Visual RGB value identification and detection system for industrial visual detection

Abstract

The invention discloses a visual RGB value recognition and detection system for industrial visual detection, and relates to the technical field of RGB value recognition and detection.Working parameters of an industrial digital camera are corrected according to environment parameters, and the RGB value of a vehicle-mounted display screen in each preset direction and environment light parameters in each preset direction are corrected according to the RGB value of the vehicle-mounted display screen in each preset direction. Analyzing the association condition between the RGB values of the preset directions and the ambient light parameters and the directions, and when the association relationship of the RGB values of the target vehicle-mounted display screen is normal, performing region division on the target vehicle-mounted display screen, and then judging whether the RGB difference value of each region is abnormal or not; and when the RGB value association relationship of the target vehicle-mounted display screen is abnormal, carrying out secondary correction on the target vehicle-mounted display screen. According to the method, the influence of the environment temperature and the environment humidity on the RGB value detection result is reduced to the maximum extent, the RGB value is detected from the perspective of user vision and the environment light parameters, and the problem that technical parameters are disjointed with user experience is solved.

Description

Technical Field

[0001] This application relates to the field of RGB value recognition and detection technology, specifically to a visual RGB value recognition and detection system for industrial visual inspection. Background Technology

[0002] With the booming development of new energy vehicles in China, the vehicles and core components have gradually shifted from electrification to intelligentization. As intelligent control technology becomes more widespread, the vehicle electrical components and in-vehicle displays are currently evolving from mainstream three-dimensional color and grayscale control functions to deep integration with vehicle-road cooperative systems, achieving ambient light adaptive color matching and building a full-scene intelligent visual ecosystem. Therefore, this application proposes a visual RGB value recognition and detection system for industrial visual inspection.

[0003] Existing technology, such as the invention application patent with announcement number CN107607207A, discloses a method, system, and electronic device for diagnosing thermal faults in power equipment. The method includes: acquiring infrared images of power equipment; constructing a convolutional neural network model based on the infrared images; inputting the infrared image to be detected into the convolutional neural network model; identifying temperature scales and power equipment in the infrared image through the convolutional neural network model; generating an RGB value and temperature reference table based on the RGB values ​​of the identified temperature scale pixels and the upper and lower boundaries of the temperature scale; extracting the RGB values ​​of the identified power equipment; comparing the extracted RGB values ​​with the RGB values ​​in the temperature reference table to obtain the temperature result of the identified power equipment; and diagnosing the temperature result according to power grid system diagnostic standards to determine whether the power equipment has a thermal fault. This invention efficiently and accurately identifies power equipment through a convolutional neural network model and precisely reads the temperature through RGB values, thereby improving the intelligence level of the power grid system.

[0004] Existing technologies, such as the invention application patent with publication number CN114494685B, disclose an automatic identification method and apparatus for continuous frame RGB images. This method includes: generating a series of continuous frame RGB images based on a set rendering trajectory, and outputting the camera attributes of each frame RGB image; after detecting and removing invalid, black, and blurred images from the RGB images, determining whether the quality of two adjacent RGB images is acceptable based on the angle amplitude and similarity calculated according to the camera attributes and pixel values; and performing post-processing on RGB images that do not meet quality standards, including manual verification, re-rendering according to the corresponding camera attributes, and re-rendering according to the corresponding rendering attributes, to further detect the RGB image quality. This method and apparatus achieve automatic identification of differences in continuous frame RGB images with high efficiency and accuracy.

[0005] The above solution has the following technical problems: 1. The current technology mainly compares the detected RGB values ​​with the standard RGB reference table to identify equipment faults. However, the current technology does not take into account the influence of the detection device on the RGB values. Since the photosensitive element of the detection device is affected by the ambient temperature, humidity and ambient light, the RGB values ​​identified by the same device in different environments are quite different. The current technology's neglect of this aspect leads to a lack of accuracy in identifying equipment faults.

