CN117542301A - Display screen detection method and system - Google Patents
Display screen detection method and system Download PDFInfo
- Publication number
- CN117542301A CN117542301A CN202410033306.0A CN202410033306A CN117542301A CN 117542301 A CN117542301 A CN 117542301A CN 202410033306 A CN202410033306 A CN 202410033306A CN 117542301 A CN117542301 A CN 117542301A
- Authority
- CN
- China
- Prior art keywords
- display screen
- detection
- module
- image
- features
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000005286 illumination Methods 0.000 claims abstract description 14
- 230000002159 abnormal effect Effects 0.000 claims abstract description 11
- 238000009826 distribution Methods 0.000 claims abstract description 11
- 238000013527 convolutional neural network Methods 0.000 claims description 29
- 238000000605 extraction Methods 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 17
- 230000007547 defect Effects 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 238000002790 cross-validation Methods 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 4
- 238000010200 validation analysis Methods 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 239000003086 colorant Substances 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000005375 photometry Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/006—Electronic inspection or testing of displays and display drivers, e.g. of LED or LCD displays
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M11/00—Testing of optical apparatus; Testing structures by optical methods not otherwise provided for
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M11/00—Testing of optical apparatus; Testing structures by optical methods not otherwise provided for
- G01M11/02—Testing optical properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30121—CRT, LCD or plasma display
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Analytical Chemistry (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a detection method and a detection system for a display screen, which relate to the technical field of display screen detection, and comprise the following steps of firstly, sensing environment, and detecting environment parameters through a sensor, wherein the environment parameters comprise at least one of illumination intensity, temperature and humidity; step two, multi-mode detection, aiming at different types of display screens, adopting corresponding detection models and parameters; and thirdly, extracting features related to the detection task of the display screen from the sample data, wherein the features are pixel value color difference and brightness distribution. By combining environment sensing and multi-mode detection, the method can adapt to different display screen models and environment conditions, so that the accuracy and reliability of detection are improved, the accuracy and robustness of a model can be improved, abnormal data and complex scene colors can be better handled, and by classifying, positioning and identifying the abnormal data, the method can timely find out product problems, take corresponding measures and improve user satisfaction.
Description
Technical Field
The invention relates to the technical field of display screen detection, in particular to a detection method and a detection system for a display screen.
Background
The detection of display screens plays a vital role in various industries, such as consumer electronics, automobile manufacturing, educational equipment, medical equipment, etc., and with the development of technology, the variety and complexity of display screens are increasing, which puts higher demands on detection methods.
In the existing display screen detection technology, the following defects still exist:
the traditional display screen detection method can detect a display screen with a specific model, but cannot adapt to different display screen models and environmental conditions, which can lead to inaccurate or unreliable detection results;
in addition, the conventional display screen detection method may only pay attention to processing and analysis of image data, and ignore data of other modes such as sound, temperature and the like, which may result in that the working state and quality of the display screen cannot be fully known. Therefore, we propose a method and a system for detecting a display screen to solve the above-mentioned technical problems.
Disclosure of Invention
In view of the above, the main objective of the present invention is to provide a method and a system for detecting a display screen, which are capable of solving the above-mentioned problems.
In order to achieve the above purpose, the present invention adopts the following technical scheme: the method comprises the following steps:
sensing environment, namely detecting environment parameters through a sensor, wherein the environment parameters comprise at least one of illumination intensity, temperature and humidity;
step two, multi-mode detection, aiming at different types of display screens, adopting corresponding detection models and parameters;
extracting features related to a detection task of the display screen from sample data, wherein the features are pixel value color difference and brightness distribution, selecting proper features according to task requirements, and removing redundant and irrelevant features;
step four, adopting a convolutional neural network to improve an automatic detection system so that screen defects and problems can be more accurately identified;
training and optimizing the model by utilizing the characteristics and the label data extracted in the step three, improving the accuracy and the robustness of the model, and performing model tuning by adopting a cross verification and grid search technology;
and step six, analyzing and processing the detection result, classifying, positioning and identifying the abnormal data, and feeding back the detection result to a user or equipment control system so as to take corresponding measures.
In a preferred embodiment, in the first step, an optimal illumination range suitable for detection of the display screen is determined, the requirements of different types of display screens for illumination may be different, an optimal temperature environment suitable for detection of the display screen is continuously determined, and an optimal humidity environment suitable for detection of the display screen is determined.
