Automatic checking method for appearance inspection
By introducing an automatic appearance inspection verification method in the optical detection system, the limit sample pictures are automatically retrieved for verification, and the traditional limit sample variation and preservation problems are solved, achieving consistency and efficient production of the detection system.
Patent Information
- Application Number
- CN202510057063.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-14
- Publication Date
- 2025-04-25
AI Technical Summary
In traditional optical detection systems, the variation and preservation of samples of limit samples lead to limited verification frequency, limited verification timeliness and effectiveness, and manual operations increase personnel workload and machine production capacity loss.
The automatic calibration method of appearance inspection is adopted. The limit sample pictures are automatically retrieved for verification during each production. By setting detection parameters and system verification standards, the inspection results are automatically compared to ensure that the clarity and grayscale of the product pictures meet the requirements.
The consistency of the limit samples is achieved, the grayscale and clarity verification is increased, the consistency of the detection system is ensured, the quality risks of limit samples failure and personnel missed operations are avoided, and the workload of personnel and machine downtime is reduced.
Smart Images
Figure CN119880925A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of integrated circuit manufacturing, and in particular relates to an automatic verification method for appearance inspection. Background Art
[0002] The limit sample is also called the golden sample. The quality characteristics of some products cannot be described in words, data or other ways, such as gloss, roughness, burrs, and scratches on the product. In order to avoid cognitive differences among different personnel, it is necessary to define a physical limit sample as a basis for judging whether the product quality is qualified or not. The limit sample is used to specifically determine the minimum requirement standard for a quality characteristic that cannot be quantified (or difficult to quantify). After comparing the quality of the delivered product with the limit sample, it can be accepted within this standard; otherwise, it will be rejected if it exceeds the limit standard.
[0003] The detection capability of the optical detection system is very sensitive to mechanical variation, photoelectric system performance degradation, and parameter setting changes. The traditional approach is to use physical limit samples for machine verification before mass production. This is affected by the variation of physical limit samples, the limited verification frequency due to the complexity of physical verification, the limited effectiveness of verification timeliness, and the loss of production capacity and increased workload. Traditional limit samples need to be prepared manually and stored separately. Each time production is carried out, the corresponding limit samples need to be found for actual detection and verification. Limit samples are often damaged during storage, causing other defects and failure.
[0004] The existing optical inspection system calibration method: using limit samples, on-machine verification, and comparing the consistency of the test results with the actual condition of the product; there are the following problems: 1. The limit samples gradually mutate as the storage time continues, resulting in calibration failure; 2. It is impossible to quantify and evaluate the changes in the grayscale and clarity of the captured images; 3. The physical space required for sample storage is large, and it takes a long time for personnel to operate the system using physical samples, resulting in large manpower consumption and machine production capacity losses; 4. Frequent operation of limit samples aggravates the process of limit sample failure, and manual execution poses the risk of missed operations. Summary of the invention
[0005] The purpose of the present invention is to provide an automatic verification method for appearance inspection, which automatically retrieves limit sample images for verification each time production is carried out, can ensure the consistency of limit samples, and at the same time increase grayscale and clarity verification to ensure the consistency of the detection system. Automatic verification replaces manual limit sample verification, avoids the quality risks of limit sample failure and personnel missing operations, reduces personnel workload, and reduces the downtime of machines due to verification.
[0006] The present invention provides an automatic verification method for appearance inspection, comprising:
[0007] S10, establishing standards: taking photos of the physical limit samples to obtain limit sample images, and establishing standards for the limit samples, wherein the standards include: set detection parameters, the limit sample images, and system calibration standards; the system calibration standards include: the detection results and allowable deviation range of the limit sample images with respect to the set detection parameters;
[0008] S20, perform system verification: switch the production program to prepare for mass production of the product corresponding to the limit sample, the machine retrieves the system verification standard, the set detection parameters and the limit sample picture; uses the set detection parameters to detect the limit sample picture, and the optical detection system automatically compares the obtained detection result with the system verification standard; if the comparison difference meets the requirements, take a picture of the product, obtain the clarity and grayscale of the product picture and compare it with the clarity and grayscale of the limit sample picture. If the clarity and grayscale of the product picture meet the requirements, the machine enters the production process and starts mass production of the product.