[0006] 2. Current technology applied to in-vehicle displays does not take into account the human eye's ability to recognize RGB values ​​on in-vehicle displays. When performing RGB value recognition and detection on in-vehicle displays, it is necessary to consider the display deviation caused by changes in viewing angle and environment, as well as the human eye's ability to recognize RGB values. The current technology's neglect of this aspect will result in the current RGB value recognition and detection not having practical utility. Summary of the Invention

[0007] The purpose of this application is to provide a visual RGB value recognition and detection system for industrial visual inspection, which solves the problems existing in the background technology.

[0008] To solve the above-mentioned technical problems, this application adopts the following technical solution: This application provides a visual RGB value recognition and detection system for industrial visual inspection, including: a detection equipment calibration module: used to collect environmental parameters during visual inspection through an environmental sensor, and then calibrate the working parameters of the industrial digital camera according to the environmental parameters.

[0009] Ambient light analysis module: After the working parameters of the industrial digital camera are calibrated, the module captures images of the target vehicle display screen in each preset direction using the industrial digital camera, thereby obtaining the RGB values ​​of the vehicle display screen in each preset direction. At the same time, it acquires the ambient light parameters of the industrial digital camera in each preset direction, and analyzes the correlation between the RGB values ​​and the ambient light parameters in each preset direction.

[0010] RGB value analysis module: When the RGB value correlation of the target vehicle display is normal, it divides the target vehicle display into regions and then determines whether the RGB difference of each region is abnormal; when the RGB value correlation of the target vehicle display is abnormal, it performs secondary calibration on the target vehicle display.

[0011] Data storage module: When an abnormal RGB value is detected in the target vehicle display, the abnormal data is stored in the corresponding path and an early warning is issued.

[0012] The beneficial effects of this application are as follows: 1. This application provides a visual RGB value recognition and detection system for industrial visual inspection. It corrects the operating parameters of an industrial digital camera based on environmental parameters, and analyzes the correlation between the RGB values ​​of the vehicle-mounted display screen in each preset direction and the ambient light parameters in each preset direction, as well as the correlation between the directions. When the correlation of the RGB values ​​of the target vehicle-mounted display screen is normal, the target vehicle-mounted display screen is divided into regions, and then it is determined whether the RGB difference in each region is abnormal. When the correlation of the RGB values ​​of the target vehicle-mounted display screen is abnormal, a secondary correction is performed on the target vehicle-mounted display screen. This application not only minimizes the impact of ambient temperature and humidity on the RGB value detection results, but also detects RGB values ​​from the user's visual perspective and ambient light parameters, solving the problem of the disconnect between technical parameters and user experience.

[0013] 2. This application eliminates the influence of ambient temperature and humidity on industrial digital cameras by calibrating them, thereby ensuring the accuracy and stability of industrial vision inspection, eliminating the influence of temperature and humidity on optical imaging quality, and minimizing the influence of ambient temperature and humidity on RGB value detection results, thus making it more suitable for complex industrial environments.

[0014] 3. This application analyzes the influence of preset directions and ambient light parameters on RGB to determine whether the RGB values ​​of the vehicle display screen are abnormal. Due to the influence of angle factors and ambient light, the RGB values ​​of the target vehicle display screen will be different when viewed from different directions. Therefore, starting from the user's visual perspective and ambient light parameters, the RGB values ​​of each preset direction are analyzed to avoid the problem of technical parameters being out of touch with user experience. At the same time, the RGB difference of each area of ​​the target vehicle display screen is analyzed to ensure the difference in RGB values ​​between the core area and non-core areas. Based on this, visual layering maximizes the efficiency of information transmission, shortens the recognition time of core information, avoids driver distraction, and ensures the optimization of driving safety. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the system structure connection of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Reference Figure 1 As shown, this application provides a visual RGB value recognition and detection system for industrial visual inspection, including the following modules: Detection equipment calibration module: used to collect environmental parameters during visual inspection through environmental sensors, and then calibrate the working parameters of the industrial digital camera based on the environmental parameters.

[0019] It should be noted that the industrial digital camera mentioned is the Daheng MER-G series industrial digital camera. The camera's dimensions are only 29mm × 29mm × 29mm, making it suitable for various testing scenarios with demanding space requirements.