In a preferred embodiment, in the second step, the multi-mode detection includes an image mode and a voice mode, the image mode detects an image, and the voice information detects the receiving and sending of the voice by the display screen.
As a preferable solution, in the third and fourth steps, feature extraction is performed on the detected content of the display screen, which includes the following parts:
a. edge detection: identifying boundaries and contours in the image;
b. color and color space analysis: analyzing color information in the image, and identifying color distribution and characteristics of different areas on the display screen;
c. and (3) extracting physical characteristics: identifying and describing textures of the surface of the display screen;
d. extracting the shape, size and geometric characteristics of an object on a display screen;
the method is used for learning the features in the image through a convolutional neural network, and capturing the local and global features of the image through a convolutional layer and a pooling layer.
In a preferred embodiment, in the fifth step, the cross-validation is performed by dividing the data set into a plurality of subsets, and alternately using a part of the data set as a validation set and the other part as a training set to evaluate the performance of the model;
the grid search finds the best performance configuration by systematically searching combinations of model hyper-parameters.
As a preferred scheme, the detection system of the display screen comprises an image acquisition module, a sound collection module, a feature extraction module, a convolutional neural network module, a decision module and a user interface module, wherein the image acquisition module, the sound collection module, the feature extraction module and the convolutional neural network module are all connected with the decision module, the decision module is connected with the user interface module, the feature extraction module is used for classifying and extracting features in the image acquisition module and the sound collection module, and the convolutional neural network module is used for identifying defects in the features.
As a preferred scheme, the image acquisition module is responsible for acquiring images from a display screen or related equipment;
the sound collection module is responsible for acquiring playback audio and voice interaction data of the display screen.
As a preferred solution, the feature extraction module extracts features of objects or regions obtained from the image to help the system better understand the content in the image;
the convolutional neural network module is used for identifying a set excess value for the characteristics of an object or region obtained in the image.
As a preferred scheme, the decision module is used for controlling the detection step of the whole display screen, judging the quality of the display screen, and determining whether the display screen is in a good state or not based on the judgment made by the convolutional neural network module;
the user interface module records the test results and system performance and provides feedback to other systems or operators to improve system performance or to conduct further analysis while the user controls the operation and stopping of the entire test system.
Compared with the prior art, the invention has obvious advantages and beneficial effects, in particular, the technical proposal can be adopted to realize that the invention mainly comprises the following steps:
according to the method, through combining environment perception and multi-mode detection, the method can adapt to different display screen models and environment conditions, so that the accuracy and reliability of detection are improved, an automatic detection system and a convolutional neural network are adopted, the method can realize rapid and automatic detection of the display screen, the production efficiency and the product quality are improved, the accuracy and the robustness of the model can be improved through training and optimizing the model, abnormal data and complex scene colors are better dealt with, and through classifying, positioning and identifying the abnormal data, the method can discover product problems in time, take corresponding measures and improve the user satisfaction.
In order to more clearly illustrate the structural features and efficacy of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and examples.
Drawings
FIG. 1 is a diagram of a method for inspecting a display screen according to an embodiment of the present invention.
Detailed Description
For the purpose of making the technical solution and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and examples of implementation. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
It will be understood that when an element is referred to as being "fixed to" another element, it can be directly on the other element or intervening elements may also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. The terms "vertical," "horizontal," "left," "right," and the like are used herein for illustrative purposes only.
Referring to fig. 1, an embodiment of the present invention provides a method and a system for detecting a display screen, including the following steps:
detecting environmental parameters including at least one of illumination intensity, temperature and humidity through a sensor, wherein the step aims to acquire relevant information of the environment where the display screen is located, and the information possibly affects the performance and reliability of the display screen;
the brightness of the display screen can be influenced by ambient illumination, under the condition of over-strong or over-weak illumination, the display effect of the display screen can be deviated, such as contrast reduction, color distortion and the like, and the real environment where the display screen is positioned can be known by detecting the illumination intensity of the environment, so that the detection result can be evaluated more accurately;
the working temperature of the display screen can influence the performance and stability, and under the condition of overhigh or overlow temperature, the display screen can have the problems of uneven display, color deviation and the like, and the temperature condition of the display screen can be known by detecting the temperature of the environment, so that the possible problems can be early warned;
the high ambient humidity may cause the internal circuit of the display screen to be wetted, thereby affecting the performance and the service life of the display screen.
