AOI visual online high-speed detection method and system

Abstract

The invention provides an AOI visual online high-speed detection method and system, and the method comprises the steps: obtaining the type information of products on a production line, confirming the features, needing to be detected, of the products, setting the features needing to be detected as target features, and obtaining a plurality of target features; setting the frame rate and the resolution of the camera according to the actual size of the target feature and the required definition; the method comprises the following steps: acquiring surface material information in pre-acquired product type information, selecting a light source according to the surface material information, and setting an irradiation angle and intensity of the light source, so that the light source irradiated on the surface of an image cannot be reflected to cause the quality problem of the image; and obtaining photos of a plurality of target features of the products, comparing the plurality of target features according to the standard parameters, and dividing the products into qualified products and unqualified products according to a comparison result. The technical problems of influence of ambient light on image quality, light source adaptation caused by reflection specificity of various materials and the like in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of AOI detection technology, and in particular to an AOI visual online high-speed detection method and system. Background Art

[0002] AOI technology is widely used in quality control in the electronics, automotive, printing and other industries. It relies on high-speed, high-precision image capture and processing capabilities to automatically detect product appearance, shape, size and color differences. Modern AOI systems usually combine computer vision and machine learning technologies to improve the accuracy and efficiency of detection. However, it is difficult to balance speed and accuracy in high-speed production environments, the image quality is unstable due to ambient light, and multiple materials cannot be effectively processed by light sources due to differences in reflective properties. The limited color difference detection capability affects the visual quality of the product, as well as deficiencies in complex geometric structure recognition and system integration. These problems affect the overall efficiency and automation of the AOI system. Summary of the invention

[0003] Based on the above technical problems, the present invention proposes an AOI visual online high-speed detection method and system, and the technical solutions adopted are as follows: An AOI visual online high-speed detection method, the method comprising: S1: Acquire product type information on the production line, confirm features of the products that need to be detected, and set the features that need to be detected as target features to obtain multiple target features; S2: Set the frame rate and resolution of the camera according to the actual size of the target feature and the required clarity; S3: obtaining surface material information from the pre-acquired product category information, selecting a light source according to the surface material information, and setting the irradiation angle and intensity of the light source so that the light source is not reflected on the image surface and thus causes image quality problems; S4: Acquire photos of multiple target features of the product, compare the multiple target features according to standard parameters, and classify the products into qualified products and unqualified products according to the comparison results.

[0004] Preferably, the S1 includes: S11: Scanning identification information of products on the production line through a scanning device, matching the collected identification information with a central database, and obtaining product type information; S12: According to the requirements for feature detection of products in the central database, the features to be detected are set as target features, wherein the target features include but are not limited to shape and size, surface defects, color and structural features, multiple target features are obtained, and an image acquisition sequence is set for each target feature.

[0005] Preferably, the S2 includes: S21: Obtain the actual size of each target feature, where the actual size represents the length of the target feature, and set the frame rate of the camera according to the minimum actual size of the multiple target features, and set the frame rate of the camera according to the following formula: , where D represents the moving distance between frames, V represents the speed of the production line, that is, the moving distance of the product in the production line per unit time, and F represents the frame rate of the camera, that is, the number of photos taken by the camera per unit time. When the camera frame rate satisfies: D≤actual size, the camera frame rate setting is completed at this time; S22: According to the required clarity of each target feature, the corresponding resolution of the camera at each clarity is set. When the current target feature image is acquired, the camera switches to the resolution required for the next target feature according to the preset image acquisition sequence.

[0006] Preferably, S3 includes: S31: Acquire surface material information of the product according to the product type information, wherein the surface material information includes but is not limited to plastic, metal, ceramic and glass and the corresponding reflective properties and surface smoothness; S32: Select a light source according to the surface material information, the light source includes but is not limited to LED, laser and infrared light; set the irradiation angle and intensity of the light source according to the reflection angle of the surface material, and arrange the light source in a ring shape.

[0007] Preferably, the S4 includes: S41: Obtain a shape and size feature photo of the product, obtain the contour information and size value of the product according to the shape and size feature photo, and compare them with the contour template and the tolerance range in the standard parameter. If the contour information is consistent with the contour template and the size value is within the tolerance range, the shape and size of the product are determined to be qualified, and the next target feature is detected. If one or both of the contour information and the size value are inconsistent, the shape and size of the product are determined to be unqualified, and the product is classified into the shape and size unqualified area; S42: Obtain a surface feature photo of the product, determine the type, size, quantity and depth of defects on the surface of the product based on the surface feature photo, and determine whether the product is qualified based on the maximum allowable size, maximum number and maximum allowable depth of the defect. If the size, quantity and depth corresponding to a defect do not reach the maximum allowable value of the defect, the product is determined to be qualified and the next target feature is detected; if any of the size, quantity and depth corresponding to a defect exceeds the maximum allowable value of the defect, the product is determined to be unqualified and the product is classified into the surface defect area; S43: Obtain a color feature photo of the product, obtain the actual color value of the product according to the color feature photo, and compare the actual color value with the standard color value required by the product, and quantify the difference through a color difference formula, and the color difference formula is as follows: , where ∆L represents the difference between the actual brightness of the product and the standard brightness, ∆a represents the difference between the actual red-green degree of the product and the standard red-green degree, and ∆b represents the difference between the actual yellow-blue degree of the product and the standard yellow-blue degree. When 0≤∆L≤1, it means that the actual color value of the product is slightly different from the standard color value, and the color of the product is slightly different from the standard color, and the next target feature is detected; when 1<∆L≤2, it means that the actual color value of the product is significantly different from the standard color value, and the color of the product is significantly different from the standard color, and the product is classified into the color difference unqualified area; S44: Obtain a structural feature photo of the product, align the structural feature photo with the standard template image, and confirm the deviation value between the structure of the product and the structure on the standard template image, wherein the deviation value includes the deviation value of the structure in the angle and the deviation value of the structure in the horizontal direction. If the deviation value is within the tolerance range, the product is qualified and the product is classified as a qualified product. If the deviation value is outside the tolerance range, the product is classified as a structural deviation area.

