Surface defect detection method of notebook computer shell, controller and equipment

By taking pictures of the six surfaces of the notebook shell at different brightness and comparing and feature extraction using AI detection models, the accuracy of visual detection under uncontrolled conditions is solved, and efficient and accurate defect detection is achieved.

CN118937332AInactive Publication Date: 2024-11-12YITONG TECH (SHENZHEN) CO LTD

Patent Information

Application Number
CN202410877473.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-02
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Under uncontrolled conditions, it is difficult to accurately detect defects on the surface of the notebook shell, and is greatly affected by factors such as light, morphology and occlusion, resulting in inaccurate detection results.

Method used

The light source component and the shooting component are used to photograph six surfaces of the notebook shell, and the image is identified by using the AI ​​detection model. Through comparison with the defect database and feature extraction, the surface recognition results are determined, and the shooting is taken at different brightnesses. Improve the fault tolerance and stability of detection.

Benefits of technology

It realizes efficient and accurate detection and identification of defect information on the surface of the notebook housing, improves the fault tolerance and stability of the detection equipment, and reduces the impact of light changes on the detection results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118937332A_ABST
    Figure CN118937332A_ABST
Patent Text Reader

Abstract

The invention discloses a notebook shell surface defect detection method, controller and equipment, and the method comprises the steps: sequentially shooting six shell surfaces of a notebook shell through a shooting assembly and a light source assembly, deploying a corresponding AI detection model according to the shell surfaces corresponding to images shot by the shooting assembly, performing defect identification on the image through the AI detection model to obtain an image defect identification result, and determining an identification result of the shell surface according to the image defect identification result belonging to the same shell surface; according to the embodiment of the invention, regional detection is carried out through a plurality of images, the error-tolerant rate and the stability of the detection equipment are improved, and the defect information condition of the surface of the notebook computer shell is efficiently and accurately detected and identified.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, a controller and a device for detecting surface defects of a notebook shell. Background Art

[0002] With the rapid development of science and technology, visual inspection has become a very common inspection tool in modern industrial production. Compared with the human eye, visual inspection can better detect errors in the production process, better detect product quality problems, improve industrial production efficiency and production automation, and improve the accuracy of industrial production, improve and speed up the workflow and save time.

[0003] However, as visual inspection applications become more and more widespread, the application scenarios are becoming more and more diverse. Under uncontrolled conditions, the content of images and videos is easily affected by lighting, shape, occlusion, etc., resulting in large changes in the images detected by vision, which are far from the actual objects in the real scene, making it difficult to accurately detect appearance defects. Summary of the invention

[0004] The embodiments of the present application provide a method, a controller, and a device for detecting surface defects of a notebook shell, which can efficiently and accurately detect and identify defect information on the surface of the notebook shell.

[0005] In a first aspect, an embodiment of the present application provides a surface defect detection method for a notebook housing, which is applied to a surface defect detection device, wherein the surface defect detection device includes a light source component and a shooting component; the surface defect detection method includes:

[0006] The six shell surfaces of the notebook shell are photographed in sequence by the photographing component and the light source component;

[0007] Deploy a corresponding AI detection model on the shell surface corresponding to the image captured by the shooting component;

[0008] Performing defect recognition on the image by using the AI ​​detection model to obtain an image defect recognition result;

[0009] Determining a recognition result of the shell surface according to the image defect recognition results belonging to the same shell surface;

[0010] Among them, the image of the front side of the notebook shell is photographed at different brightnesses of the light source assembly, the shell structure surface of the notebook shell is photographed at the preset brightness of the light source assembly, and the images of the left side, right side, front side and rear side of the notebook shell are all photographed in different areas at the preset brightness of the light source assembly.

[0011] In some embodiments, the performing defect recognition on the image by the AI ​​detection model to obtain an image defect recognition result includes:

[0012] Compare the image with the defect template in the defect database by using the AI ​​detection model to obtain defect similarity;

[0013] Determining the image defect recognition result according to the defect similarity and a preset threshold;

[0014] The defect database is used to store sample templates of various defects.