[0020] In one specific example, the environmental sensor includes a temperature sensor and a humidity sensor; the environmental parameters include ambient temperature and ambient humidity; and the operating parameters of the industrial digital camera include exposure time, gain, and focal length.

[0021] In a specific example, the process of collecting environmental parameters during visual inspection using environmental sensors and then correcting the operating parameters of the industrial digital camera based on these parameters is as follows: A temperature sensor and a humidity sensor are installed at the photosensitive chip inside the industrial digital camera. When the industrial digital camera acquires an RGB image of the target vehicle display screen, the temperature and humidity of the photosensitive chip at the time of acquisition are simultaneously obtained through the temperature and humidity sensors. The temperature and humidity of the photosensitive chip at the time of acquisition are compared with a set temperature threshold range and a set humidity threshold range. If the temperature or humidity of the photosensitive chip at the time of acquisition does not fall within the set temperature threshold range and a set humidity threshold range, the temperature and humidity of the photosensitive chip are substituted into a pre-trained temperature and humidity correction model to correct the operating parameters of the industrial digital camera. Conversely, if the temperature and humidity of the photosensitive chip at the time of acquisition both fall within the set temperature threshold range and a set humidity threshold range, the operating parameters of the industrial digital camera are not corrected.

[0022] It should be noted that the temperature threshold range and humidity threshold range were obtained by consulting the user manual of the photosensitive chip.

[0023] In a specific example, the temperature and humidity correction model is trained as follows: S1. Using an industrial digital camera, test several different models of vehicle-mounted displays with known standard RGB values. First, adjust the ambient temperature according to a preset gradient within the set temperature threshold range, and at the same time, adjust the ambient humidity according to a preset gradient outside the set humidity threshold range. Based on this, obtain the first set of test environment combinations, and then use an industrial digital camera to detect the RGB values ​​of each vehicle-mounted display.

[0024] S2. Adjust the ambient temperature according to a preset gradient outside the set temperature threshold range, and adjust the ambient humidity according to a preset gradient within the set humidity threshold range to obtain the second set of test environment combinations. Then, use an industrial digital camera to detect the RGB values ​​of each vehicle display screen.

[0025] S3. Adjust the ambient temperature and humidity according to a preset gradient outside the set temperature threshold range and humidity threshold range to obtain the third set of test environment combinations, and then use an industrial digital camera to detect the RGB values ​​of each vehicle display screen.

[0026] S4. The first test environment combination, the second test environment combination, and the third test environment combination, as well as the RGB values ​​and standard RGB values ​​of each vehicle display screen collected accordingly, are divided into training set and validation set according to a preset ratio. Based on the BP neural network model, the BP neural network model is trained and validated using the training set and validation set to obtain the temperature and humidity correction model of the industrial digital camera.

[0027] It should be noted that when industrial digital cameras capture images of various vehicle displays, all ambient light parameters, acquisition angle, and acquisition height are the same, except for ambient temperature and humidity.

[0028] It should be noted that the preset gradient and preset ratio are set by the relevant staff themselves, and no specific restrictions are imposed here.

[0029] It should be noted that the BP network model's input layer is configured with two neurons, corresponding to ambient temperature and humidity; the activation layer has two neurons, with ReLU as the activation function; and the output layer has three neurons, corresponding to camera parameters exposure time, gain, and focal length, all of which are normalized. During model training, the Adam optimizer is used, with the loss function being the mean squared error between the predicted and actual optimal camera parameters. Iterative training is performed until the validation set MSE < 0.001. After training, the model is deployed to an industrial digital camera device. For example, when the environmental temperature and humidity are high... The sensor collects ambient temperature T of 35℃ and ambient humidity H of 65%RH. Normalization yields temperature data T_norm = 0.625 and humidity data H_norm = 0.7. Substituting the normalized temperature and humidity data into the temperature and humidity correction model outputs normalized camera parameters: exposure time of 0.58, gain of 0.32, and focal length of 0.65. After denormalization, the actual camera parameters are obtained: exposure time of 1370µs, gain of 10.24dB, and focal length of 11.5mm. The above example is merely illustrative and not the only possible interpretation.