Step two, multi-mode detection, adopting corresponding detection models and parameters for different types of display screens, wherein the different types of display screens possibly have different performance characteristics and technical requirements, so that special detection and evaluation are required for the characteristics of the different types of display screens, the different types of display screens possibly need to be evaluated by using different detection models, and a detection method based on image processing can be considered for a Liquid Crystal Display (LCD);
aiming at different types of display screens, a plurality of detection technologies can be combined to obtain a more comprehensive evaluation result, and an image processing technology and a photometry technology are combined to evaluate the color and brightness performance of the display screen;
the multi-mode detection can be realized through an automatic detection system and a machine learning algorithm, and through selecting proper detection models and parameters and combining multiple detection technologies, the comprehensive and accurate detection and evaluation of different types of display screens can be realized, so that the production efficiency and the product quality can be improved, and meanwhile, the cost and the error of manual detection can be reduced.
Extracting features related to a detection task of the display screen from sample data, wherein the features are pixel value color difference and brightness distribution, selecting proper features according to task requirements, and removing redundant and irrelevant features;
by comparing the color values of different pixel points, the characteristics of color deviation, color uniformity and the like of the display screen can be known, specifically, the color difference between adjacent pixel points can be calculated, or the color distribution condition of the whole image can be calculated so as to reflect the whole color performance of the display screen, the brightness distribution characteristic describes the brightness condition of a dark area in the image, and the brightness and contrast performance of the display screen can be known by analyzing the distribution of the brightness and the dark area;
after feature extraction, the obtained result is used as input data for subsequent model training and prediction. By selecting appropriate features and removing redundant and irrelevant features, the efficiency and accuracy of the model can be improved, thereby better solving the detection problem of a specific display screen.
And fourthly, adopting a convolutional neural network to improve an automatic detection system, so that the screen defects and problems can be more accurately identified, adopting the CNN has the advantages of strong feature learning and classifying capability, and automatically learning useful features from a large amount of image data, thereby improving the accuracy and robustness of the automatic detection system, and in addition, the CNN can also process various types of image data without being limited by specific data formats or task types, and has better generalization capability.
Training and optimizing the model by utilizing the features and the label data extracted in the step three, improving the accuracy and the robustness of the model, performing model tuning by adopting a cross verification and grid search technology, and using the features extracted in the step one as input data for training and optimizing the model, wherein the features can comprise pixel value color difference and brightness distribution;
training the model by using label data, wherein the label data refers to classification results of display screen defects and problems corresponding to the extracted features, and the model can learn a mapping relation from the features to the labels through a training process;
in order to improve the accuracy and robustness of the model, various optimization techniques may be employed, and the model may be evaluated and tuned using cross-validation techniques, which may be performed by dividing the data set into multiple subsets and performing model training and testing on each subset separately to obtain a more accurate performance evaluation;
grid search is a tuning technique, an optimal model parameter combination is found by searching grid points in a parameter space, a group of super parameters such as learning rate, batch size, iteration number and the like are set in the model training process, different parameter combinations are tried by the grid search, and the parameter combination with the optimal performance is selected as the optimal parameter;
evaluating the performance of the optimized model by using the test data set, and comparing the predicted result of the model on the test data set with the real label to know the performance of the model and further adjust the parameters of the model or change the architecture of the model according to the requirement;
the optimized model is applied to an actual display screen detection task, and the model outputs recognition and classification results of screen defects and problems by inputting images to be detected.
And step six, analyzing and processing the detection result, classifying, positioning and identifying the abnormal data, and feeding back the detection result to a user or equipment control system so as to take corresponding measures.
In the embodiment, the method can adapt to different display screen models and environment conditions by combining environment sensing and multi-mode detection, so that the accuracy and reliability of detection are improved, an automatic detection system and a convolutional neural network are adopted, the method can realize rapid and automatic detection of the display screen, the production efficiency and the product quality are improved, the accuracy and the robustness of the model can be improved by training and optimizing the model, abnormal data and complex scene colors are better dealt with, and the abnormal data are classified, positioned and marked, so that the method can discover product problems in time, take corresponding measures and improve the user satisfaction.
Referring to fig. 1, in the first step, an optimal illumination range suitable for detection of a display screen is determined, the requirements of different types of display screens for illumination may be different, and an optimal temperature environment suitable for detection of the display screen is continuously determined, and an optimal humidity environment suitable for detection of the display screen is determined.
Referring to fig. 1, in the second step, the multi-mode detection includes an image mode and a voice mode, the image mode detects an image, and the voice information detects the receiving and sending of the voice by the display screen.