[0008] An AOI visual online high-speed inspection system, the system comprising: Product information identification system: obtains the type information of products on the production line, confirms the features of the products that need to be detected, and sets the features that need to be detected as target features to obtain multiple target features; Camera configuration system: Set the frame rate and resolution of the camera according to the actual size of the target feature and the required clarity; Light source setting system: obtains surface material information from the pre-acquired product type information, selects a light source according to the surface material information, and sets the irradiation angle and intensity of the light source so that the light source will not be reflected on the image surface and cause image quality problems; Image detection and product classification system: obtain photos of multiple target features of the product, compare the multiple target features according to standard parameters, and classify the products into qualified and unqualified products based on the comparison results.

[0009] Preferably, the product information identification system comprises: Product identification information matching system: Scan the identification information of products on the production line through scanning equipment, match the collected identification information with the central database, and obtain product type information; Capture sequence setting system: According to the requirements for feature detection of products in the central database, the features to be detected are set as target features, and the target features include but are not limited to shape and size, surface defects, color and structural features. Multiple target features are obtained, and the image capture sequence is set for each target feature.

[0010] Preferably, the camera configuration system comprises: Camera frame rate setting system: obtain the actual size of each target feature, the actual size represents the size of the target feature in length, and set the frame rate of the camera according to the minimum actual size of multiple target features, and set the frame rate of the camera according to the following formula, , where D represents the moving distance between frames, V represents the speed of the production line, that is, the moving distance of the product in the production line per unit time, and F represents the frame rate of the camera, that is, the number of photos taken by the camera per unit time. When the camera frame rate satisfies: D≤actual size, the camera frame rate setting is completed at this time; Camera resolution setting system: According to the required clarity of each target feature, set the corresponding resolution of the camera at each clarity. When the current target feature image is acquired, the camera switches to the resolution required for the next target feature according to the preset image acquisition sequence.

[0011] Preferably, the light source setting system comprises: Surface material analysis system: obtains surface material information of the product according to the product type information, and the surface material information includes but is not limited to plastic, metal, ceramic and glass and the corresponding reflective properties and surface smoothness; Intelligent light source configuration system: selects light sources according to surface material information, the light sources include but are not limited to LED, laser and infrared light; sets the irradiation angle and intensity of the light source according to the reflection angle of the surface material, and arranges the light sources in a ring shape.

[0012] Preferably, the image detection and product classification system comprises: Shape and size detection system: obtains the shape and size feature photos of the product, obtains the contour information and size value of the product based on the shape and size feature photos, and compares them with the contour template and the tolerance range in the standard parameters. If the contour information is consistent with the contour template and the size value is within the tolerance range, the shape and size of the product are determined to be qualified, and the next target feature is detected. If one or both of the contour information and the size value do not meet the requirements, the shape and size of the product are determined to be unqualified, and the product is classified into the unqualified shape and size area; Surface defect detection system: obtains surface feature photos of the product, determines the type, size, quantity and depth of defects on the product surface based on the surface feature photos, and determines whether the product is qualified based on the maximum allowable size, maximum number and maximum allowable depth of the defects. If the size, quantity and depth corresponding to a defect do not reach the maximum allowable value of the defect, the product is determined to be qualified and the next target feature is detected; if any of the size, quantity and depth corresponding to a defect exceeds the maximum allowable value of the defect, the product is determined to be unqualified and the product is classified into the surface defect area; Color difference analysis system: obtain the color characteristic photo of the product, obtain the actual color value of the product based on the color characteristic photo, and compare the actual color value with the standard color value required by the product, and quantify the difference through the color difference formula, and the color difference formula is as follows: , where ∆L represents the difference between the actual brightness of the product and the standard brightness, ∆a represents the difference between the actual red-green degree of the product and the standard red-green degree, and ∆b represents the difference between the actual yellow-blue degree of the product and the standard yellow-blue degree. When 0≤∆L≤1, it means that the actual color value of the product is slightly different from the standard color value, and the color of the product is slightly different from the standard color, and the next target feature is detected; when 1<∆L≤2, it means that the actual color value of the product is significantly different from the standard color value, and the color of the product is significantly different from the standard color, and the product is classified into the color difference unqualified area; Structural feature alignment system: obtain the structural feature photo of the product, align the structural feature photo with the standard template image, and confirm the deviation value between the structure of the product and the structure on the standard template image. The deviation value includes the deviation value of the structure in the angle and the deviation value of the structure in the horizontal direction. If the deviation value is within the tolerance range, the product is qualified and the product is classified as a qualified product. If the deviation value is outside the tolerance range, the product is classified as a structural deviation area.

[0013] Beneficial effects of the present invention: The AOI visual online high-speed testing method and system described in the present invention significantly improves the detection accuracy and speed by optimizing camera settings, intelligent light source configuration, and multi-target feature detection. Using advanced color difference algorithms and structural feature alignment technology, the system can accurately judge complex geometric structures and color differences, while enhancing the optical adaptability to different materials. In addition, through the systematic integration of detection modules, manual intervention is reduced, the overall automation level and the reliability of product quality control are improved, and ultimately the efficiency of the production line is improved and the defect rate and rework costs are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is an AOI vision online high-speed detection method described in the present invention. DETAILED DESCRIPTION

[0015] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0016] One embodiment of the present invention provides an AOI visual online high-speed detection method, the method comprising: S1: Acquire product type information on the production line, confirm features of the products that need to be detected, and set the features that need to be detected as target features to obtain multiple target features; S2: Set the frame rate and resolution of the camera according to the actual size of the target feature and the required clarity; S3: obtaining surface material information from the pre-acquired product category information, selecting a light source according to the surface material information, and setting the irradiation angle and intensity of the light source so that the light source is not reflected on the image surface and thus causes image quality problems; S4: Acquire photos of multiple target features of the product, compare the multiple target features according to standard parameters, and classify the products into qualified products and unqualified products according to the comparison results.