[0030] Ambient light analysis module: After the working parameters of the industrial digital camera are calibrated, the module captures images of the target vehicle display screen in each preset direction using the industrial digital camera, thereby obtaining the RGB values ​​of the vehicle display screen in each preset direction. At the same time, it acquires the ambient light parameters of the industrial digital camera in each preset direction, and analyzes the correlation between the RGB values ​​and the ambient light parameters in each preset direction.

[0031] In a specific example, the analysis of the correlation between RGB values ​​and ambient light parameters in each preset direction is carried out as follows: An industrial digital camera is used to capture images of the target vehicle display screen in each preset direction, and the RGB values ​​of the target vehicle display screen in each preset direction are then converted. Simultaneously, ambient light parameters in each preset direction are collected using a photosensitive sensor. The RGB values ​​of the target vehicle display screen in each preset direction and the ambient light parameters are input into a pre-trained RGB value correlation model. The RGB correlation model expression then outputs the RGB correlation feature values ​​of the target vehicle display screen in each preset direction. These correlation feature values ​​include data of 1 and -1. When the correlation feature value is 1, it indicates that the RGB value correlation in each preset direction is normal; when the correlation feature value is -1, it indicates that the RGB value correlation in each preset direction is abnormal.

[0032] It should be noted that the RGB values ​​of the target vehicle display screen vary depending on the orientation and ambient light. Analyzing the relationship between the RGB values ​​of the target vehicle display screen and the orientation and ambient light parameters is beneficial for optimizing the visual performance of the vehicle display system, improving the user viewing experience, and solving the display deviation problem caused by changes in viewing angle and environment in practical applications.

[0033] It should be noted that the preset directions are set by the relevant staff according to the actual application of the vehicle display screen, and no specific restrictions are imposed here.

[0034] In a specific example, the training process for the RGB value association model is as follows: An industrial digital camera is used to test several different models of vehicle-mounted displays with known standard RGB values. The ambient light parameters in each preset direction are changed according to a preset gradient to obtain the test ambient light parameters. Under these test ambient light parameters, the industrial digital camera acquires RGB images of each vehicle-mounted display in each preset direction, thus obtaining the RGB values ​​of each vehicle-mounted display in each preset direction under the same test ambient light parameters. Simultaneously, the Fromm model serves as the basic framework, calculating the interference of ambient light parameters on RGB values ​​based on Lambert's law of reflection. The Lambert law of reflection and the human visual system model are superimposed onto the Fromm model. The test ambient light parameters, each preset direction, and the corresponding RGB images of each vehicle-mounted display in each preset direction are divided into training and testing sets according to a preset ratio. The model is then trained and validated to obtain the RGB value range visible to the human eye under the ambient light parameters in each preset direction, as well as the association model expression.

[0035] It should be noted that the expression for the RGB value association model is as follows: ,in Here, S represents the associated feature value, and S represents the detected RGB value. This represents the range of RGB values ​​that the human eye can see under the influence of ambient light parameters. It is obtained by training an RGB value association model.

[0036] It should be noted that the Fromm model, Lamb's law of reflection, and the human visual system model mentioned are all existing model technologies, and will not be elaborated upon here.

[0037] RGB value analysis module: When the RGB value correlation of the target vehicle display is normal, it divides the target vehicle display into regions and then determines whether the RGB difference of each region is abnormal; when the RGB value correlation of the target vehicle display is abnormal, it performs secondary calibration on the target vehicle display.