Referring to fig. 1, in the third and fourth steps, feature extraction is performed on the detected content of the display screen, which includes the following parts:
a. edge detection: the boundaries and the outlines in the image are identified, and common algorithms comprise Sobel, canny and the like, so that the boundaries and the areas of the display screen can be positioned;
b. color and color space analysis: analyzing color information in the image, identifying color distribution and characteristics of different areas on the display screen, and color space conversion (such as RGB to HSV, LAB and the like) is helpful for better describing the color information;
c. and (3) extracting physical characteristics: identifying and describing the texture of the surface of the display screen, wherein the texture features describe the appearance and texture of local areas of the image, which helps to distinguish the material differences of different areas;
d. extracting the shape, size and geometric characteristics of an object on the display screen, which can be realized by contour detection, shape matching and other technologies;
the method is used for learning the features in the image through a convolutional neural network, and capturing the local and global features of the image through a convolutional layer and a pooling layer.
Referring to fig. 1, in the fifth step, the cross-validation is performed by dividing the data set into a plurality of subsets, and alternately using a part of the data set as a validation set and the other part as a training set to evaluate the performance of the model; this helps to prevent the model from over-fitting or under-fitting on a particular dataset, improving the generalization ability of the model to unknown data by systematically searching combinations of model hyper-parameters, which are parameters that need to be set before model training, such as learning rate, regularization parameters, etc., to find the best performance configuration. Grid searching attempts different combinations of hyper-parameters to evaluate their performance on the validation set to find the best hyper-parameter configuration.
Referring to fig. 1, the system comprises an image acquisition module, a sound collection module, a feature extraction module, a convolutional neural network module, a decision module and a user interface module, wherein the image acquisition module, the sound collection module, the feature extraction module and the convolutional neural network module are all connected with the decision module, the decision module is connected with the user interface module, the feature extraction module is used for classifying and extracting features in the image acquisition module and the sound collection module, and the convolutional neural network module is used for identifying defects in the features;
the image acquisition module is responsible for acquiring images from a display screen or related equipment;
the sound collection module is responsible for acquiring playback audio and voice interaction data of the display screen;
the feature extraction module extracts features of objects or regions obtained from the image to help the system better understand the content in the image;
the convolutional neural network module is used for identifying a set excess value for the characteristics of an object or an area obtained from the image;
the decision module is used for controlling the detection step of the whole display screen, judging the quality of the display screen, and determining whether the display screen is in a good state or not based on the judgment made by the convolutional neural network module;
the user interface module records the detection result and the system performance and provides feedback to other systems or operators to improve the system performance or perform further analysis, and simultaneously, the user controls the operation and stop of the whole detection system;
in this embodiment, the image acquisition module is responsible for capturing and acquiring image data of the display screen, and may include one or more high-definition cameras, which may capture images at different angles and under different illumination conditions of the display screen, and may adjust settings of the cameras, such as focal length, resolution, exposure, etc., as required;
the sound collecting module is responsible for collecting and recording sound emitted by the display screen during operation, and can know the working state and potential problems of the display screen by analyzing sound data, and the module can comprise a sound sensor and a recording device so as to capture and store the sound data;
the feature extraction module is responsible for carrying out feature extraction and processing on the data collected by the image acquisition module and the sound collection module, can extract features related to detection of a display screen, such as color, brightness, contrast and the like, and carries out necessary pretreatment and analysis on sound data;
a convolutional neural network module (CNN) is a deep learning algorithm that processes and analyzes input feature data, by training a CNN model, can identify and classify defects and problems in the features of a display screen that will receive input from a feature extraction module and output recognition results that will be used by a decision module.
In summary, by combining environment sensing and multi-mode detection, the method can adapt to different display screen models and environment conditions, so that the accuracy and reliability of detection are improved, an automatic detection system and a convolutional neural network are adopted, the method can realize rapid and automatic detection of the display screen, the production efficiency and the product quality are improved, the accuracy and the robustness of the model can be improved by training and optimizing the model, abnormal data and complex scene colors are better dealt with, and by classifying, positioning and identifying the abnormal data, the method can discover product problems in time, take corresponding measures and improve the user satisfaction.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalents, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (9)
1. A detection method of a display screen is characterized in that: the method comprises the following steps:
sensing environment, namely detecting environment parameters through a sensor, wherein the environment parameters comprise at least one of illumination intensity, temperature and humidity;
step two, multi-mode detection, aiming at different types of display screens, adopting corresponding detection models and parameters;
extracting features related to a detection task of the display screen from sample data, wherein the features are pixel value color difference and brightness distribution, selecting proper features according to task requirements, and removing redundant and irrelevant features;
step four, adopting a convolutional neural network to improve an automatic detection system so that screen defects and problems can be more accurately identified;
training and optimizing the model by utilizing the characteristics and the label data extracted in the step three, improving the accuracy and the robustness of the model, and performing model tuning by adopting a cross verification and grid search technology;
and step six, analyzing and processing the detection result, classifying, positioning and identifying the abnormal data, and feeding back the detection result to a user or equipment control system so as to take corresponding measures.