[0017] The working principle and effect of the above technical solution are as follows: the system first obtains the type information of the products on the production line and identifies the product features that need to be detected. These features often include size, shape, color, and surface features, etc. According to these features, the system sets the detection target. According to the actual size of the target feature and the required clarity, the system adjusts the frame rate and resolution of the camera. High frame rate effectively captures fast-moving products, while high resolution ensures that the details of the image are sufficient for accurate analysis. Using the acquired product material information, select the appropriate light source type, including LED or laser, and adjust the illumination angle and intensity of the light source. Through these adjustments, the system avoids the negative impact of light reflection on image quality and ensures that the image is clear and usable. The camera takes photos for each target feature and processes these photos. The system determines whether the product meets the quality requirements by comparing with standard parameters (such as tolerance range and color difference standard) and templates. Through automated detection, product quality can be monitored and evaluated in real time during the production process. This system reduces the need for manual detection errors and human intervention, and greatly improves production efficiency and consistency. It can quickly adapt to the detection requirements of a variety of products, and achieves comprehensive non-destructive inspection of product features by flexibly adjusting camera and light source parameters, as well as efficient image processing and analysis methods. Ultimately, this automated system not only helps manufacturers maintain high standards of product quality, but also provides them with more immediate and accurate quality management information, improving overall production line efficiency and reliability.

[0018] In one embodiment of the present invention, the S1 includes: S11: Scanning identification information of products on the production line through a scanning device, matching the collected identification information with a central database, and obtaining product type information; S12: According to the requirements for feature detection of products in the central database, the features to be detected are set as target features, wherein the target features include but are not limited to shape and size, surface defects, color and structural features, multiple target features are obtained, and an image acquisition sequence is set for each target feature.

[0019] The working principle and effect of the above technical solution are as follows: On the production line, a scanning device is used to automatically scan the identification information of each product. The scanned identification information is transmitted to the system and matched with the records in the central database. The database contains detailed information about each product, including its type, specification, inspection standard, etc. Once the match is successful, the system retrieves the category information of the specific product to form a basic information set. Based on the product category information extracted from the central database, the system identifies the features that need to be inspected for the product. This includes shape, size, surface defects, color, and structural features. The identified inspection features are set as system target features. In order to complete the inspection process in an orderly manner, the system sets the order of image acquisition for each target feature. The order is set based on the priority of the inspection and the needs of process optimization to ensure that the inspection of each feature can be carried out under the best conditions. Therefore, the inspection order of the target features includes shape and size, surface defects, color, and structural features. The main advantage of the AOI system in the target feature identification and setting stage lies in its highly intelligent and automated capabilities. By scanning the product identification information on the production line and matching it with the central database, the system can quickly and accurately identify the category information of each product and extract the corresponding inspection requirements. Accurate data matching and transmission avoids possible errors caused by manual identification and greatly improves production efficiency. In addition, the system automatically sets the detection target features and the appropriate image acquisition sequence according to the database instructions to ensure the accuracy and priority of various features during the detection process. This fully automated process design enables the production line to perform efficient quality control in real time and reliably, ultimately improving the product qualification rate and overall production efficiency.

[0020] In one embodiment of the present invention, S2 includes: S21: Obtain the actual size of each target feature, where the actual size represents the length of the target feature, and set the frame rate of the camera according to the minimum actual size of the multiple target features, and set the frame rate of the camera according to the following formula: , where D represents the moving distance between frames, V represents the speed of the production line, that is, the moving distance of the product in the production line per unit time, and F represents the frame rate of the camera, that is, the number of photos taken by the camera per unit time. When the camera frame rate satisfies: D≤actual size, the camera frame rate setting is completed at this time; S22: According to the required clarity of each target feature, the corresponding resolution of the camera at each clarity is set. When the current target feature image is acquired, the camera switches to the resolution required for the next target feature according to the preset image acquisition sequence.

[0021] The working principle and effect of the above technical solution are as follows: the system first identifies and obtains the actual size of each target feature, especially its length. Among multiple target features, find the feature with the smallest actual size as the adjustment standard. Set the camera frame rate (F) according to the smallest actual size to ensure that the camera can capture the complete imaging of the target feature. The frame rate is calculated based on the pixel movement and the speed (V) of the product on the production line, and must meet the condition that the inter-frame movement distance (D) is less than or equal to the actual minimum feature size. The detection accuracy required for each target feature will affect the camera resolution that needs to be set. Each target feature will have different clarity standards. According to the clarity requirements of the current target feature, the system sets the camera resolution to obtain the best quality image. After completing the detection of a target feature, the system dynamically switches the camera resolution to the setting required for the next target feature according to the preset image acquisition sequence and scheme. The advantage of the AOI system in the dynamic setting of camera frame rate and resolution lies in its flexibility and accuracy, and optimizes for different target features by adjusting camera parameters in real time. The system first identifies the size and clarity requirements of each feature, automatically sets the appropriate frame rate and resolution, and ensures that the camera can fully capture high-quality images. It not only ensures the integrity of details and the accuracy of detection, but also adapts to the diverse detection requirements of different products. Dynamically switching camera settings enables the system to maintain stable detection capabilities in a rapidly changing production environment and avoid unnecessary resource consumption, thereby improving overall production efficiency and product quality, and effectively supporting quality control and optimization processes in modern manufacturing processes.

[0022] In one embodiment of the present invention, S3 includes: S31: Acquire surface material information of the product according to the product type information, wherein the surface material information includes but is not limited to plastic, metal, ceramic and glass and the corresponding reflective properties and surface smoothness; S32: Select a light source according to the surface material information, the light source includes but is not limited to LED, laser and infrared light; set the irradiation angle and intensity of the light source according to the reflection angle of the surface material, and arrange the light source in a ring shape.