[0038] In a specific example, the process of dividing the target vehicle display screen into regions and then determining whether the RGB differences between each region are abnormal is as follows: The target vehicle display screen is divided into regions according to its display function. RGB values ​​of each region of the target vehicle display screen are collected under strong light and low light conditions using an industrial digital camera. The RGB values ​​of each region of the target vehicle display screen under strong light conditions are compared to obtain the RGB differences between each region. If the RGB differences between the regions of the target vehicle display screen meet the set RGB tolerance value under strong light conditions, it indicates that the RGB differences between the regions of the target vehicle display screen under strong light conditions are normal; otherwise, it indicates that the RGB differences between the regions of the target vehicle display screen under strong light conditions are abnormal. Similarly, the RGB differences between the regions of the target vehicle display screen under low light conditions are compared with the set RGB tolerance value under low light conditions to determine whether the RGB differences between the regions of the target vehicle display screen under low light conditions are abnormal.

[0039] It should be noted that dividing the target vehicle display screen into areas according to its display function can result in the core area of ​​the instrument panel, the functional area of ​​the central control screen, the HUD projection area, and the warning information area, etc.

[0040] It should be noted that the values ​​for strong and weak light are set by the relevant staff. For example, a light intensity of 5000 lux or higher may be defined as strong light, and a light intensity of less than 5000 lux may be defined as weak light. No specific restrictions are imposed here.

[0041] It should be noted that the RGB tolerance value for strong light includes both the illuminance tolerance value and the chromaticity tolerance value, while the RGB tolerance value for weak light includes both the illuminance tolerance value and the chromaticity tolerance value; for example, setting the illuminance tolerance value for strong light to 30 cd / m². 2 The permissible tolerance for chromaticity under strong light is 4.0; the permissible tolerance for illuminance under weak light is 15 cd / m². 2 The permissible tolerance for colorimetry under low light is 5.0. The above example is for illustrative purposes only. The specific value shall be set by the relevant personnel and no specific restrictions are imposed here.

[0042] In a specific example, when the correlation of the RGB values ​​of the target vehicle display screen is abnormal, a secondary calibration is performed on the target vehicle display screen, including calibration of the environmental sensor, temperature and humidity calibration of the industrial digital camera, and correlation analysis of the RGB values ​​of the vehicle display screen.

[0043] Data storage module: When an abnormal RGB value is detected in the target vehicle display, the abnormal data is stored in the corresponding path and an early warning is issued.

[0044] In a specific example, when an abnormal RGB value is detected in the target vehicle display screen, the abnormal data is stored in the corresponding path. The specific process is as follows: after secondary calibration of the target vehicle display screen, the correlation between RGB values ​​and ambient light parameters is analyzed for several consecutive frames of images in each preset direction, and the RGB difference of several consecutive frames of images in each region of the target vehicle display screen is analyzed for anomalies, thereby obtaining each abnormal frame image and storing each abnormal frame image in the corresponding path.

[0045] In a specific example, the warning prompt includes a voice prompt and a text prompt, wherein the voice prompt reads "RGB value abnormal"; and the text prompt includes the path of the abnormal frame image, the reason for the abnormality, and the RGB value of the abnormal frame image.

[0046] This application provides a visual RGB value recognition and detection system for industrial visual inspection. It corrects the operating parameters of an industrial digital camera based on environmental parameters and analyzes the correlation between the RGB values ​​of a vehicle-mounted display screen in each preset direction and the ambient light parameters in each preset direction. When the correlation of the RGB values ​​of the target vehicle-mounted display screen is normal, the display screen is divided into regions, and the RGB difference between each region is determined to be abnormal. When the correlation of the RGB values ​​of the target vehicle-mounted display screen is abnormal, a secondary correction is performed. This application not only minimizes the impact of ambient temperature and humidity on the RGB value detection results but also detects RGB values ​​from the user's visual perspective and ambient light parameters, solving the problem of the disconnect between technical parameters and user experience.

[0047] The above content is merely an example and illustration of the concept of this application. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the inventive concept or exceed the scope defined in this application, they should all fall within the protection scope of this application.