[0015] In some embodiments, the defect comparison of the image using the AI ​​detection model and the defect template in the defect database to obtain the defect similarity includes:

[0016] Extract features from the image using the AI ​​detection model to obtain image features;

[0017] The defect similarity is obtained by comparing the image features with the defect features in the defect template through the AI ​​detection model.

[0018] In some embodiments, the AI ​​detection model corresponding to the shell surface corresponding to the image captured by the shooting component is deployed, including:

[0019] Numbering the images according to the shell surface corresponding to the images captured by the shooting component;

[0020] An AI detection model corresponding to the shell surface is deployed according to the serial number of the image.

[0021] In some embodiments, determining the recognition result of the shell surface according to the image defect recognition results belonging to the same shell surface includes:

[0022] When at least one of the image defect recognition results belonging to the same shell surface is defective, determining that the recognition result of the shell surface is defective;

[0023] When all the image defect recognition results belonging to the same shell surface are defect-free, it is determined that the recognition result of the shell surface is defect-free.

[0024] In some embodiments, after determining the recognition result of the shell surface according to the image defect recognition result belonging to the same shell surface, the surface defect detection method further includes:

[0025] updating the AI ​​detection model according to the recognition result of the shell surface;

[0026] The recognition results of the six shell surfaces of the notebook shell are summarized and outputted.

[0027] In some embodiments, after summarizing and outputting the recognition results of the six shell surfaces of the notebook shell, the surface defect detection method further includes:

[0028] Obtaining manual re-judgment results with the recognition results of the six shell surfaces of the notebook shell as a comparison;

[0029] The AI ​​detection model is updated according to the manual re-judgment result.

[0030] In a second aspect, an embodiment of the present application provides a controller including a memory and a processor, wherein the memory stores a computer program, and the processor implements the surface defect detection method described in the first aspect when executing the computer program.

[0031] In a third aspect, an embodiment of the present application provides a surface defect detection device, including a light source component, a shooting component and a controller as described in the second aspect.

[0032] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, the computer executes the method described in the first aspect.

[0033] A surface defect detection method, controller and device for a notebook shell provided by an embodiment of the present application have at least the following beneficial effects: in the embodiment of the present application, six shell surfaces of the notebook shell are photographed in sequence by a shooting component and a light source component, and a corresponding AI detection model is deployed according to the shell surface corresponding to the image photographed by the shooting component, and the image is defect-recognized by the AI ​​detection model to obtain an image defect recognition result, and the recognition result of the shell surface is determined according to the image defect recognition result belonging to the same shell surface; therefore, the embodiment of the present application obtains an image of the front of the notebook shell at different brightnesses, an image of the structural surface of the notebook shell at a preset brightness, and partial area images of the left side, right side, front side and rear side of the notebook shell at a preset brightness through the shooting component and the light source component, and obtains more comprehensive information of the shell surface of the notebook shell by acquiring images of the six surfaces, and deploys a corresponding AI detection model according to the shell surface corresponding to the image, and detects the image by the AI ​​detection model corresponding to the shell surface, thereby preventing the AI ​​detection model from misrecognizing in the process of recognizing the image, and determining the recognition result of the shell surface according to the image defect recognition result belonging to the same shell surface, thereby improving the fault tolerance and stability of the detection equipment, and realizing efficient and accurate detection and recognition of defect information on the surface of the notebook shell.