[0023] The working principle and effect of the above technical solution are as follows: the product type information obtained from the production line is used to query the material data of the product. The system obtains the surface material characteristics of the product from the database, covering the reflective characteristics and surface smoothness of common materials including but not limited to plastics, metals, ceramics, and glass. According to the obtained surface material characteristics, select the appropriate light source type. LED, laser and infrared light each have different optical properties and are suitable for different surface materials. For example, LED is suitable for general use, laser is suitable for occasions requiring higher resolution, and infrared light is suitable for perspective or detection of features under the surface. The angle and intensity of the light source are adjusted based on the reflective characteristics and surface smoothness. By setting the incident angle of the light, glare and interference caused by direct light are avoided. The light source is set to a ring layout to ensure uniform lighting and reduce the impact of shadows. This arrangement helps to reduce the highlights caused by direct reflection and increase the uniformity of light distribution in the detection area. The advantage of the AOI system in terms of light source configuration lies in its ability to customize and flexibly adjust for different materials. The system first obtains the surface material information of the product to ensure that the light source type and irradiation parameters can optimize the reflective characteristics of the matching material. This process prevents image quality degradation due to excessive or improper light reflection, and enhances light uniformity and feature clarity in the inspection area. By selecting a suitable light source such as LED, laser or infrared light, and adjusting the incident angle and intensity, the ring light source can provide uniform illumination from multiple directions, avoiding shadows and uneven lighting that may be caused by a single-direction light source. Through the ring arrangement, light can be better dispersed incident on the inspection surface, reducing the problem of direct light reflection on highly reflective surfaces and reducing possible glare, thereby improving the quality of images and inspection results.

[0024] In one embodiment of the present invention, the S4 includes: S41: Obtain a shape and size feature photo of the product, obtain the contour information and size value of the product according to the shape and size feature photo, and compare them with the contour template and the tolerance range in the standard parameter. If the contour information is consistent with the contour template and the size value is within the tolerance range, the shape and size of the product are determined to be qualified, and the next target feature is detected. If one or both of the contour information and the size value are inconsistent, the shape and size of the product are determined to be unqualified, and the product is classified into the shape and size unqualified area; S42: Obtain a surface feature photo of the product, determine the type, size, quantity and depth of defects on the surface of the product based on the surface feature photo, and determine whether the product is qualified based on the maximum allowable size, maximum number and maximum allowable depth of the defect. If the size, quantity and depth corresponding to a defect do not reach the maximum allowable value of the defect, the product is determined to be qualified and the next target feature is detected; if any of the size, quantity and depth corresponding to a defect exceeds the maximum allowable value of the defect, the product is determined to be unqualified and the product is classified into the surface defect area; S43: Obtain a color feature photo of the product, obtain the actual color value of the product according to the color feature photo, and compare the actual color value with the standard color value required by the product, and quantify the difference through a color difference formula, and the color difference formula is as follows: , where ∆L represents the difference between the actual brightness of the product and the standard brightness, ∆a represents the difference between the actual red-green degree of the product and the standard red-green degree, and ∆b represents the difference between the actual yellow-blue degree of the product and the standard yellow-blue degree. When 0≤∆L≤1, it means that the actual color value of the product is slightly different from the standard color value, and the color of the product is slightly different from the standard color, and the next target feature is detected; when 1<∆L≤2, it means that the actual color value of the product is significantly different from the standard color value, and the color of the product is significantly different from the standard color, and the product is classified into the color difference unqualified area; S44: Obtain a structural feature photo of the product, align the structural feature photo with the standard template image, and confirm the deviation value between the structure of the product and the structure on the standard template image, wherein the deviation value includes the deviation value of the structure in the angle and the deviation value of the structure in the horizontal direction. If the deviation value is within the tolerance range, the product is qualified and the product is classified as a qualified product. If the deviation value is outside the tolerance range, the product is classified as a structural deviation area.

[0025] The working principle and effect of the above technical solution are: using image acquisition technology to obtain photos of product shape and size characteristics. Extracting the contour information and size values ​​of the product based on these photos. Comparing the obtained contour information and size with the tolerance range in the predetermined template and standard parameters. If the contour and size are within the allowable range, it is judged as qualified in this dimension. If not, it is marked as unqualified and enters the shape and size unqualified area; take photos of the product surface for detailed analysis to determine the type, size, quantity and depth of defects on the surface, and compare these test results with the maximum size, quantity and depth allowed for defects. If all dimensions are within the allowable range, it is qualified; if any one exceeds the standard, it is marked as unqualified and classified into the surface defect area; take photos of the product's color characteristics for detection, obtain the actual color value of the product through the photo, use the color difference formula to measure the difference between the actual color value and the standard color value with ∆L, ∆a, ∆b, if the difference is within the specified range, continue to detect, if it exceeds the upper limit, the product is classified into the unqualified area due to color mismatch; capture photos of the product's structural features, align the acquired image with the standard template image, and compare the angle and horizontal position deviation of the structure. If the deviation is within the tolerance range, it indicates that the product structure is qualified; if any aspect of the deviation exceeds the standard, it is marked as a structural deviation product. By analyzing the key features of the product one by one, the system can fully identify potential defects and deviations, and quickly classify products that do not meet the standards to ensure that only qualified products enter the next production link. This refined detection mechanism reduces human detection errors with its high reliability and improves overall production efficiency. At the same time, the system automatically compares standard parameters to ensure that the color difference, surface defects and structural deviations of complex products are within a controllable range, thereby improving the accuracy and consistency of the quality assurance process.

[0026] One embodiment of the present invention provides an AOI visual online high-speed inspection system, the system comprising: Product information identification system: obtains the type information of products on the production line, confirms the features of the products that need to be detected, and sets the features that need to be detected as target features to obtain multiple target features; Camera configuration system: Set the frame rate and resolution of the camera according to the actual size of the target feature and the required clarity; Light source setting system: obtains surface material information from the pre-acquired product type information, selects a light source according to the surface material information, and sets the irradiation angle and intensity of the light source so that the light source will not be reflected on the image surface and cause image quality problems; Image detection and product classification system: obtain photos of multiple target features of the product, compare the multiple target features according to standard parameters, and classify the products into qualified and unqualified products based on the comparison results.