[0009] Furthermore, S10, establishing the standard specifically includes: taking out the limit sample of the physical object from the storage position, loading and transferring the limit sample to the optical detection position, and taking a picture of the limit sample of the physical object using a camera of the optical detection system.
[0010] Further, S10, establishing standards specifically includes: debugging detection parameters and setting detection specifications through the limit samples; the detection specifications include: detection items, detection conditions and detection specifications;
[0011] The optical detection system uses a preset optical algorithm to detect the limit sample image according to the detection parameters and the detection specifications to obtain a first detection result;
[0012] Sending the limit sample of the physical object to a tool microscope for inspection, performing inspection according to the inspection parameters and the inspection specifications, and obtaining a second inspection result;
[0013] Comparing the first test result with the second test result, if the difference between the two test results is within a threshold range, the test parameter is used as the set test parameter; and the test specification is used as the final test specification corresponding to the limit sample;
[0014] If the difference between the two test results exceeds the threshold range, the test parameters and / or relevant contents in the test specifications are adjusted until the difference between the two test results is within the threshold range; and the adjusted test parameters are used as the set test parameters; the adjusted test specifications are used as the final test specifications corresponding to the limit sample.
[0015] Furthermore, the detection parameters include: the detection area of the limit sample; the grayscale, background base grayscale and threshold of the limit sample in the limit sample image; the shape, template selection and feature matching of the key pattern in the limit sample image; the target object pattern quality and pixel contour matching of the limit sample.
[0016] Furthermore, if at least one of the clarity and grayscale of the product image does not meet the requirements, the limit sample image and / or the set detection parameters are adjusted.
[0017] Furthermore, if the optical detection system automatically determines that the comparison difference does not meet the requirements, the production is reported for repair and the hardware and / or parameter adjustments are performed.
[0018] Furthermore, each of the products automatically learns the grayscale value of a fixed area on the chip surface and the clarity of a characteristic area when it is first produced;
[0019] The surface grayscale of the fixed area of the first N chips is automatically checked in each subsequent production to ensure that the brightness of the light source used in the production of each batch of materials is consistent;
[0020] The surface clarity of the fixed area of the first N chips is automatically checked during the production of each batch of materials to ensure that the camera focal length is correct and the capture clarity is consistent during the production of each batch of materials.
[0021] Furthermore, before the production of each batch of the products begins and after the machine parameters are debugged, the limit sample image is automatically retrieved for automatic verification to confirm whether the verification result is consistent with the limit sample defect.
[0022] Furthermore, the automatic verification includes: sampling a fixed area and calibrating the grayscale of the screenshot; sampling a fixed area and judging the image clarity; retrieving the digital archive of the screenshot of the limit sample image when the production file is established, and performing calculations on the limit sample image according to the set detection parameters to obtain the inspection result; confirming whether the inspection result is consistent with the limit sample defect.
[0023] Furthermore, the limit sample image is stored in a designated document of the detection computer.
[0024] Compared with the prior art, the present invention has the following beneficial effects:
[0025] The present invention provides an automatic verification method for appearance inspection, comprising: S10, establishing standards: taking photos of limit samples of a physical object to obtain limit sample images, establishing standards for the limit samples, the standards comprising: setting detection parameters, the limit sample images and system verification standards; the system verification standards comprising: detection results and allowable deviation ranges of the limit sample images with respect to the set detection parameters; S20, performing system verification: switching the production program to prepare for mass production of products corresponding to the limit samples, a machine retrieves the system verification standards, the set detection parameters and the limit sample images; the limit sample images are detected using the set detection parameters, and an optical detection system automatically compares the obtained detection results with the system verification standards; if the comparison difference meets the requirements, photographing the product, obtaining the clarity and grayscale of the product image and comparing them with the clarity and grayscale of the limit sample image; if the clarity and grayscale of the product image meet the requirements, the machine enters the production process and starts mass production of the product.