[0034] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic flow chart of a method for detecting surface defects of a notebook shell provided in an embodiment of the present application;

[0036] Figure 2 It is a flowchart of a defect identification method provided in an embodiment of the present application;

[0037] Figure 3 It is a flowchart of a defect comparison method provided in an embodiment of the present application;

[0038] Figure 4 It is a flowchart of an AI detection model deployment method provided in an embodiment of the present application;

[0039] Figure 5 It is a flowchart of a method for determining a shell surface recognition result provided by an embodiment of the present application;

[0040] Figure 6 It is a flowchart of a post-detection update summary method provided in an embodiment of the present application;

[0041] Figure 7 It is a flowchart of a re-judgment update method provided in an embodiment of the present application;

[0042] Figure 8 is a structural diagram of a controller provided in an embodiment of the present application;

[0043] Fig. 9 It is a structural schematic diagram of a surface defect detection device provided in an embodiment of the present application;

[0044] Fig.10 It is a schematic diagram of the overall process of a surface defect detection method for a notebook shell provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. In addition, the characteristics, operations or features described in the specification can be combined in any appropriate manner to form various implementation methods. At the same time, the steps or actions in the method description can also be replaced or adjusted in order in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the accompanying drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a necessary sequence, unless otherwise specified that a certain sequence must be followed.

[0046] In the description of this application, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed", etc. are understood to exclude the number itself, and "above", "below", "within", etc. are understood to include the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0047] The serial numbers of the components in this document, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in this application, unless otherwise specified, include direct and indirect connections (couplings).

[0048] With the rapid development of science and technology, visual inspection has become a very common inspection tool in modern industrial production. The characteristics of machine vision inspection are to improve the flexibility and automation of production. In some dangerous working environments that are not suitable for manual operation or where artificial vision cannot meet the requirements, machine vision is often used to replace artificial vision.

[0049] Compared with manual vision, visual inspection can better detect errors in the production process and better detect product quality problems. At the same time, in large-scale industrial production processes, using manual vision to check product quality is inefficient and not very accurate. Using machine vision inspection methods can greatly improve production efficiency and the degree of production automation, improve the accuracy of industrial production, speed up work processes and save time.

[0050] However, as the application of visual inspection becomes more and more extensive, the application scenarios are also becoming more and more diverse. Visual inspection is sensitive to lighting, angles and occlusions, and has limited adaptability to complex backgrounds and changing environments, resulting in large changes in the images detected by vision, which are far from the actual objects in the real scene. In addition, the visual inspection system may have data bias or overfitting problems when only performing a single inspection on the inspection image, which affects its accuracy and fault tolerance in actual applications, making it difficult to accurately detect appearance defects.

[0051] Based on this, the embodiments of the present application provide a surface defect detection method, controller and device for a laptop shell, which improve the fault tolerance and stability of the detection equipment and achieve efficient and accurate detection and identification of defect information on the surface of the laptop shell.

[0052] The following is a description of a surface defect detection method, controller and device for a notebook shell provided by an embodiment of the present application in conjunction with the accompanying drawings:

[0053] Reference Figure 1 As shown, Figure 1 : is a flow chart of a method for detecting surface defects of a notebook shell provided by an embodiment of the present application. The method may include but is not limited to the following steps:

[0054] Step S110, photographing six shell surfaces of the notebook shell in sequence by using the photographing component and the light source component;

[0055] Among them, the image of the front side of the notebook shell is captured at different brightness of the light source assembly, the shell structure surface of the notebook shell is captured at the preset brightness of the light source assembly, and the images of the left side, right side, front side and rear side of the notebook shell are captured in different areas at the preset brightness of the light source assembly.

[0056] It should be noted that the front image of the notebook shell is captured by setting the light source component to high, medium and low brightness. The front of the notebook shell can have reflective properties. By setting high, medium and low brightness, various possible shooting lighting conditions are covered, such as daytime, cloudy days and night. In addition, an image is obtained at different brightnesses to avoid model bias caused by training biased towards a certain lighting condition, thereby improving the accuracy of detection and recognition, being able to better cope with working environments under different lighting conditions, and reducing the impact of lighting changes on detection results.

[0057] It should be noted that the structural surface image of the notebook shell is captured under the preset brightness of the light source assembly, which can help capture the details and features of the structural surface of the notebook shell under standard lighting conditions, and is used to analyze and record the appearance of the structural surface, such as the layout of positioning structures and interfaces.