[0027] The working principle and effect of the above technical solution are as follows: the system first obtains the type information of the products on the production line and identifies the product features that need to be detected. These features often include size, shape, color, and surface features, etc. According to these features, the system sets the detection target. According to the actual size of the target feature and the required clarity, the system adjusts the frame rate and resolution of the camera. High frame rate effectively captures fast-moving products, while high resolution ensures that the details of the image are sufficient for accurate analysis. Using the acquired product material information, select the appropriate light source type, including LED or laser, and adjust the illumination angle and intensity of the light source. Through these adjustments, the system avoids the negative impact of light reflection on image quality and ensures that the image is clear and usable. The camera takes photos for each target feature and processes these photos. The system determines whether the product meets the quality requirements by comparing with standard parameters (such as tolerance range and color difference standard) and templates. Through automated detection, product quality can be monitored and evaluated in real time during the production process. This system reduces the need for manual detection errors and human intervention, and greatly improves production efficiency and consistency. It can quickly adapt to the detection requirements of a variety of products, and achieves comprehensive non-destructive inspection of product features by flexibly adjusting camera and light source parameters, as well as efficient image processing and analysis methods. Ultimately, this automated system not only helps manufacturers maintain high standards of product quality, but also provides them with more immediate and accurate quality management information, improving overall production line efficiency and reliability.

[0028] In one embodiment of the present invention, the product information identification system comprises: Product identification information matching system: Scan the identification information of products on the production line through scanning equipment, match the collected identification information with the central database, and obtain product type information; Capture sequence setting system: According to the requirements for feature detection of products in the central database, the features to be detected are set as target features, and the target features include but are not limited to shape and size, surface defects, color and structural features. Multiple target features are obtained, and the image capture sequence is set for each target feature.

[0029] The working principle and effect of the above technical solution are as follows: On the production line, a scanning device is used to automatically scan the identification information of each product. The scanned identification information is transmitted to the system and matched with the records in the central database. The database contains detailed information about each product, including its type, specification, inspection standard, etc. Once the match is successful, the system retrieves the category information of the specific product to form a basic information set. Based on the product category information extracted from the central database, the system identifies the features that need to be inspected for the product. This includes shape, size, surface defects, color, and structural features. The identified inspection features are set as system target features. In order to complete the inspection process in an orderly manner, the system sets the order of image acquisition for each target feature. The order is set based on the priority of the inspection and the needs of process optimization to ensure that the inspection of each feature can be carried out under the best conditions. Therefore, the inspection order of the target features includes shape and size, surface defects, color, and structural features. The main advantage of the AOI system in the target feature identification and setting stage lies in its highly intelligent and automated capabilities. By scanning the product identification information on the production line and matching it with the central database, the system can quickly and accurately identify the category information of each product and extract the corresponding inspection requirements. Accurate data matching and transmission avoids possible errors caused by manual identification and greatly improves production efficiency. In addition, the system automatically sets the detection target features and the appropriate image acquisition sequence according to the database instructions to ensure the accuracy and priority of various features during the detection process. This fully automated process design enables the production line to perform efficient quality control in real time and reliably, ultimately improving the product qualification rate and overall production efficiency.

[0030] In one embodiment of the present invention, the camera configuration system comprises: Camera frame rate setting system: obtain the actual size of each target feature, the actual size represents the size of the target feature in length, and set the frame rate of the camera according to the minimum actual size of multiple target features, and set the frame rate of the camera according to the following formula, , where D represents the moving distance between frames, V represents the speed of the production line, that is, the moving distance of the product in the production line per unit time, and F represents the frame rate of the camera, that is, the number of photos taken by the camera per unit time. When the camera frame rate satisfies: D≤actual size, the camera frame rate setting is completed at this time; Camera resolution setting system: According to the required clarity of each target feature, set the corresponding resolution of the camera at each clarity. When the current target feature image is acquired, the camera switches to the resolution required for the next target feature according to the preset image acquisition sequence.

[0031] The working principle and effect of the above technical solution are as follows: the system first identifies and obtains the actual size of each target feature, especially its length. Among multiple target features, find the feature with the smallest actual size as the adjustment standard. Set the camera frame rate (F) according to the smallest actual size to ensure that the camera can capture the complete imaging of the target feature. The frame rate is calculated based on the pixel movement and the speed (V) of the product on the production line, and must meet the condition that the inter-frame movement distance (D) is less than or equal to the actual minimum feature size. The detection accuracy required for each target feature will affect the camera resolution that needs to be set. Each target feature will have different clarity standards. According to the clarity requirements of the current target feature, the system sets the camera resolution to obtain the best quality image. After completing the detection of a target feature, the system dynamically switches the camera resolution to the setting required for the next target feature according to the preset image acquisition sequence and scheme. The advantage of the AOI system in the dynamic setting of camera frame rate and resolution lies in its flexibility and accuracy, and optimizes for different target features by adjusting camera parameters in real time. The system first identifies the size and clarity requirements of each feature, automatically sets the appropriate frame rate and resolution, and ensures that the camera can fully capture high-quality images. It not only ensures the integrity of details and the accuracy of detection, but also adapts to the diverse detection requirements of different products. Dynamically switching camera settings enables the system to maintain stable detection capabilities in a rapidly changing production environment and avoid unnecessary resource consumption, thereby improving overall production efficiency and product quality, and effectively supporting quality control and optimization processes in modern manufacturing processes.

[0032] In one embodiment of the present invention, the light source setting system comprises: Surface material analysis system: obtains surface material information of the product according to the product type information, and the surface material information includes but is not limited to plastic, metal, ceramic and glass and the corresponding reflective properties and surface smoothness; Intelligent light source configuration system: selects light sources according to surface material information, the light sources include but are not limited to LED, laser and infrared light; sets the irradiation angle and intensity of the light source according to the reflection angle of the surface material, and arranges the light sources in a ring shape.

[0033] Furthermore, the camera obtains the intensity of reflected light reflected by irradiating the surface of the product through its own photometer, converts the intensity of reflected light into an electrical signal, and transmits the electrical signal to the intelligent light source configuration system. After receiving the electrical signal, the intelligent light source configuration system compares the intensity of reflected light on the surface of the product with the maximum intensity of reflected light allowed by the surface of the product according to the electrical signal, and then dynamically adjusts the intensity of the incident light. Furthermore, the adjustment method is obtained through the following formula:

[0034] in, Indicates the intensity of reflected light (watts per square meter), Represents the intensity of incident light, R represents the surface reflectivity of the product, θ represents the angle of incident light, and s represents the diffuse reflectance. The value range of s is (0,1) when When the reflected light intensity is less than or equal to the minimum allowed reflected light intensity on the product surface, it means that the reflected light is too weak, which will cause some details in the image to be unclear due to insufficient light, increasing the risk of false judgment during the detection process. The intelligent light source configuration system enhances the intensity of the incident light; When the minimum reflected light intensity allowed on the product surface is less than When the reflected light intensity is less than or equal to the maximum allowed reflected light intensity on the product surface, it means that the reflected light intensity is just right at this time; when >When the maximum reflected light intensity allowed on the product surface is reached, it means that the reflected light is too strong, which will cause overexposure of the image, and the intelligent light source configuration system reduces the intensity of the incident light.

[0035] The working principle and effect of the above technical solution are as follows: the product type information obtained from the production line is used to query the material data of the product. The system obtains the surface material characteristics of the product from the database, covering the reflective characteristics and surface smoothness of common materials including but not limited to plastics, metals, ceramics, and glass. According to the obtained surface material characteristics, select the appropriate light source type. LED, laser and infrared light each have different optical properties and are suitable for different surface materials. For example, LED is suitable for general use, laser is suitable for occasions requiring higher resolution, and infrared light is suitable for perspective or detection of features under the surface. The angle and intensity of the light source are adjusted based on the reflective characteristics and surface smoothness. By setting the incident angle of the light, glare and interference caused by direct light are avoided. The light source is set to a ring layout to ensure uniform lighting and reduce the impact of shadows. This arrangement helps to reduce the highlights caused by direct reflection and increase the uniformity of light distribution in the detection area. The advantage of the AOI system in terms of light source configuration lies in its ability to customize and flexibly adjust for different materials. The system first obtains the surface material information of the product to ensure that the light source type and irradiation parameters can optimize the reflective characteristics of the matching material. This process prevents image quality degradation due to excessive or improper light reflection, and enhances light uniformity and feature clarity in the inspection area. By selecting a suitable light source such as LED, laser or infrared light, and adjusting the incident angle and intensity, the ring light source can provide uniform illumination from multiple directions, avoiding shadows and uneven lighting that may be caused by a single-direction light source. Through the ring arrangement, light can be better dispersed incident on the inspection surface, reducing the problem of direct light reflection on highly reflective surfaces and reducing possible glare, thereby improving the quality of images and inspection results.

[0036] In this formula for calculating the intensity of reflected light, by combining specular reflection R and diffuse reflection s, the formula can comprehensively represent the reflection characteristics of light on different types of surfaces. Specular reflection describes the situation where light is reflected at a specific angle, while diffuse reflection takes into account the characteristics of uniform scattering of light on the surface. θ is introduced in the formula to describe the effect of the incident angle on the intensity of reflected light. Smaller incident angles (i.e., close to parallel to the surface) increase the specular effect, while larger incident angles (i.e., vertical incidence) reduce this effect. The reflectivity R represents the reflective ability of the surface material. Different materials have different reflectivities, but they are all between 0 and 1. Metal materials have the highest reflectivity, which is 60%-90%, and natural materials have the lowest reflectivity, which is 15%-30%. The diffuse reflection factor is introduced because the diffuse reflection factor s is used to quantify the scattering ratio of light in diffuse reflection, so that the formula effectively reflects the light scattering of various surfaces (such as rough or textured surfaces). By changing the intensity of the incident light, the intensity of the reflected light is adjusted according to real-time detection needs. In one embodiment of the present invention, the image detection and product classification system comprises: Shape and size detection system: obtains the shape and size feature photos of the product, obtains the contour information and size value of the product based on the shape and size feature photos, and compares them with the contour template and the tolerance range in the standard parameters. If the contour information is consistent with the contour template and the size value is within the tolerance range, the shape and size of the product are determined to be qualified, and the next target feature is detected. If one or both of the contour information and the size value do not meet the requirements, the shape and size of the product are determined to be unqualified, and the product is classified into the unqualified shape and size area; Surface defect detection system: obtains surface feature photos of the product, determines the type, size, quantity and depth of defects on the product surface based on the surface feature photos, and determines whether the product is qualified based on the maximum allowable size, maximum number and maximum allowable depth of the defects. If the size, quantity and depth corresponding to a defect do not reach the maximum allowable value of the defect, the product is determined to be qualified and the next target feature is detected; if any of the size, quantity and depth corresponding to a defect exceeds the maximum allowable value of the defect, the product is determined to be unqualified and the product is classified into the surface defect area; Color difference analysis system: obtain the color characteristic photo of the product, obtain the actual color value of the product based on the color characteristic photo, and compare the actual color value with the standard color value required by the product, and quantify the difference through the color difference formula, and the color difference formula is as follows: , where ∆L represents the difference between the actual brightness of the product and the standard brightness, ∆a represents the difference between the actual red-green degree of the product and the standard red-green degree, and ∆b represents the difference between the actual yellow-blue degree of the product and the standard yellow-blue degree. When 0≤∆L≤1, it means that the actual color value of the product is slightly different from the standard color value, and the color of the product is slightly different from the standard color, and the next target feature is detected; when 1<∆L≤2, it means that the actual color value of the product is significantly different from the standard color value, and the color of the product is significantly different from the standard color, and the product is classified into the color difference unqualified area; Structural feature alignment system: obtain the structural feature photo of the product, align the structural feature photo with the standard template image, and confirm the deviation value between the structure of the product and the structure on the standard template image. The deviation value includes the deviation value of the structure in the angle and the deviation value of the structure in the horizontal direction. If the deviation value is within the tolerance range, the product is qualified and the product is classified as a qualified product. If the deviation value is outside the tolerance range, the product is classified as a structural deviation area.