[0026] The automatic appearance inspection method of the present invention can automatically retrieve the limit sample image for verification each time production is carried out, which can ensure the consistency of the limit sample, and at the same time increase the grayscale and clarity verification to ensure the consistency of the detection system. Automatic verification replaces manual limit sample verification, avoids the quality risks of limit sample failure and personnel missing operations, reduces the workload of personnel, and reduces the downtime of the machine due to verification. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 The figure is a flow chart of an automatic verification method for appearance inspection according to an embodiment of the present invention.
[0028] Figure 2 The figure is a schematic diagram of the standard establishment process in the automatic verification method for appearance inspection according to an embodiment of the present invention.
[0029] Figure 3 The figure is a schematic diagram of the system verification process in the automatic verification method for appearance inspection according to an embodiment of the present invention. DETAILED DESCRIPTION
[0030] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description. It should be noted that the accompanying drawings are in a very simplified form and use an inaccurate scale, which is only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.
[0031] For ease of description, some embodiments of the present application may use spatially relative terms such as "above", "below", "top", "below", etc. to describe the relationship between one element or component and another (or other) elements or components as shown in the various figures of the embodiments. It should be understood that in addition to the orientations described in the drawings, the spatially relative terms are also intended to include different orientations of the device in use or operation. For example, if the device in the drawings is turned over, the elements or components described as being "below" or "below" other elements or components will subsequently be positioned as being "above" or "above" other elements or components. The terms "first", "second", etc. below are used to distinguish between similar elements and are not necessarily used to describe a specific order or time sequence. It is to be understood that these terms used in this way are interchangeable where appropriate.
[0032] The embodiment of the present invention provides an automatic verification method for appearance inspection, such as Figure 1 As shown, including:
[0033] S10, establishing standards: taking photos of the physical limit samples to obtain limit sample images, and establishing standards for the limit samples, the standards include: setting detection parameters, limit sample images and system calibration standards; the system calibration standards include: the detection results of the limit sample images with respect to the set detection parameters and the allowable deviation range;
[0034] S20, perform system calibration: switch the production program to prepare for mass production of the product corresponding to the limit sample, the machine retrieves the system calibration standard, sets the detection parameters and the limit sample image; uses the set detection parameters to detect the limit sample image, and the optical detection system automatically compares the obtained detection result with the system calibration standard; if the comparison difference meets the requirements, take a picture of the product, obtain the clarity and grayscale of the product image and compare it with the clarity and grayscale of the limit sample image; if the clarity and grayscale of the product image meet the requirements, the machine enters the production process and starts mass production of the product.
[0035] Combine the following Figure 2 and Figure 3 The steps of the automatic verification method for appearance inspection according to an embodiment of the present invention are introduced in detail.
[0036] Step S10: Establish standards, such as Figure 2 As shown, the physical limit sample (gold sample) is taken out from the storage position, and the limit sample is manually loaded and transferred to the optical detection position. Through the limit sample, the detection parameters are debugged and the detection specifications are set; the detection specifications include: detection items, detection conditions and detection specifications. The detection parameters include: the detection area of the limit sample; the grayscale of the limit sample in the limit sample image, the background basic grayscale and threshold; the shape, template selection and feature matching of the key pattern in the limit sample image; the target object pattern quality and pixel contour matching of the limit sample, etc.
[0037] The camera of the optical detection system is used to shoot the limit sample of the physical object to obtain the limit sample picture. The optical detection system uses a preset optical algorithm to detect the limit sample picture according to the detection parameters and the preset detection specifications to obtain the first detection result; the physical limit sample is sent to the tool microscope for inspection, and is detected according to the detection parameters and the preset detection specifications to obtain the second detection result; the first detection result is compared with the second detection result. If the difference between the two detection results is within the threshold range, the detection parameter is used as the set detection parameter F; the detection specification is used as the final detection specification corresponding to the limit sample. If the difference between the two detection results exceeds the threshold range, the detection parameters and / or the relevant content in the detection specification are adjusted until the difference between the two detection results is within the threshold range; and the adjusted detection parameters are used as the set detection parameters F; the adjusted detection specifications are used as the final detection specifications corresponding to the limit sample.