2. The method for detecting a display screen according to claim 1, wherein: in the first step, an optimal illumination range suitable for detection of the display screen is determined, the requirements of different types of display screens on illumination may be different, an optimal temperature environment suitable for detection of the display screen is continuously determined, and an optimal humidity environment suitable for detection of the display screen is determined.
3. The method for detecting a display screen according to claim 1, wherein: in the second step, the multi-mode detection includes an image mode and a voice mode, the image mode detects an image, and the voice information detects the receiving and sending of voice by the display screen.
4. The method for detecting a display screen according to claim 1, wherein: in the third and fourth steps, feature extraction is performed on the detected content of the display screen, and the method comprises the following steps:
a. edge detection: identifying boundaries and contours in the image;
b. color and color space analysis: analyzing color information in the image, and identifying color distribution and characteristics of different areas on the display screen;
c. and (3) extracting physical characteristics: identifying and describing textures of the surface of the display screen;
d. extracting the shape, size and geometric characteristics of an object on a display screen;
the method is used for learning the features in the image through a convolutional neural network, and capturing the local and global features of the image through a convolutional layer and a pooling layer.
5. The method for detecting a display screen according to claim 1, wherein: in the fifth step, the cross-validation is a method for evaluating the performance of the model by dividing the data set into a plurality of subsets and alternately using one part of the data set as a validation set and the other part as a training set;
the grid search finds the best performance configuration by systematically searching combinations of model hyper-parameters.
6. A display screen detection system, applied to a display screen detection method as claimed in any one of claims 1 to 5, characterized in that: the device comprises an image acquisition module, a sound collection module, a feature extraction module, a convolutional neural network module, a decision module and a user interface module, wherein the image acquisition module, the sound collection module, the feature extraction module and the convolutional neural network module are all connected with the decision module, the decision module is connected with the user interface module, the feature extraction module is used for classifying and extracting features in the image acquisition module and the sound collection module, and the convolutional neural network module is used for identifying defects in the features.
7. The display screen detection system of claim 6, wherein: the image acquisition module is responsible for acquiring images from a display screen or related equipment;
the sound collection module is responsible for acquiring playback audio and voice interaction data of the display screen.
8. The display screen detection system of claim 6, wherein: the feature extraction module extracts features of objects or regions obtained from the image to help the system better understand the content in the image;
the convolutional neural network module is used for identifying a set excess value for the characteristics of an object or region obtained in the image.
9. The display screen detection system of claim 6, wherein: the decision module is used for controlling the detection step of the whole display screen, judging the quality of the display screen, and determining whether the display screen is in a good state or not based on the judgment made by the convolutional neural network module;
the user interface module records the test results and system performance and provides feedback to other systems or operators to improve system performance or to conduct further analysis while the user controls the operation and stopping of the entire test system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410033306.0A CN117542301A (en) | 2024-01-10 | 2024-01-10 | Display screen detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410033306.0A CN117542301A (en) | 2024-01-10 | 2024-01-10 | Display screen detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117542301A true CN117542301A (en) | 2024-02-09 |
Family
ID=89786599
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410033306.0A Pending CN117542301A (en) | 2024-01-10 | 2024-01-10 | Display screen detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117542301A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118192112A (en) * | 2024-03-27 | 2024-06-14 | 深圳市爱富佳科技有限公司 | AI-based liquid crystal display fault detection method, device, equipment and medium |