[0058] It should be noted that the four side images of the notebook case are captured in different areas under the preset brightness of the light source component. The front and rear sides are divided into three areas for shooting, and the left and right sides are divided into two areas for shooting. The details and features of each side can be clearly captured, which helps to fully understand the design and structure of the notebook case. Through detailed images, the details and structures of each side can be detected more accurately, which improves the accuracy of detection and recognition.

[0059] Step S120, deploying a corresponding AI detection model on the shell surface corresponding to the image captured by the shooting component;

[0060] It should be noted that, based on the front image of the notebook shell taken by the shooting component, a front AI detection model corresponding to the front image of the notebook shell is deployed, based on the structural surface image of the notebook shell taken by the shooting component, a structural surface AI detection model corresponding to the structural surface image of the notebook shell is deployed, and based on the side image of the notebook shell taken by the shooting component, a side AI detection model corresponding to the side image of the notebook shell is deployed.

[0061] It can be understood that the corresponding AI detection model is trained and optimized for a specific shell surface and its possible defects or features. For example, the front AI detection model is trained and optimized for the front of the notebook shell and its possible defects or features, thereby improving the accuracy and applicability of the detection and ensuring the reliability and effectiveness of the detection results.

[0062] Step S130, performing defect recognition on the image through the AI ​​detection model to obtain an image defect recognition result;

[0063] It should be noted that the trained AI detection model is deployed on the production line to perform real-time detection on the image of the notebook shell, identify the defects in the detected image, and obtain the image defect recognition result corresponding to the image. Each image is passed through the AI ​​detection model corresponding to the image to obtain the image defect recognition result corresponding to the image, which improves the detection speed and efficiency on the production line, is more accurate than manual detection, and reduces labor costs.

[0064] Step S140 , determining the shell surface recognition result according to the image defect recognition results belonging to the same shell surface.

[0065] It should be noted that the number of images acquired on the six surfaces of the notebook shell is different. The same surface includes one or more images, each image corresponds to an image defect recognition result, and one surface corresponds to one or more image defect recognition results. The recognition result of the shell surface is determined based on the image defect recognition result of the same shell surface. By comprehensively considering all defect recognition results on the same shell surface, it helps to reduce missed detections and false detections, such as undetected defects and erroneous identification of normal parts as defects, thereby improving the accuracy of detection.

[0066] Reference Figure 2 As shown, Figure 2 : is a flow chart of a defect identification method provided by an embodiment of the present application, and the method may include but is not limited to the following steps:

[0067] Step S210, comparing the image with the defect template in the defect database using the AI ​​detection model to obtain defect similarity;

[0068] Among them, the defect database is used to store sample templates of various defects.

[0069] It should be noted that the defect template in the defect database is retrieved through the AI ​​detection model, and the image and the defect template are compared to obtain the defect similarity of this image. The defect similarity is used to characterize the similarity between the detection image and the defect template.

[0070] Step S220, determining the image defect recognition result according to the defect similarity and a preset threshold.

[0071] It should be noted that when the similarity between the detection image and the defect template is greater than or equal to a preset threshold, the detection image can be regarded as having the defect of the compared defect template, thereby determining the image defect recognition result of the image.

[0072] It can be understood that by comparing the defect similarity, the type and location of defects in the image can be determined more accurately, reducing the possibility of misjudgment and improving the accuracy of recognition. In addition, automatically determining the image defect recognition results can save human resources and time costs. Determining the recognition results based on defect similarity and preset thresholds can achieve automated quality control and improve production efficiency, especially in large-scale production, where a large number of images can be processed quickly.

[0073] Reference Figure 3 As shown, Figure 3 : is a flow chart of a defect comparison method provided in an embodiment of the present application, and the method may include but is not limited to the following steps:

[0074] Step S310, extracting features from the image using an AI detection model to obtain image features;

[0075] It should be noted that the AI ​​detection model is used to extract features from images and convert them into feature vectors or feature descriptions of the images, such as color histograms, texture features, etc.