[0038] Establish the standard of limit sample, which includes: setting detection parameter F, limit sample picture (snapshot) P and system verification standard. Store the limit sample picture P in the specified path and mark it with the name of the production file; record the clarity C and grayscale G of the limit sample picture. The system verification standard includes: the detection result R and the allowable deviation range T of the limit sample picture regarding the set detection parameters. The detection result R is specifically the first detection result mentioned above. The system verification standard is stored in the specified path file and can be retrieved by the machine.
[0039] Step S20: perform system verification, such as Figure 3 As shown, the production program is switched to prepare for mass production of the product corresponding to the limit sample. The machine automatically retrieves the system verification standard, retrieves the set detection parameter F, and automatically retrieves the limit sample picture P; the set detection parameter F is used to detect the limit sample picture P to avoid detection failure caused by inappropriate revision of the detection parameter. The optical detection system automatically compares the obtained detection results with the system verification standard; the optical detection system automatically determines whether the difference meets the requirements. If the compared difference meets the requirements, the camera of the optical detection system shoots the product to be mass-produced, obtains the clarity and grayscale of the product picture and compares it with the clarity C and grayscale G of the evaluation standard limit sample picture. If the clarity and grayscale of the product picture meet the requirements, the machine enters the production process and starts mass production of the product. If at least one of the clarity and grayscale of the product picture to be mass-produced does not meet the requirements, the hardware and / or parameter adjustments are performed, specifically adjusting the limit sample picture P and / or setting the detection parameter F. If the optical detection system automatically determines that the compared difference does not meet the requirements, the production is repaired, and the hardware and / or parameter adjustments are performed, specifically adjusting the limit sample picture P and / or setting the detection parameter F.
[0040] The camera of the optical inspection system takes pictures of the products to be mass-produced, obtains the clarity and grayscale of the product pictures, and produces direct and quantitative verification results to ensure the effectiveness of the verification. Based on the quantitative results, the camera quality variation caused by hardware and setting variations can be effectively determined.
[0041] When each product is produced for the first time, the grayscale value of the fixed area on the chip surface and the clarity of the feature area are automatically learned; in each subsequent production, the surface grayscale of the fixed area of the first N chips is automatically verified to ensure that the brightness of the light source used in the production of each batch of materials is consistent; when each batch of materials is produced, the surface clarity of the fixed area of the first N chips is automatically verified to ensure that the camera focal length is correct and the capture clarity is consistent when each batch of materials is produced.
[0042] Working principle of optical inspection system: The camera takes a picture of the object to be inspected, and obtains a picture whose grayscale and clarity meet the inspection requirements. The optical algorithm system performs calculations according to the set parameters, and outputs the inspection results according to the defined type to classify the object to be inspected.
[0043] Before the improvement, each calibration required manual loading of the physical limit samples and transporting them to the optical inspection position, where the camera would take photos of the physical limit samples. Physical limit samples require special sites and personnel to manage, and there are many types and multiple physical limit samples, which is laborious and time-consuming to manage. The defect types, results, and allowable errors of the limit samples need to be managed. Moreover, the physical limit samples are subject to damage over time and use, so the limit sample photos taken by the camera each time change with the changes in the physical limit samples, and the consistency of the limit sample photos cannot be guaranteed.
[0044] The automatic verification method for appearance inspection of the present invention can automatically retrieve limit sample images for verification each time production is carried out. Each time, limit sample images are retrieved, and limit sample images are stored in a designated document of a detection computer, which is easier to manage and save than physical objects. The present invention only uses physical limit samples when establishing the standard for the first time, and calls limit sample images for each subsequent verification. No human operation is required, and the consistency of limit sample images can be guaranteed. At the same time, grayscale and clarity verification are increased to ensure the consistency of the detection system. Fixed limit sample images are used to avoid verification failure due to physical limit sample failure. Fixed limit sample images are used to eliminate changes in the detection results caused by variations in each photo, and it can be ensured that the changes in the detection results come from changes in parameter settings. The machine automatically compares to determine whether the current detection result is consistent with the allowed detection result. The standard is stored in a special document of the detection system to establish a closed loop of the detection system itself. The system is automatically executed without manpower consumption. The limit samples are stored near the production site for a long time, and the verification is integrated into the production operation process, which takes less than 1 minute; each batch uses a fixed limit sample image machine to automatically perform verification, without the risk of failure and missed operation.
[0045] The automatic verification system of the present invention automatically retrieves the limit sample image for automatic verification before the start of production of each batch of products and after the machine parameters are debugged to confirm whether the verification result is consistent with the limit sample defect. The automatic verification includes: sampling a fixed area and calibrating the grayscale of the captured image; sampling a fixed area and judging the image clarity; retrieving the digital archive of the captured image of the archived limit sample image when the production file is established, and performing calculations and processing on the limit sample image according to the set detection parameters to obtain the inspection result; confirming whether the inspection result is consistent with the limit sample defect. The present invention eliminates the influence of limit sample variation, reduces storage space and management costs; automatically performs verification, avoids missed operations, saves labor costs, shortens execution cycles, and reduces production capacity losses.
[0046] The present invention can automatically verify and monitor the indicators of the optical detection system, and effectively solve the problems of detection failure caused by mechanical variation of the optical detection system, performance attenuation of the photoelectric system, inappropriate revision of detection parameters, and detection failure caused by factors such as human error operation.
[0047] The present invention establishes an online inspection system for light source brightness and automatic focus inspection for captured images. When the product is changed, the effectiveness of the light source and the quality of the camera are automatically inspected to ensure the quality of the captured images. A standard defect image library and specifications are established. When the product is switched, the production file is retrieved to inspect the standard defect image. When the inspection result is inconsistent with the specification, an alarm is issued and the machine is shut down to ensure the effectiveness of the production file. The automatic verification system replaces manual limit sample verification to avoid the quality risks of limit sample failure and personnel omission, reduce the workload of personnel, and reduce the downtime of the machine due to verification.
[0048] In summary, the present invention provides an automatic verification method for appearance inspection, including: S10, establishing standards: taking photos of limit samples of the actual object to obtain limit sample pictures, establishing standards for the limit samples, the standards including: setting detection parameters, the limit sample pictures and system verification standards; the system verification standards including: the detection results and allowable deviation ranges of the limit sample pictures regarding the set detection parameters; S20, performing system verification: switching the production program to prepare for mass production of the product corresponding to the limit sample, the machine retrieves the system verification standards, the set detection parameters and the limit sample pictures; the limit sample pictures are detected using the set detection parameters, and the optical detection system automatically compares the obtained detection results with the system verification standards; if the comparison difference meets the requirements, photograph the product, obtain the clarity and grayscale of the product picture and compare them with the clarity and grayscale of the limit sample picture, if the clarity and grayscale of the product picture meet the requirements, the machine enters the production process and starts mass production of the product.
[0049] The automatic appearance inspection method of the present invention can automatically retrieve the limit sample image for verification each time production is carried out, which can ensure the consistency of the limit sample, and at the same time increase the grayscale and clarity verification to ensure the consistency of the detection system. Automatic verification replaces manual limit sample verification, avoids the quality risks of limit sample failure and personnel missing operations, reduces the workload of personnel, and reduces the downtime of the machine due to verification.
[0050] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the method disclosed in the embodiment, since it corresponds to the device disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.
[0051] The above description is only a description of the preferred embodiment of the present invention, and is not any limitation on the scope of rights of the present invention. Any technical personnel in this field can make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.