[0037] The working principle and effect of the above technical solution are: using image acquisition technology to obtain photos of product shape and size characteristics. Extracting the contour information and size values ​​of the product based on these photos. Comparing the obtained contour information and size with the tolerance range in the predetermined template and standard parameters. If the contour and size are within the allowable range, it is judged as qualified in this dimension. If not, it is marked as unqualified and enters the shape and size unqualified area; take photos of the product surface for detailed analysis to determine the type, size, quantity and depth of defects on the surface, and compare these test results with the maximum size, quantity and depth allowed for defects. If all dimensions are within the allowable range, it is qualified; if any one exceeds the standard, it is marked as unqualified and classified into the surface defect area; take photos of the product's color characteristics for detection, obtain the actual color value of the product through the photo, use the color difference formula to measure the difference between the actual color value and the standard color value with ∆L, ∆a, ∆b, if the difference is within the specified range, continue to detect, if it exceeds the upper limit, the product is classified into the unqualified area due to color mismatch; capture photos of the product's structural features, align the acquired image with the standard template image, and compare the angle and horizontal position deviation of the structure. If the deviation is within the tolerance range, it indicates that the product structure is qualified; if any aspect of the deviation exceeds the standard, it is marked as a structural deviation product. By analyzing the key features of the product one by one, the system can fully identify potential defects and deviations, and quickly classify products that do not meet the standards to ensure that only qualified products enter the next production link. This refined detection mechanism reduces human detection errors with its high reliability and improves overall production efficiency. At the same time, the system automatically compares standard parameters to ensure that the color difference, surface defects and structural deviations of complex products are within a controllable range, thereby improving the accuracy and consistency of the quality assurance process.

[0038] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An AOI visual online high-speed detection method, characterized in that: The method comprises: S1: Acquire product type information on the production line, confirm features of the products that need to be detected, and set the features that need to be detected as target features to obtain multiple target features; S2: Set the frame rate and resolution of the camera according to the actual size of the target feature and the required clarity; S3: obtaining surface material information from the pre-acquired product category information, selecting a light source according to the surface material information, and setting the irradiation angle and intensity of the light source so that the light source is not reflected on the image surface and thus causes image quality problems; S4: Acquire photos of multiple target features of the product, compare the multiple target features according to standard parameters, and classify the products into qualified products and unqualified products according to the comparison results.

2. The AOI visual online high-speed detection method according to claim 1, characterized in that: The S1 includes: S11: Scanning identification information of products on the production line through a scanning device, matching the collected identification information with a central database, and obtaining product type information; S12: According to the requirements for feature detection of products in the central database, the features to be detected are set as target features, wherein the target features include but are not limited to shape and size, surface defects, color and structural features, multiple target features are obtained, and an image acquisition sequence is set for each target feature.

3. The AOI visual online high-speed detection method according to claim 1, characterized in that: The S2 includes: S21: Obtain the actual size of each target feature, where the actual size represents the length of the target feature, and set the frame rate of the camera according to the minimum actual size of the multiple target features, and set the frame rate of the camera according to the following formula: , where D represents the moving distance between frames, V represents the speed of the production line, i.e., the moving distance of the product on the production line per unit time, F represents the frame rate of the camera, in frames per second, θ represents the angle formed by the direction of product movement and the direction of the camera's optical axis, and α represents the camera's viewing angle, i.e., the angle range that the camera lens can capture the image; when the camera frame rate satisfies: D≤actual size, the camera frame rate setting is completed; however, when D≥actual size, the camera increases the frame rate while dynamically adjusting its own shooting angle until D≤actual size; S22: According to the required clarity of each target feature, the corresponding resolution of the camera at each clarity is set. When the current target feature image is acquired, the camera switches to the resolution required for the next target feature according to the preset image acquisition sequence.

4. The AOI visual online high-speed detection method according to claim 1, characterized in that: The S3 includes: S31: Acquire surface material information of the product according to the product type information, wherein the surface material information includes but is not limited to plastic, metal, ceramic and glass and the corresponding reflective properties and surface smoothness; S32: Select a light source according to the surface material information, the light source includes but is not limited to LED, laser and infrared light; set the irradiation angle and intensity of the light source according to the reflection angle of the surface material, and arrange the light source in a ring shape.

5. The AOI visual online high-speed detection method according to claim 1, characterized in that: The S4 includes: S41: Obtain a shape and size feature photo of the product, obtain the contour information and size value of the product according to the shape and size feature photo, and compare them with the contour template and the tolerance range in the standard parameter. If the contour information is consistent with the contour template and the size value is within the tolerance range, the shape and size of the product are determined to be qualified, and the next target feature is detected. If one or both of the contour information and the size value are inconsistent, the shape and size of the product are determined to be unqualified, and the product is classified into the shape and size unqualified area; S42: Obtain a surface feature photo of the product, determine the type, size, quantity and depth of defects on the surface of the product based on the surface feature photo, and determine whether the product is qualified based on the maximum allowable size, maximum number and maximum allowable depth of the defect. If the size, quantity and depth corresponding to a defect do not reach the maximum allowable value of the defect, the product is determined to be qualified and the next target feature is detected; if any of the size, quantity and depth corresponding to a defect exceeds the maximum allowable value of the defect, the product is determined to be unqualified and the product is classified into the surface defect area; S43: Obtain a color feature photo of the product, obtain the actual color value of the product according to the color feature photo, and compare the actual color value with the standard color value required by the product, and quantify the difference through a color difference formula, and the color difference formula is as follows: , where ∆L represents the difference between the actual brightness of the product and the standard brightness, ∆a represents the difference between the actual red-green degree of the product and the standard red-green degree, and ∆b represents the difference between the actual yellow-blue degree of the product and the standard yellow-blue degree. When 0≤∆L≤1, it means that the actual color value of the product is slightly different from the standard color value, and the color of the product is slightly different from the standard color, and the next target feature is detected; when 1<∆L≤2, it means that the actual color value of the product is significantly different from the standard color value, and the color of the product is significantly different from the standard color, and the product is classified into the color difference unqualified area; S44: Obtain a structural feature photo of the product, align the structural feature photo with the standard template image, and confirm the deviation value between the structure of the product and the structure on the standard template image, wherein the deviation value includes the deviation value of the structure in the angle and the deviation value of the structure in the horizontal direction. If the deviation value is within the tolerance range, the product is qualified and the product is classified as a qualified product. If the deviation value is outside the tolerance range, the product is classified as a structural deviation area.

6. An AOI visual online high-speed inspection system, characterized in that: The system comprises: Product information identification system: obtains the type information of products on the production line, confirms the features of the products that need to be detected, and sets the features that need to be detected as target features to obtain multiple target features; Camera configuration system: Set the frame rate and resolution of the camera according to the actual size of the target feature and the required clarity; Light source setting system: obtains the surface material information in the pre-acquired product type information, selects the light source according to the surface material information, and sets the irradiation angle and intensity of the light source so that the light source will not be reflected on the image surface and cause image quality problems; Image detection and product classification system: obtain photos of multiple target features of the product, compare the multiple target features according to standard parameters, and classify the products into qualified and unqualified products based on the comparison results.

7. The AOI visual online high-speed inspection system according to claim 6, characterized in that: The product information identification system comprises: Product identification information matching system: Scan the identification information of products on the production line through scanning equipment, match the collected identification information with the central database, and obtain product type information; Capture sequence setting system: according to the requirements for feature detection of products in the central database, the features to be detected are set as target features, and the target features include but are not limited to shape and size, surface defects, color and structural features, multiple target features are obtained, and the image capture sequence is set for each target feature.

8. The AOI visual online high-speed inspection system according to claim 6, characterized in that: The camera configuration system comprises: Camera frame rate setting system: obtain the actual size of each target feature, the actual size represents the size of the target feature in length, and set the frame rate of the camera according to the minimum actual size of multiple target features, and set the frame rate of the camera according to the following formula, , where D represents the moving distance between frames, V represents the speed of the production line, that is, the moving distance of the product in the production line per unit time, and F represents the frame rate of the camera, that is, the number of photos taken by the camera per unit time. When the camera frame rate satisfies: D≤actual size, the camera frame rate setting is completed at this time; Camera resolution setting system: According to the required clarity of each target feature, set the corresponding resolution of the camera at each clarity. When the current target feature image is acquired, the camera switches to the resolution required for the next target feature according to the preset image acquisition sequence.

9. The AOI visual online high-speed inspection system according to claim 6, characterized in that: The light source setting system comprises: Surface material analysis system: obtains surface material information of the product according to the product type information, and the surface material information includes but is not limited to plastic, metal, ceramic and glass and the corresponding reflective properties and surface smoothness; Intelligent light source configuration system: selects light sources according to surface material information, the light sources include but are not limited to LED, laser and infrared light; sets the irradiation angle and intensity of the light source according to the reflection angle of the surface material, and arranges the light sources in a ring shape.

10. The AOI visual online high-speed inspection system according to claim 6, characterized in that: The image detection and product classification system comprises: Shape and size detection system: obtains the shape and size feature photos of the product, obtains the contour information and size value of the product based on the shape and size feature photos, and compares them with the contour template and the tolerance range in the standard parameters. If the contour information is consistent with the contour template and the size value is within the tolerance range, the shape and size of the product are determined to be qualified, and the next target feature is detected. If one or both of the contour information and the size value do not meet the requirements, the shape and size of the product are determined to be unqualified, and the product is classified into the unqualified shape and size area; Surface defect detection system: obtains surface feature photos of the product, determines the type, size, quantity and depth of defects on the product surface based on the surface feature photos, and determines whether the product is qualified based on the maximum allowable size, maximum number and maximum allowable depth of the defects. If the size, quantity and depth corresponding to a defect do not reach the maximum allowable value of the defect, the product is determined to be qualified and the next target feature is detected; if any of the size, quantity and depth corresponding to a defect exceeds the maximum allowable value of the defect, the product is determined to be unqualified and the product is classified into the surface defect area; Color difference analysis system: obtain the color characteristic photo of the product, obtain the actual color value of the product based on the color characteristic photo, and compare the actual color value with the standard color value required by the product, and quantify the difference through the color difference formula, and the color difference formula is as follows: , where ∆L represents the difference between the actual brightness of the product and the standard brightness, ∆a represents the difference between the actual red-green degree of the product and the standard red-green degree, and ∆b represents the difference between the actual yellow-blue degree of the product and the standard yellow-blue degree. When 0≤∆L≤1, it means that the actual color value of the product is slightly different from the standard color value, and the color of the product is slightly different from the standard color, and the next target feature is detected; when 1<∆L≤2, it means that the actual color value of the product is significantly different from the standard color value, and the color of the product is significantly different from the standard color, and the product is classified into the color difference unqualified area; Structural feature alignment system: obtain the structural feature photo of the product, align the structural feature photo with the standard template image, and confirm the deviation value between the structure of the product and the structure on the standard template image. The deviation value includes the deviation value of the structure in the angle and the deviation value of the structure in the horizontal direction. If the deviation value is within the tolerance range, the product is qualified and the product is classified as a qualified product. If the deviation value is outside the tolerance range, the product is classified as a structural deviation area.

Images