Claims

1. A visual RGB value recognition detection system for industrial visual inspection, characterized in that, The application relates to a vehicle-mounted display screen detection device and method. The device comprises: a detection device correction module for collecting environmental parameters during visual detection through an environmental sensor, and then correcting working parameters of an industrial digital camera according to the environmental parameters; an ambient light analysis module for taking images of a target vehicle-mounted display screen in each preset direction through the industrial digital camera after the working parameters of the industrial digital camera are corrected, obtaining RGB values of the vehicle-mounted display screen in each preset direction, and acquiring environmental light parameters of the industrial digital camera in each preset direction, so as to analyze the correlation between the RGB values and the environmental light parameters in each preset direction; an RGB value analysis module for dividing the target vehicle-mounted display screen into regions when the correlation between the RGB values of the target vehicle-mounted display screen is normal, and then judging whether the RGB difference values of the regions are abnormal; when the correlation between the RGB values of the target vehicle-mounted display screen is abnormal, the target vehicle-mounted display screen is corrected again; 2. The visual RGB value recognition detection system for industrial visual inspection according to claim 1, characterized in that, a data storage module for storing abnormal data to a corresponding path and giving a warning prompt when the RGB values of the target vehicle-mounted display screen are detected to be abnormal.

3. The visual RGB value recognition detection system for industrial visual inspection according to claim 2, characterized in that, The environmental sensor comprises a temperature sensor and a humidity sensor; the environmental parameters comprise environmental temperature and environmental humidity; and the working parameters of the industrial digital camera comprise exposure time, gain and focal length. The specific process of collecting the environmental parameters during visual detection through the environmental sensor and then correcting the working parameters of the industrial digital camera according to the environmental parameters is as follows:

4. The visual RGB value recognition detection system for industrial visual inspection according to claim 3, characterized in that, temperature and humidity sensors are installed at an internal photosensitive chip of the industrial digital camera, the temperature and humidity of the photosensitive chip at the time of collecting the RGB images of the target vehicle-mounted display screen are acquired through the temperature and humidity sensors, the temperature and humidity of the photosensitive chip at the time of collecting the images are compared with a set temperature threshold interval and a set humidity threshold interval, if the temperature or humidity of the photosensitive chip at the time of collecting the images does not belong to the set temperature threshold interval and the set humidity threshold interval, the temperature and humidity of the photosensitive chip are substituted into a pre-trained temperature and humidity correction model to correct the working parameters of the industrial digital camera, otherwise, if the temperature and humidity of the photosensitive chip at the time of collecting the images both belong to the set temperature threshold interval and the set humidity threshold interval, the working parameters of the industrial digital camera are not corrected. The specific training process of the temperature and humidity correction model is as follows: S1, using the industrial digital camera to test vehicle-mounted display screens of different types with known standard RGB values, first adjusting the environmental temperature in the set temperature threshold interval according to a preset gradient, and adjusting the environmental humidity outside the set humidity threshold interval according to a preset gradient, thereby obtaining a first group of test environment combinations, and then using the industrial digital camera to detect the RGB values of the vehicle-mounted display screens; S2, adjusting the environmental temperature outside the set temperature threshold interval according to a preset gradient, and adjusting the environmental humidity in the set humidity threshold interval according to a preset gradient, thereby obtaining a second group of test environment combinations, and then using the industrial digital camera to detect the RGB values of the vehicle-mounted display screens; S3, adjusting the ambient temperature and the ambient humidity according to a preset gradient outside a set temperature threshold interval and a humidity threshold interval, so as to obtain a third group of test environment combinations, and then using an industrial digital camera to detect the RGB values of each vehicle-mounted display screen; S4, dividing the first group of test environment combinations, the second group of test environment combinations and the third group of test environment combinations, and the corresponding collected RGB values and standard RGB values of each vehicle-mounted display screen into a training set and a verification set according to a preset proportion, and using a BP neural network model as a basic framework, training and verifying the BP neural network model using the training set and the verification set, and then obtaining a temperature and humidity correction model of the industrial digital camera.

