1 to 7.

Images