[0076] Step S320, comparing the image features with the defect features in the defect template through the AI ​​detection model to obtain defect similarity.

[0077] It should be noted that the images to be compared are collected and preprocessed, including noise removal, image size and color adjustment, etc., to ensure the accuracy and reliability of the comparison. The image features are extracted through the AI ​​detection model, and the obtained image features are matched with the defect templates in the defect database to find defect templates similar to the image features. The similarity between them is calculated by comparing the feature vector of the image to be compared with the feature vector of the defect template. Taking all the matched defect templates into consideration, the final defect similarity can be determined by weighted average or other methods.

[0078] It can be understood that by comparing defect features, defects can be classified according to the similarity of the defect features, such as scratches, dents, cracks, etc., which helps to more accurately identify and handle different types of defects. On the other hand, by comparing defect features, the location of the defect in the image can be determined, which is very important for subsequent processing and repair.

[0079] Reference Figure 4 As shown, Figure 4 : is a flowchart of an AI detection model deployment method provided in an embodiment of the present application. The method may include but is not limited to the following steps:

[0080] Step S410, numbering images according to the shell surface corresponding to the images captured by the shooting component;

[0081] It should be noted that the image is numbered according to the shell surface corresponding to the image captured by the shooting component. For example, the first character of the number is used to indicate the shell surface to which the image belongs, and the last four characters are used to indicate which image of the shell surface the image corresponds to.

[0082] It is understandable that the numbering can clearly indicate the shell surface to which the image belongs, which is helpful for classifying and managing the images in subsequent processing. In addition, the numbering can be used as a unique identifier of the image, which is convenient for retrieving and indexing a large number of images, avoiding confusion caused by mixing images of different shell surfaces, and is conducive to accurately distinguishing and processing images from different sources.

[0083] Step S420: deploy an AI detection model corresponding to the shell surface according to the image number.

[0084] It should be noted that the shell surface corresponding to the image is determined according to the first character in the image number, and the front AI detection model corresponding to the front image of the notebook shell is deployed according to the first character representing the front image, and the structural surface AI detection model corresponding to the structural surface image of the notebook shell is deployed according to the first character representing the structural surface image, and the side AI detection model corresponding to the side image of the notebook shell is deployed according to the first character representing the side image. According to the characteristics of different shell surfaces and possible defects, the corresponding AI detection model can be customized and deployed to improve the accuracy and efficiency of detection, and facilitate the subsequent use of the corresponding shell surface defect identification method.

[0085] Reference Figure 5 As shown, Figure 5 : is a flow chart of a method for determining a shell surface recognition result provided by an embodiment of the present application. The method may include but is not limited to the following steps:

[0086] Step S510, when at least one image defect recognition result belonging to the same shell surface is defective, determining that the shell surface recognition result is defective;

[0087] It should be noted that different shell surfaces have different numbers of corresponding images, but when at least one of the one or more image defect recognition results belonging to the same shell surface is identified as defective, the AI ​​detection model recognizes that the shell surface has defects. For example, the front shell surface includes three images with different brightness. If one of the images with different brightness is identified as defective, the front shell is judged to be defective.

[0088] Step S520: When all image defect recognition results belonging to the same shell surface are defect-free, it is determined that the recognition result of the shell surface is defect-free.

[0089] It can be understood that by unifying the recognition results of the shell surface and simplifying the decision-making process, when the recognition results of multiple images are consistent with no defects, it is more likely to reflect the actual situation, thereby improving the accuracy and credibility of the recognition of shell surface defects. In addition, by combining the recognition results of multiple images to determine the recognition results of the shell surface, the misjudgment rate can be reduced, and the possibility of missed detection and false detection can be reduced.

[0090] Reference Figure 6 As shown, Figure 6 : is a flow chart of a post-detection update summary method provided in an embodiment of the present application, which method may include but is not limited to the following steps:

[0091] Step S610, updating the AI ​​detection model according to the recognition result of the shell surface;

[0092] It should be noted that the AI ​​detection model of the corresponding shell surface is updated according to the recognition result of the shell surface. The AI ​​detection model is retrained using the collected new data and the original data set to update the recognition ability and accuracy of the model. In addition, the images with defective recognition results are updated as defect templates in the defect database to ensure that the AI ​​detection model can be compared with the latest defect template.

[0093] It is understandable that based on the recognition results, the AI ​​detection model is continuously optimized and updated to improve its performance and stability, ensuring that the system can continue to provide accurate and reliable detection services.

[0094] Step S620 , summarizing and outputting the recognition results of the six shell surfaces of the notebook shell.

[0095] It should be noted that the recognition results of the four shell sides are displayed surrounding the recognition results of the shell front and the shell structure surface, wherein the four shell sides, the front side, the right side, the rear side and the left side are arranged in a counterclockwise direction from bottom to top and then to bottom, showing the effect of the notebook shell structure, and better viewing the recognition results of the six shell surfaces of the notebook shell.

[0096] Reference Figure 7 As shown, Figure 7 : is a flowchart of a re-judgment update method provided in an embodiment of the present application, and the method may include but is not limited to the following steps:

[0097] Step S710, obtaining a manual re-judgment result with the recognition result of six shell surfaces of the notebook shell as a reference;

[0098] Step S720, updating the AI ​​detection model according to the manual review result.

[0099] It should be noted that the recognition results of the six shell surfaces of the notebook shell are obtained and summarized, and the recognition results of the six shell surfaces are displayed on the equipment of the re-inspection personnel. The re-inspection personnel re-inspect the notebook shell with reference to the recognition results of the six shell surfaces, and judge the recognition results of the six shell surfaces separately, and summarize them into manual re-judgment results. For example, the AI ​​detection model recognizes that the front of the shell has no defects, but the manual re-judgment result is that there are defects. It is determined that the AI ​​detection model recognizes that the front of the shell has an error. The AI ​​detection model is updated according to the manual re-judgment result, and a new defect template is added to the defect database, so that more appearance defects can be identified in subsequent use.

[0100] Reference Figure 8 As shown, Figure 8 It is a structural schematic diagram of a controller provided in an embodiment of the present application.

[0101] The controller 800 includes:

[0102] The processor 810 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant instructions to implement the technical solutions provided in the embodiments of the present application;

[0103] The memory 820 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 820 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 820, and the processor 810 calls and executes the technical solutions of the embodiments of this application;

[0104] Input / output interface 830, used to implement information input and output;

[0105] The communication interface 840 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0106] bus 850 , which transmits information between the various components of the device (e.g., processor 810 , memory 820 , input / output interface 830 , and communication interface 840 );

[0107] The processor 810 , the memory 820 , the input / output interface 830 and the communication interface 840 are connected to each other in communication within the device via a bus 850 .

[0108] Reference Fig. 9 As shown, Fig. 9 It is a structural schematic diagram of a surface defect detection device provided in an embodiment of the present application.

[0109] The surface defect detection device 900 includes a light source component 910 , a photographing component 920 and a controller 800 .

[0110] The embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, the computer executes the method described above.

[0111] Reference Fig.10 As shown, Fig.10 This is a schematic diagram of the overall process of a surface defect detection method for a notebook shell provided by an embodiment of the present application. The method may include but is not limited to the following steps:

[0112] Step S1001, photographing six shell surfaces of the notebook shell in sequence by using a photographing component and a light source component;

[0113] Step S1002, numbering images according to the shell surface corresponding to the images captured by the shooting component;

[0114] Step S1003, deploying an AI detection model corresponding to the shell surface according to the image number;

[0115] Step S1004, extracting features from the image using an AI detection model to obtain image features;

[0116] Step S1005, comparing the image features with the defect features in the defect template through the AI ​​detection model to obtain defect similarity;

[0117] Step S1006, determining the image defect recognition result according to the defect similarity and a preset threshold;

[0118] Step S1007, when at least one image defect recognition result belonging to the same shell surface is defective, determining that the shell surface recognition result is defective;

[0119] Step S1008, when all image defect recognition results belonging to the same shell surface are all defect-free, determining that the shell surface recognition result is defect-free;

[0120] Step S1009, updating the AI ​​detection model according to the recognition result of the shell surface;

[0121] Step S1010 , summarizing and outputting the recognition results of the six shell surfaces of the notebook shell.

[0122] In one embodiment of the present application, one of the six shell surfaces of the notebook shell is photographed by a shooting component and a light source component, such as the front image of the notebook shell is photographed and obtained by setting the light source component to high, medium and low brightness; the structural surface image of the notebook shell is photographed at the preset brightness of the light source component; the four side images of the notebook shell are all photographed in different areas at the preset brightness of the light source component, and the front side and the rear side are divided into 3 areas for photographing respectively by the shooting component, and the left side and the right side are divided into 2 areas for photographing. One or more images of the shell surface are obtained, and the images are numbered according to the shell surface, such as the first character in the number is used to indicate the shell surface to which the image belongs, and the last 4 are digits used to indicate the image of the corresponding shell surface. The shell surface corresponding to the image is determined according to the first character in the image number, and the AI ​​detection model corresponding to the shell surface is deployed. The image of the shell surface is feature extracted by the corresponding AI detection model, and the image is converted into a feature vector or feature description of the image. The obtained image features are matched with the defect template in the defect database, and a defect template similar to the image features is found. The similarity between them is calculated by comparing the feature vector of the image to be compared with the feature vector of the defect template. When the similarity between the detected image and the defect template is greater than or equal to the preset threshold, the detected image can be regarded as having the defect of the compared defect template, thereby determining the image defect recognition result of the image. When at least one of the one or more image defect recognition results belonging to the same shell surface is identified as having a defect, the AI ​​detection model recognizes the shell surface as having a defect; when all image defect recognition results belonging to the same shell surface are not defective, the recognition result of the shell surface is determined to be non-defective. After the recognition result of the shell surface is obtained, the AI ​​detection model corresponding to the shell surface is updated according to the recognition result of the shell surface, and the next shell surface is photographed to obtain an image, the corresponding AI detection model is deployed, the defect is identified, and the recognition result of the shell surface is obtained and the corresponding AI detection model is updated until the six shell surfaces of the notebook shell are detected. The recognition results of the four shell sides are displayed surrounding the recognition results of the shell front and the shell structure surface, wherein the four shell sides, the front side, the right side, the rear side and the left side are arranged in a counterclockwise direction from bottom to top and then to bottom, showing the effect of the notebook shell structure and better viewing the recognition results of the six shell surfaces of the notebook shell.

[0123] In an embodiment of the present application, six shell surfaces of the notebook shell are photographed in sequence by a shooting component and a light source component, and a corresponding AI detection model is deployed according to the shell surface corresponding to the image photographed by the shooting component, and defects are identified on the image by the AI ​​detection model to obtain an image defect recognition result, and the recognition result of the shell surface is determined according to the image defect recognition result belonging to the same shell surface; therefore, the embodiment of the present application obtains an image of the front of the notebook shell at different brightnesses, an image of the structural surface of the notebook shell at a preset brightness, and partial area images of the left side, right side, front side and rear side of the notebook shell at a preset brightness through a shooting component and a light source component. By acquiring images of the six surfaces, more comprehensive information of the shell surface of the notebook shell is obtained, and a corresponding AI detection model is deployed according to the shell surface corresponding to the image. The image is detected by the AI ​​detection model corresponding to the shell surface to prevent the AI ​​detection model from misrecognizing in the process of recognizing the image, and the recognition result of the shell surface is determined according to the image defect recognition result belonging to the same shell surface, thereby improving the fault tolerance and stability of the detection equipment, and realizing efficient and accurate detection and identification of defect information on the surface of the notebook shell.

[0124] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0125] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0126] It should also be understood that the various implementations provided in the embodiments of the present application can be combined arbitrarily to achieve different technical effects.

[0127] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions without violating the spirit of the present application. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A method for detecting surface defects of a notebook shell, It is characterized in that Applied to a surface defect detection device, the surface defect detection device comprises a light source component and a shooting component; The surface defect detection method comprises: The six shell surfaces of the notebook shell are photographed in sequence by the photographing component and the light source component; Deploy a corresponding AI detection model on the shell surface corresponding to the image captured by the shooting component; Performing defect recognition on the image by using the AI ​​detection model to obtain an image defect recognition result; Determining the recognition result of the shell surface according to the image defect recognition result belonging to the same shell surface; Among them, the image of the front side of the notebook shell is photographed at different brightnesses of the light source assembly, the shell structure surface of the notebook shell is photographed at the preset brightness of the light source assembly, and the images of the left side, right side, front side and rear side of the notebook shell are all photographed in different areas at the preset brightness of the light source assembly.

2. The surface defect detection method according to claim 1, It is characterized in that The performing defect recognition on the image by using the AI ​​detection model to obtain an image defect recognition result includes: Compare the image with the defect template in the defect database by using the AI ​​detection model to obtain defect similarity; Determining the image defect recognition result according to the defect similarity and a preset threshold; The defect database is used to store sample templates of various defects.

3. The surface defect detection method according to claim 2, It is characterized in that The defect comparison of the image by using the AI ​​detection model and the defect template in the defect database to obtain the defect similarity includes: Extract features from the image using the AI ​​detection model to obtain image features; The defect similarity is obtained by comparing the image features with the defect features in the defect template through the AI ​​detection model.

4. The surface defect detection method according to claim 1, It is characterized in that The deploying a corresponding AI detection model on the shell surface corresponding to the image captured by the shooting component includes: Numbering the images according to the shell surface corresponding to the images captured by the shooting component; An AI detection model corresponding to the shell surface is deployed according to the serial number of the image.

5. The surface defect detection method according to claim 1, It is characterized in that The determining the recognition result of the shell surface according to the image defect recognition result belonging to the same shell surface includes: When at least one of the image defect recognition results belonging to the same shell surface is defective, determining that the recognition result of the shell surface is defective; When all the image defect recognition results belonging to the same shell surface are defect-free, it is determined that the recognition result of the shell surface is defect-free.

6. The surface defect detection method according to claim 1, It is characterized in that After determining the recognition result of the shell surface according to the image defect recognition result belonging to the same shell surface, the surface defect detection method further includes: updating the AI ​​detection model according to the recognition result of the shell surface; The recognition results of the six shell surfaces of the notebook shell are summarized and outputted.

7. The surface defect detection method according to claim 6, It is characterized in that After summarizing and outputting the recognition results of the six shell surfaces of the notebook shell, the surface defect detection method further includes: Obtaining manual re-judgment results with the recognition results of the six shell surfaces of the notebook shell as a comparison; The AI ​​detection model is updated according to the manual re-judgment result.

8. A controller, It is characterized in that The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the surface defect detection method according to any one of claims 1 to 7 when executing the computer program.

9. A surface defect detection device, It is characterized in that It comprises a light source assembly, a shooting assembly and the controller as claimed in claim 8.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Quality detection method device and system, and electronic equipment

    CN110261392A

  • Defect detection method, system, and apparatus

    CN110726724A

  • AI visual detection method and system based on deep learning, and readable storage medium

    CN114705685A

  • Wafer defect detection method and system

    CN116385398A

  • Detection method, device, system and equipment of hexahedral detected piece and storage medium

    CN116642892A

Cited By

  • Notebook computer shell surface defect detection method and device and storage medium

    CN121577536A