5. The visual RGB value recognition detection system for industrial visual inspection according to claim 4, characterized in that, The correlation between the RGB values in each preset direction and the ambient light parameters is analyzed, and the specific analysis process is as follows: The images of the target vehicle-mounted display screen in each preset direction are captured using an industrial digital camera, and the RGB values of the target vehicle-mounted display screen in each preset direction are obtained by conversion. The ambient light parameters in each preset direction are collected by a photosensitive sensor, and the RGB values and ambient light parameters of the target vehicle-mounted display screen in each preset direction are input into a pre-trained RGB value correlation model. The RGB correlation characteristic value of the target vehicle-mounted display screen in each preset direction is output through the RGB value correlation model expression. The correlation characteristic value contains data of 1 and -1. When the correlation characteristic value is 1, it indicates that the RGB value correlation in each preset direction is normal, and when the correlation characteristic value is -1, it indicates that the RGB value correlation in each preset direction is abnormal.

6. The visual RGB value recognition detection system for industrial visual inspection according to claim 5, characterized in that, The RGB value correlation model is trained as follows: A number of vehicle-mounted display screens of different models with known standard RGB values are tested using an industrial digital camera. The ambient light parameters in each preset direction are changed according to a preset gradient, and the test ambient light parameters are obtained. The RGB images of each vehicle-mounted display screen in each preset direction are collected under the condition of each test ambient light parameter using an industrial digital camera, so as to obtain the RGB values of each vehicle-mounted display screen in each preset direction under the condition of each test ambient light parameter. The Flome model is used as the model basic framework, the interference of ambient light parameters on RGB values is calculated based on the Lambertian reflection law, and the Lambertian reflection law and the human eye visual system model are superimposed to the Flome model. Each test ambient light parameter, each preset direction and the corresponding RGB image of each vehicle-mounted display screen in each preset direction are divided into a training set and a test set according to a preset proportion, and the model is trained and verified, so as to obtain the RGB value interval that can be seen by the human eye under the ambient light parameters in each preset direction and the correlation model expression.

7. The visual RGB value recognition detection system for industrial visual inspection according to claim 6, characterized in that, The target vehicle-mounted display screen is divided into regions, and then it is determined whether the RGB difference value of each region is abnormal, and the specific analysis process is as follows: According to the display function, the target vehicle display screen is divided into regions, and the RGB values of each region of the target vehicle display screen under strong light and weak light conditions are collected by an industrial digital camera. The RGB values of each region of the target vehicle display screen under strong light conditions are compared with each other to obtain the RGB difference values of each region. If the RGB difference values between each region of the target vehicle display screen meet the set strong light RGB tolerance values, it indicates that the RGB difference values between each region of the target vehicle display screen under strong light conditions are normal. Otherwise, it indicates that the RGB difference values between each region of the target vehicle display screen under strong light conditions are abnormal. Similarly, the RGB difference values of each region of the target vehicle display screen under weak light conditions are compared with the set weak light RGB tolerance values to determine whether the RGB difference values between each region of the target vehicle display screen under weak light conditions are abnormal.

8. The visual RGB value recognition detection system for industrial visual inspection according to claim 7, characterized in that, When the correlation between the RGB values of the target vehicle display screen is abnormal, the target vehicle display screen is corrected again, including correcting the environmental sensor, correcting the industrial digital camera for temperature and humidity, and analyzing the correlation between the RGB values of the vehicle display screen.

9. The visual RGB value recognition detection system for industrial visual inspection according to claim 8, characterized in that, When the RGB values of the target vehicle display screen are detected to be abnormal, the abnormal data is stored in the corresponding path, and the specific process is as follows: After the target vehicle display screen is corrected again, the correlation between the RGB values and the ambient light parameters of a plurality of frames of continuous images in each preset direction is analyzed, and the RGB difference values of a plurality of frames of continuous images in each region of the target vehicle display screen are analyzed for abnormality. Then, each abnormal frame image is obtained and stored in the corresponding path.

10. The visual RGB value recognition detection system for industrial visual inspection according to claim 9, wherein, The early warning prompt includes voice prompt and text prompt, wherein the voice prompt content is "RGB value abnormality"; the text prompt content includes the abnormal frame image path, the abnormal reason and the RGB value of the abnormal frame image.

Images