Claims
1. An automatic verification method for appearance inspection, characterized in that: include: S10, establishing standards: taking photos of the physical limit samples to obtain limit sample images, and establishing standards for the limit samples, wherein the standards include: set detection parameters, the limit sample images, and system calibration standards; the system calibration standards include: the detection results and allowable deviation range of the limit sample images with respect to the set detection parameters; S20, perform system verification: switch the production program to prepare for mass production of the product corresponding to the limit sample, the machine retrieves the system verification standard, the set detection parameters and the limit sample picture; uses the set detection parameters to detect the limit sample picture, and the optical detection system automatically compares the obtained detection result with the system verification standard; if the comparison difference meets the requirements, take a picture of the product, obtain the clarity and grayscale of the product picture and compare it with the clarity and grayscale of the limit sample picture. If the clarity and grayscale of the product picture meet the requirements, the machine enters the production process and starts mass production of the product.
2. The automatic appearance inspection method according to claim 1, characterized in that: S10, establishing the standard specifically includes: taking out the limit sample of the physical object from the storage position, loading and transferring the limit sample to the optical detection position, and taking a picture of the limit sample of the physical object using a camera of the optical detection system.
3. The automatic appearance inspection method according to claim 1, characterized in that: S10, establishing standards specifically includes: debugging detection parameters and setting detection specifications through the limit samples; the detection specifications include: detection items, detection conditions and detection specifications; The optical detection system uses a preset optical algorithm to detect the limit sample image according to the detection parameters and the detection specifications to obtain a first detection result; Sending the limit sample of the physical object to a tool microscope for inspection, performing inspection according to the inspection parameters and the inspection specifications, and obtaining a second inspection result; Comparing the first test result with the second test result, if the difference between the two test results is within a threshold range, the test parameter is used as the set test parameter; and the test specification is used as the final test specification corresponding to the limit sample; If the difference between the two test results exceeds the threshold range, the test parameters and / or relevant contents in the test specifications are adjusted until the difference between the two test results is within the threshold range; and the adjusted test parameters are used as the set test parameters; the adjusted test specifications are used as the final test specifications corresponding to the limit sample.
4. The automatic appearance inspection method according to claim 3, characterized in that: The detection parameters include: the detection area of the limit sample; the grayscale, background base grayscale and threshold of the limit sample in the limit sample image; the shape, template selection and feature matching of the key pattern in the limit sample image; the target object pattern quality and pixel contour matching of the limit sample.
5. The automatic appearance inspection method according to claim 1, characterized in that: If at least one of the clarity and grayscale of the product image does not meet the requirements, adjust the limit sample image and / or the set detection parameters.
6. The automatic verification method for appearance inspection according to claim 1, characterized in that: If the optical detection system automatically determines that the comparison difference does not meet the requirements, the production will report it for repair and make hardware and / or parameter adjustments.
7. The automatic appearance inspection method according to claim 1, characterized in that: Also includes: Each of the products automatically learns the grayscale value of a fixed area on the chip surface and the clarity of the characteristic area when it is first produced; The surface grayscale of the fixed area of the first N chips is automatically checked in each subsequent production to ensure that the brightness of the light source used in the production of each batch of materials is consistent; The surface clarity of the fixed area of the first N chips is automatically checked during the production of each batch of materials to ensure that the camera focal length is correct and the capture clarity is consistent during the production of each batch of materials.
8. The automatic appearance inspection method according to claim 1, characterized in that: Also includes: Before the production of each batch of the products begins and after the machine parameters are debugged, the limit sample image is automatically retrieved for automatic verification to confirm whether the verification result is consistent with the limit sample defects.
9. The automatic appearance inspection method according to claim 8, characterized in that: Also includes: The automatic verification includes: sampling a fixed area and calibrating the grayscale of the screenshot; sampling a fixed area and judging the image clarity; retrieving the archived digital screenshot of the limit sample image when the production file is established, performing calculations on the limit sample image according to the set detection parameters to obtain the inspection result; and confirming whether the inspection result is consistent with the limit sample defect.
10. The automatic appearance inspection method according to claim 1, characterized in that: The limit sample image is stored in a designated file on the detection computer.