| CN118294113A (en) * | 2024-03-29 | 2024-07-05 | 深圳市云之声科技有限公司 | Digital LED illumination quality evaluation method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105301810A (en) * | 2015-11-24 | 2016-02-03 | 上海斐讯数据通信技术有限公司 | Screen defect detecting method and screen defect detecting device |
| CN206271394U (en) * | 2016-11-02 | 2017-06-20 | 清投智能（北京）科技股份有限公司 | A kind of giant-screen of automatic fault detection |
| CN108802041A (en) * | 2018-03-16 | 2018-11-13 | 浙江大学 | A kind of method that the small sample set of screen detection is quickly remodeled |
| CN108961238A (en) * | 2018-07-02 | 2018-12-07 | 北京百度网讯科技有限公司 | Display screen quality determining method, device, electronic equipment and storage medium |
| CN116681995A (en) * | 2023-06-13 | 2023-09-01 | 上海帆声图像科技有限公司 | Application of grpc-based multi-model fusion scheme in display screen detection |
| CN116739991A (en) * | 2023-05-16 | 2023-09-12 | 东莞理工学院 | Liquid crystal display screen surface defect detection method based on deep learning and electronic device |
-
2024
- 2024-01-10 CN CN202410033306.0A patent/CN117542301A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105301810A (en) * | 2015-11-24 | 2016-02-03 | 上海斐讯数据通信技术有限公司 | Screen defect detecting method and screen defect detecting device |
| CN206271394U (en) * | 2016-11-02 | 2017-06-20 | 清投智能（北京）科技股份有限公司 | A kind of giant-screen of automatic fault detection |
| CN108802041A (en) * | 2018-03-16 | 2018-11-13 | 浙江大学 | A kind of method that the small sample set of screen detection is quickly remodeled |
| CN108961238A (en) * | 2018-07-02 | 2018-12-07 | 北京百度网讯科技有限公司 | Display screen quality determining method, device, electronic equipment and storage medium |
| CN116739991A (en) * | 2023-05-16 | 2023-09-12 | 东莞理工学院 | Liquid crystal display screen surface defect detection method based on deep learning and electronic device |
| CN116681995A (en) * | 2023-06-13 | 2023-09-01 | 上海帆声图像科技有限公司 | Application of grpc-based multi-model fusion scheme in display screen detection |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118192112A (en) * | 2024-03-27 | 2024-06-14 | 深圳市爱富佳科技有限公司 | AI-based liquid crystal display fault detection method, device, equipment and medium |
| CN118192112B (en) * | 2024-03-27 | 2024-11-08 | 深圳市爱富佳科技有限公司 | AI-based liquid crystal display fault detection method, device, equipment and medium |
| CN118294113A (en) * | 2024-03-29 | 2024-07-05 | 深圳市云之声科技有限公司 | Digital LED illumination quality evaluation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117542301A (en) | Display screen detection method and system | |
| CN114549522A (en) | Textile quality detection method based on target detection | |
| US11657599B2 (en) | Method for detecting appearance of six sides of chip multi-layer ceramic capacitor based on artificial intelligence | |
| CN111667455A (en) | AI detection method for various defects of brush | |
| JP2022027473A (en) | Generation of training data usable for inspection of semiconductor sample | |
| CN108711148A (en) | A kind of wheel tyre defect intelligent detecting method based on deep learning | |
| JP2021515885A (en) | Methods, devices, systems and programs for setting lighting conditions and storage media | |
| CN113177924A (en) | Industrial production line product flaw detection method | |
| CN112862744A (en) | Intelligent detection method for internal defects of capacitor based on ultrasonic image | |
| CN114926441A (en) | Defect detection method and system for machining and molding injection molding part | |
| CN115294033A (en) | Tire belt layer difference level and misalignment defect detection method based on semantic segmentation network | |
| CN116559111A (en) | Sorghum variety identification method based on hyperspectral imaging technology | |
| CN115205209A (en) | Monochrome cloth flaw detection method based on weak supervised learning | |
| CN107610119A (en) | The accurate detection method of steel strip surface defect decomposed based on histogram | |
| CN113390882A (en) | Tire inner side defect detector based on machine vision and deep learning algorithm | |
| CN114226262A (en) | Flaw detection method, flaw classification method and flaw detection system | |
| CN112070762A (en) | Mura defect detection method and device for liquid crystal panel, storage medium and terminal | |
| CN112102253A (en) | Non-woven fabric surface defect automatic detection method and system based on machine vision | |
| JPH08189904A (en) | Surface defect detector | |
| CN114549414A (en) | Abnormal change detection method and system for track data | |
| CN113435444B (en) | Immunochromatography detection method, immunochromatography detection device, storage medium and computer equipment | |
| CN116091506A (en) | Machine vision defect quality inspection method based on YOLOV5 | |
| CN212646436U (en) | Artificial board surface flaw detection device | |
| CN113642473A (en) | Mining coal machine state identification method based on computer vision | |
| CN113963230A (en) | Parking space detection method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |