Product yield monitoring system and method applied to high-speed package testing machine

Abstract

The invention discloses a product yield monitoring system and method applied to a high-speed package testing machine, and relates to the technical field of electronic component packaging monitoring control, and the system comprises the following modules: a pre-packaging detection module, which is used for arranging a conveyor belt on the outer side of a main console to convey unpackaged to-be-detected electronic components, arranging an indoor light source right above the main console, and arranging a light source right above the main console; the electronic components are shot through a high-definition camera on the upper side of the conveying belt to obtain pictures before packaging; the packaged marked products are identified through the reflection judgment module, the indoor lamp source is arranged over the main console, the conveying belt is arranged on the outer side of the main console, and the marked products are conveyed and transported by turning the conveying belt to penetrate through the main console, so that the movement direction of the conveying belt is changed, and then the movement direction of the products is changed; and finally, marking the misjudged electronic component as a to-be-identified product according to a judgment result, so that whether the electronic component covered by the reflective lines has scratches or not can be further determined subsequently.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic component packaging monitoring and control, and in particular to a product yield monitoring system and method applied to a high-speed package testing machine. Background Art

[0002] The high-speed electronic component packaging machine is a highly efficient device that integrates optical inspection, electrical testing, and automated sorting. It is designed specifically for assembly line packaging. It uses a high-resolution industrial camera combined with AI vision algorithms to scan the appearance of components in real time, and synchronously triggers a contact probe or non-contact LCR tester to verify electrical parameters. The inspection speed can reach 3,000 to 5,000 pieces per minute. It uses a multi-axis robotic arm and a vibration plate feeding system to achieve precise sorting, automatically reject defective products, and support SPC data statistics. It is widely used for final inspection and packaging of SMD components such as chips, resistors, and capacitors.

[0003] When packaging electronic components, the package testing machine uses a robot to feed the electronic components into anti-static bags for sealed packaging. After packaging, the packaging needs to be inspected again to prevent the surface of the electrical components from being scratched by the robot during feeding, leaving scratches and resulting in defective products. However, the surface of the anti-static bag is smooth and easily affected by light during the photo inspection process, resulting in reflective lines in the inspection image. On the one hand, these reflective lines will cause misjudgment of the inspection results, mistakenly thinking that they are scratches, resulting in waste of good products. On the other hand, the reflective lines will obscure the original appearance of the electronic components, making it difficult to correctly judge and inspect the surface of the electronic components. To this end, we propose a product yield monitoring system and method for high-speed package testing machines. Summary of the Invention

[0004] (1) Technical problems solved In view of the deficiencies in the prior art, the present invention provides a product yield monitoring system and method applied to a high-speed package testing machine to solve the above-mentioned problems in the prior art.

[0005] (2) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a product yield monitoring system applied to a high-speed package testing machine, comprising the following modules: The pre-pack inspection module uses a conveyor belt outside the main console to transport unpacked electronic components to be inspected. An indoor light source is set directly above the main console. A high-definition camera is used on the upper side of the conveyor belt to capture pre-pack images of the electronic components. The pre-pack images are then used to determine whether the electronic components are qualified, and unqualified electronic components are removed from the conveyor belt. The post-packaging inspection module uses a robotic arm to pack qualified electronic components, and uses a high-definition camera to take pictures of the packaged electronic components to determine whether the electronic components are qualified. Unqualified packaged electronic components are marked to obtain marked products, and qualified packaged electronic components are removed from the conveyor belt and recorded as qualified products. The reflection judgment module diverts the marked products from the conveyor belt to the main console for transportation. A high-definition camera captures the marked products to obtain a re-inspection image. The re-inspection image and the packaged image are used to determine whether there are any misjudgments of reflections in the packaged image of the marked products. Based on the judgment result, unqualified electronic components in the marked products are removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as products to be identified. The edge recognition module uses the re-inspection image and the pre-package image to determine whether there are scratches on the edge of the product to be identified. The scratched product to be identified is removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as secondary qualified products; The yield statistics module sets a statistical period, obtains the total number of electrical components delivered to the test package machine during the statistical period and records it as the total number of products, obtains the yield percentage through the total number of products, the first qualified products and the second qualified products, establishes a yield database and stores the yield percentage in the yield database.

[0006] Preferably, in the pre-packaging inspection module, whether the electronic components are qualified is determined by identifying the pre-packaging pictures and unqualified electronic components are separated from the conveyor belt, specifically: Step 1: Obtain a pre-package image, grayscale the pre-package image, and use the brightness characteristics of the grayscale-processed pre-package image to identify whether the electronic components in the pre-package image are qualified, and mark and record the locations of unqualified electronic components; Step 2: Obtain the position of the marked unqualified electronic components, and use the pneumatic rejection valve to blow the marked unqualified electronic components off the conveyor belt and collect them.

[0007] Preferably, in the post-packaging inspection module, whether the electronic components are qualified is determined by the post-packaging pictures, unqualified post-packaging electronic components are marked to obtain marked products, and qualified post-packaging electronic components are separated from the conveyor belt and recorded as qualified products. Specifically, Step 1: Obtain a packaged image, grayscale the packaged image, and use the brightness characteristics of the grayscale-processed packaged image to identify whether the electronic components in the packaged image are qualified; Step 2: Obtain the position information of the electronic components in the qualified packaged pictures and mark them as qualified products. Obtain the position information of the electronic components in the unqualified packaged pictures, mark the unqualified electronic components in the packaged pictures and record them as marked products. Step 3: Obtain the position information of the electronic components in the qualified packaged picture, and blow the qualified electronic components in the packaged picture off the conveyor belt through the pneumatic rejection valve and collect them.

[0008] Preferably, in the reflection judgment module, whether there is a reflection misjudgment in the packaged picture of the marked product is judged by re-checking the picture and the packaged picture, specifically: Step 1: Obtain a post-packaging image and a re-inspection image, grayscale-process the post-packaging image and the re-inspection image, respectively capture reflective lines in the grayscale-processed post-packaging image and the re-inspection image, obtain the extension direction of the reflective lines in the re-inspection image and the extension direction of the reflective lines in the post-packaging image, respectively capture the extension direction of the conveyor belt edge in the grayscale-processed post-packaging image and the re-inspection image, obtain a virtual angle of the re-inspection image by the extension direction of the reflective lines in the re-inspection image and the extension direction of the conveyor belt edge in the re-inspection image, and obtain a virtual angle of the post-packaging image by the extension direction of the reflective lines in the post-packaging image and the extension direction of the conveyor belt edge in the post-packaging image; Step 2: Measure the virtual angle between the re-inspection image and the packaged image respectively, and determine whether a virtual angle exists. If no virtual angle exists, define the packaged image or the re-inspection image as if the reflective line is parallel to the conveyor belt edge. If a virtual angle exists, proceed to step 3: Step 3: Determine the relationship between the virtual angle between the re-inspected image and the packaged image and 45°. If the virtual angle between the re-inspected image or the packaged image is less than or equal to 45°, define the re-inspected image or the packaged image as having its reflective line parallel to the conveyor belt edge. If the virtual angle between the re-inspected image or the packaged image is greater than 45°, define the re-inspected image or the packaged image as having its reflective line perpendicular to the conveyor belt edge. Step 4: Determine whether the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is consistent. If the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is consistent, the marked product is defined as having no misjudgment of reflection. If the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is inconsistent, the marked product is defined as having misjudgment of reflection.

[0009] Preferably, in the reflection judgment module, unqualified electronic components in the marked product are separated from the conveyor belt according to the judgment result, and the remaining electronic components on the conveyor belt are recorded as products to be identified. Specifically, the judgment result is obtained. If the judgment result is a reflection misjudgment, the marked product is defined as a product to be identified. If the judgment result is no reflection misjudgment, the marked product is defined as an unqualified electronic component, and the unqualified electronic component is blown off the conveyor belt and collected through the jet of the pneumatic rejection valve.

[0010] Preferably, in the edge recognition module, specifically: Step 1: Obtain a re-inspection image and a pre-packaging image of the product to be identified, and grayscale process the re-inspection image and the pre-packaging image of the product to be identified to obtain a re-inspection grayscale image and a pre-packaging grayscale image; Step 2: Obtain the position of the reflective line in the re-inspection grayscale image, capture the two uncovered edge positions of the electronic component parallel to the reflective line position in the re-inspection grayscale image, and identify whether there are scratches on the two uncovered edge positions of the electronic component based on brightness. If the identification result shows scratches, the product to be identified is defined as unqualified, and the pneumatic rejection valve is used to blow the unqualified product to be identified off the conveyor belt and collect it. If the identification result shows no scratches, proceed to step 3: Step three: obtain the position of the reflective lines in the grayscale image before the package, capture the two uncovered edge positions of the electronic components parallel to the reflective line positions in the grayscale image before the package, and identify whether the two uncovered edge positions of the electronic components have scratches by brightness. If the identification result is that there are scratches, the product to be identified is defined as unqualified, and the unqualified product to be identified is blown off the conveyor belt and collected by the pneumatic rejection valve. If the identification result is that there are no scratches, the product to be identified is defined as a secondary qualified product.

[0011] Preferably, in the yield statistics module, the yield percentage is obtained by the total number of products, first-pass products, and second-pass products, specifically: Step 1: Obtain the total number of electrical components delivered to the test charter aircraft during the statistical period and record it as the total number of products. Obtain the location information of the first-pass products and count the first-pass products. Obtain the location information of the second-pass products and count the second-pass products. Step 2: Sum the first qualified product counting result and the second qualified product counting result to obtain the total qualified number, divide the total qualified number by the total number of products, and multiply the product by 100% to obtain the yield percentage.

[0012] A product yield monitoring method applied to a high-speed package testing machine comprises the following steps: S1: A conveyor belt is set up outside the main console to transport unpacked electronic components to be tested. An indoor light source is set up directly above the main console. A high-definition camera is used on the upper side of the conveyor belt to capture pre-pack images of the electronic components. The pre-pack images are then used to determine whether the electronic components are qualified, and unqualified electronic components are removed from the conveyor belt. S2: The qualified electronic components are packaged by a robot, and the packaged electronic components are photographed by a high-definition camera to obtain a packaged image. The packaged image is used to determine whether the electronic components are qualified. Unqualified packaged electronic components are marked to obtain marked products. Qualified packaged electronic components are removed from the conveyor belt and recorded as qualified products. S3: The marked product is diverted and transported through the main control station via a conveyor belt. A high-definition camera is used to capture a re-inspection image of the marked product. The re-inspection image and the packaged image are used to determine whether there is any reflection misjudgment in the packaged image of the marked product. Based on the judgment result, unqualified electronic components in the marked product are removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as products to be identified; S4: Determine whether there are scratches on the edge of the product to be identified based on the re-inspection image and the pre-packaging image. Remove the scratched product from the conveyor belt and record the remaining electronic components on the conveyor belt as secondary qualified products. S5: Set a statistical period, obtain the total number of electrical components delivered to the test package machine during the statistical period and record it as the total number of products, obtain the yield percentage through the total number of products, the first qualified products and the second qualified products, establish a yield database and store the yield percentage in the yield database.

[0013] (3) Beneficial effects The present invention provides a product yield monitoring system and method applied to a high-speed package testing machine, which has the following beneficial effects: (1) In this solution, the marked product after packaging is identified by the reflective judgment module. Since the indoor light source is set just above the main console and the conveyor belt is set outside the main console, the marked product is transported through the conveyor belt through the main console, thereby changing the movement direction of the conveyor belt and then changing the movement direction of the product. Since the position of the light source is fixed, the position change of the reflective lines in the re-inspected picture is compared with the post-packaging picture to determine whether the defect in the unqualified product is a reflective line. According to the judgment result, the misjudged electronic components are marked as products to be identified, so as to facilitate subsequent further confirmation of whether the electronic components covered by the reflective lines have scratches.

[0014] (2) In this solution, the edge recognition module is used to detect the products to be identified that are misjudged as non-reflective. Since the products to be identified are all qualified products before packaging, the defects on the products to be identified can only be left by the robot during the handling and bagging process. The robot may misalign and scratch the surface of the electronic components during the clamping process. Therefore, at least one end of such scratches is located on the edge line of the electronic component surface. Since the position relationship between the reflective lines in the re-inspection image and the pre-package image is a cross, the two uncovered edges of the electronic components in the re-inspection image are first detected to see if there are scratches, and two of the four edges of the electronic components are detected. Then, the two uncovered edges of the electronic components in the pre-package image are detected to see if there are scratches, and the other two of the four edges of the electronic components are detected. By combining the two detection results, it is possible to finally determine whether there are scratches on the four edges of the electronic components, and then it is convenient to determine whether there are scratches under the reflective lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a product yield monitoring method applied to a high-speed package testing machine according to the present invention; Figure 2 This is a schematic diagram of the module structure of a product yield monitoring system applied to a high-speed package testing machine according to the present invention; Figure 3 This is a logical structure diagram of a product yield monitoring system applied to a high-speed package testing machine according to the present invention; Figure 4 The present invention is a schematic diagram of a top view of the structure of a package testing machine of a product yield monitoring system applied to a high-speed package testing machine.

[0016] In the picture: 1. Main console; 2. High-definition camera; 3. Pneumatic rejection valve; 4. Light source; 5. Conveyor belt; 6. Electronic component packaging equipment. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] See also Figures 1-4 The present invention provides a product yield monitoring system for a high-speed package testing machine, comprising the following modules: The pre-pack inspection module uses a conveyor belt outside the main console to transport unpacked electronic components to be inspected. An indoor light source is set directly above the main console. A high-definition camera is used on the upper side of the conveyor belt to capture pre-pack images of the electronic components. The pre-pack images are then used to determine whether the electronic components are qualified, and unqualified electronic components are removed from the conveyor belt. The post-packaging inspection module uses a robotic arm to pack qualified electronic components, and uses a high-definition camera to take pictures of the packaged electronic components to determine whether the electronic components are qualified. Unqualified packaged electronic components are marked to obtain marked products, and qualified packaged electronic components are removed from the conveyor belt and recorded as qualified products. The reflection judgment module diverts the marked products from the conveyor belt to the main console for transportation. A high-definition camera captures the marked products to obtain a re-inspection image. The re-inspection image and the packaged image are used to determine whether there are any misjudgments of reflections in the packaged image of the marked products. Based on the judgment result, unqualified electronic components in the marked products are removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as products to be identified. The edge recognition module uses the re-inspection image and the pre-package image to determine whether there are scratches on the edge of the product to be identified. The scratched product to be identified is removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as secondary qualified products; The yield statistics module sets a statistical period, obtains the total number of electrical components delivered to the test package machine during the statistical period and records it as the total number of products, obtains the yield percentage through the total number of products, the first qualified products and the second qualified products, establishes a yield database and stores the yield percentage in the yield database.

[0019] In this embodiment, the pre-packaging inspection module takes a picture of the unpacked electronic components, and then compares the picture with the qualified templates in the database to determine whether the electronic components are qualified. Unqualified electronic components are then removed from the conveyor belt. By placing the indoor light source directly above the main console, it is easier to control the uniform position of the light source, thereby facilitating the subsequent identification of whether the reflected light in the picture is a product defect. In this solution, the post-packaging inspection module takes photos of the packaged electronic components and compares the photos with the qualified template in the database to determine whether the packaged electronic components are qualified. Qualified products are collected and unqualified products are marked, making it easier to determine whether these unqualified products are misjudged due to reflection or are truly unqualified. In this solution, a reflective identification module is used to identify packaged marked products. Since the indoor light source is located directly above the main console and the conveyor belt is located outside the main console, the marked product is diverted through the main console for transportation, thereby changing the direction of movement of the conveyor belt and, in turn, the direction of movement of the product. Since the position of the light source is fixed, the position change of the reflective lines in the re-inspected image is determined by comparing the re-inspected image with the packaged image to determine whether the defect in the unqualified product is a reflective line. Based on the judgment result, the misidentified electronic components are marked as products to be identified, thereby facilitating subsequent further confirmation of whether the electronic components covered by the reflective lines have scratches. In this solution, the edge recognition module is used to detect the non-reflective misjudgment products to be identified. Since the products to be identified are qualified products before packaging, the defects on the products to be identified can only be left by the robot during the handling and bagging process. The robot may be misaligned and scratch the surface of the electronic component during the clamping process. Therefore, at least one end of such scratches is located on the edge line of the surface of the electronic component. Since the position relationship between the reflective lines in the re-inspection image and the pre-package image is a cross, the two uncovered edges of the electronic component in the re-inspection image are first detected to see if there are scratches, and two of the four edges of the electronic component are detected. Then, the two uncovered edges of the electronic component in the pre-package image are detected to see if there are scratches, and the other two of the four edges of the electronic component are detected. By combining the two detection results, it is possible to finally determine whether there are scratches on the four edges of the electronic component, and then it is convenient to determine whether there are scratches under the reflective lines. This solution calculates the yield percentage in the yield statistics module by the total number of products, first-pass products, and second-pass products, and stores the data, which makes it easier to count the yield in the production line and facilitate the maintenance and update of the equipment.

[0020] In the pre-packaging inspection module, the quality of electronic components is determined by identifying the pre-packaging image and unqualified electronic components are removed from the conveyor belt. Specifically: Step 1: Obtain a pre-package image, grayscale the pre-package image, and use the brightness characteristics of the grayscale-processed pre-package image to identify whether the electronic components in the pre-package image are qualified, and mark and record the locations of unqualified electronic components; Step 2: Obtain the position of the marked unqualified electronic components, and use the pneumatic rejection valve to blow the marked unqualified electronic components off the conveyor belt and collect them.

[0021] In this embodiment, the pre-packaging picture of the electronic component is obtained by taking a picture of the electronic component before packaging, and then the grayscale of the picture is processed and the defects in the grayscale picture are identified by brightness, so as to determine whether the electronic component is qualified; It is worth mentioning that the specific judgment method is to compare the captured image with the qualified images in the database. If they are consistent, they are qualified, and if they are inconsistent, they are unqualified. This technical means is a widely used existing technology, and the specific details will not be elaborated here. The pneumatic rejection valve is a commonly used screening and rejection method in this field. The application of this existing technology is very mature and the specific operating details will not be elaborated here.

[0022] In the post-packaging inspection module, the quality of electronic components is determined by post-packaging images. Unqualified post-packaging electronic components are marked to obtain marked products. Qualified post-packaging electronic components are removed from the conveyor belt and recorded as qualified products. Specifically: Step 1: Obtain a packaged image, grayscale the packaged image, and use the brightness characteristics of the grayscale-processed packaged image to identify whether the electronic components in the packaged image are qualified; Step 2: Obtain the position information of the electronic components in the qualified packaged pictures and mark them as qualified products. Obtain the position information of the electronic components in the unqualified packaged pictures, mark the unqualified electronic components in the packaged pictures and record them as marked products. Step 3: Obtain the position information of the electronic components in the qualified packaged picture, and blow the qualified electronic components in the packaged picture off the conveyor belt through the pneumatic rejection valve and collect them.

[0023] In the reflection judgment module, the re-inspection image and the packaged image are used to determine whether there is any reflection misjudgment in the packaged image of the marked product. Specifically: Step 1: Obtain a post-packaging image and a re-inspection image, grayscale-process the post-packaging image and the re-inspection image, respectively capture reflective lines in the grayscale-processed post-packaging image and the re-inspection image, obtain the extension direction of the reflective lines in the re-inspection image and the extension direction of the reflective lines in the post-packaging image, respectively capture the extension direction of the conveyor belt edge in the grayscale-processed post-packaging image and the re-inspection image, obtain a virtual angle of the re-inspection image by the extension direction of the reflective lines in the re-inspection image and the extension direction of the conveyor belt edge in the re-inspection image, and obtain a virtual angle of the post-packaging image by the extension direction of the reflective lines in the post-packaging image and the extension direction of the conveyor belt edge in the post-packaging image; Step 2: Measure the virtual angle between the re-inspection image and the packaged image respectively, and determine whether a virtual angle exists. If no virtual angle exists, define the packaged image or the re-inspection image as if the reflective line is parallel to the conveyor belt edge. If a virtual angle exists, proceed to step 3: Step 3: Determine the relationship between the virtual angle between the re-inspected image and the packaged image and 45°. If the virtual angle between the re-inspected image or the packaged image is less than or equal to 45°, define the re-inspected image or the packaged image as having its reflective line parallel to the conveyor belt edge. If the virtual angle between the re-inspected image or the packaged image is greater than 45°, define the re-inspected image or the packaged image as having its reflective line perpendicular to the conveyor belt edge. Step 4: Determine whether the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is consistent. If the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is consistent, the marked product is defined as having no misjudgment of reflection. If the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is inconsistent, the marked product is defined as having misjudgment of reflection.

[0024] In this embodiment, the position of the reflective lines on the anti-static bag is identified by capturing the reflective lines in the post-packaging picture and the re-inspection picture. Since the indoor light source is set directly above the main console and the conveyor belt is set outside the main console, the marked product is turned through the conveyor belt for transportation through the main console, thereby changing the movement direction of the conveyor belt and then changing the movement direction of the product. Since the position of the light source is fixed, the position change of the reflective lines in the re-inspection picture and the post-packaging picture is judged to determine whether the defects in the unqualified product are reflective lines. According to the judgment result, the misjudged electronic components are marked as products to be identified, so as to facilitate subsequent further confirmation of whether the electronic components covered by the reflective lines have scratches.

[0025] In the reflection judgment module, unqualified electronic components in the marked product are separated from the conveyor belt according to the judgment result, and the remaining electronic components on the conveyor belt are recorded as products to be identified. Specifically, the judgment result is obtained. If the judgment result is a reflection misjudgment, the marked product is defined as a product to be identified. If the judgment result is no reflection misjudgment, the marked product is defined as an unqualified electronic component, and the unqualified electronic component is blown off the conveyor belt and collected through the pneumatic rejection valve.

[0026] In the edge recognition module, specifically: Step 1: Obtain a re-inspection image and a pre-packaging image of the product to be identified, and grayscale process the re-inspection image and the pre-packaging image of the product to be identified to obtain a re-inspection grayscale image and a pre-packaging grayscale image; Step 2: Obtain the position of the reflective line in the re-inspection grayscale image, capture the two uncovered edge positions of the electronic component parallel to the reflective line position in the re-inspection grayscale image, and identify whether there are scratches on the two uncovered edge positions of the electronic component based on brightness. If the identification result shows scratches, the product to be identified is defined as unqualified, and the pneumatic rejection valve is used to blow the unqualified product to be identified off the conveyor belt and collect it. If the identification result shows no scratches, proceed to step 3: Step three: obtain the position of the reflective lines in the grayscale image before the package, capture the two uncovered edge positions of the electronic components parallel to the reflective line positions in the grayscale image before the package, and identify whether the two uncovered edge positions of the electronic components have scratches by brightness. If the identification result is that there are scratches, the product to be identified is defined as unqualified, and the unqualified product to be identified is blown off the conveyor belt and collected by the pneumatic rejection valve. If the identification result is that there are no scratches, the product to be identified is defined as a secondary qualified product.

[0027] In this embodiment, since the products to be identified are all qualified products before packaging, the defects on the products to be identified can only be left by the robot during the handling and bagging process. The robot may be misaligned and scratch the surface of the electronic component during the clamping and detachment process. Therefore, at least one end of such scratches is located on the edge line of the surface of the electronic component. Since the positional relationship between the reflective lines in the re-inspection picture and the pre-package picture is a cross, the two uncovered edges of the electronic component in the re-inspection picture are first detected to see if there are scratches, two of the four edges of the electronic component are detected, and then the two uncovered edges of the electronic component in the pre-package picture are detected to see if there are scratches, and the other two of the four edges of the electronic component are detected. The two detection results are combined to finally determine whether there are scratches on the four edges of the electronic component, and then it is convenient to determine whether there are scratches under the reflective lines.

[0028] In the yield statistics module, the yield percentage is obtained by the total number of products, first-pass products, and second-pass products. Specifically: Step 1: Obtain the total number of electrical components delivered to the test charter aircraft during the statistical period and record it as the total number of products. Obtain the location information of the first-pass products and count the first-pass products. Obtain the location information of the second-pass products and count the second-pass products. Step 2: Sum the first qualified product counting result and the second qualified product counting result to obtain the total qualified number, divide the total qualified number by the total number of products, and multiply the product by 100% to obtain the yield percentage.

[0029] In this embodiment, the yield percentage is calculated based on the total number of products, first-pass products, and second-pass products, and the data is stored, thereby facilitating statistics on the yield rate in the production line and further facilitating maintenance and updating of equipment.

[0030] See also Figures 1-4 The present invention provides a product yield monitoring method applied to a high-speed package testing machine, comprising the following steps: S1: A conveyor belt is set up outside the main console to transport unpacked electronic components to be tested. An indoor light source is set up directly above the main console. A high-definition camera is used on the upper side of the conveyor belt to capture pre-pack images of the electronic components. The pre-pack images are then used to determine whether the electronic components are qualified, and unqualified electronic components are removed from the conveyor belt. S2: The qualified electronic components are packaged by a robot, and the packaged electronic components are photographed by a high-definition camera to obtain a packaged image. The packaged image is used to determine whether the electronic components are qualified. Unqualified packaged electronic components are marked to obtain marked products. Qualified packaged electronic components are removed from the conveyor belt and recorded as qualified products. S3: The marked product is diverted and transported through the main control station via a conveyor belt. A high-definition camera is used to capture a re-inspection image of the marked product. The re-inspection image and the packaged image are used to determine whether there is any reflection misjudgment in the packaged image of the marked product. Based on the judgment result, unqualified electronic components in the marked product are removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as products to be identified; S4: Determine whether there are scratches on the edge of the product to be identified based on the re-inspection image and the pre-packaging image. Remove the scratched product from the conveyor belt and record the remaining electronic components on the conveyor belt as secondary qualified products. S5: Set a statistical period, obtain the total number of electrical components delivered to the test package machine during the statistical period and record it as the total number of products, obtain the yield percentage through the total number of products, the first qualified products and the second qualified products, establish a yield database and store the yield percentage in the yield database.

[0031] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0032] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0033] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A product yield monitoring system applied to a high-speed package testing machine, characterized in that: Includes the following modules: The pre-pack inspection module uses a conveyor belt outside the main console to transport unpacked electronic components to be inspected. An indoor light source is set directly above the main console. A high-definition camera is used on the upper side of the conveyor belt to capture pre-pack images of the electronic components. The pre-pack images are then used to determine whether the electronic components are qualified, and unqualified electronic components are removed from the conveyor belt. The post-packaging inspection module uses a robotic arm to pack qualified electronic components, and uses a high-definition camera to take pictures of the packaged electronic components to determine whether the electronic components are qualified. Unqualified packaged electronic components are marked to obtain marked products, and qualified packaged electronic components are removed from the conveyor belt and recorded as qualified products. The reflection judgment module diverts the marked products from the conveyor belt to the main console for transportation. A high-definition camera captures the marked products to obtain a re-inspection image. The re-inspection image and the packaged image are used to determine whether there are any misjudgments of reflections in the packaged image of the marked products. Based on the judgment result, unqualified electronic components in the marked products are removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as products to be identified. The edge recognition module uses the re-inspection image and the pre-package image to determine whether there are scratches on the edge of the product to be identified. The scratched product to be identified is removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as secondary qualified products; The yield statistics module sets a statistical period, obtains the total number of electrical components delivered to the test package machine during the statistical period and records it as the total number of products, obtains the yield percentage through the total number of products, the first qualified products and the second qualified products, establishes a yield database and stores the yield percentage in the yield database.

2. The product yield monitoring system for a high-speed package testing machine according to claim 1, characterized in that: In the pre-packaging inspection module, the quality of electronic components is determined by identifying the pre-packaging image and unqualified electronic components are removed from the conveyor belt. Specifically: Step 1: Obtain a pre-package image, grayscale the pre-package image, and use the brightness characteristics of the grayscale-processed pre-package image to identify whether the electronic components in the pre-package image are qualified, and mark and record the locations of unqualified electronic components; Step 2: Obtain the position of the marked unqualified electronic components, and use the pneumatic rejection valve to blow the marked unqualified electronic components off the conveyor belt and collect them.

3. The product yield monitoring system for a high-speed package testing machine according to claim 1, characterized in that: In the post-packaging inspection module, the quality of electronic components is determined by post-packaging images. Unqualified post-packaging electronic components are marked to obtain marked products. Qualified post-packaging electronic components are removed from the conveyor belt and recorded as qualified products. Specifically: Step 1: Obtain a packaged image, grayscale the packaged image, and use the brightness characteristics of the grayscale-processed packaged image to identify whether the electronic components in the packaged image are qualified; Step 2: Obtain the position information of the electronic components in the qualified packaged pictures and mark them as qualified products. Obtain the position information of the electronic components in the unqualified packaged pictures, mark the unqualified electronic components in the packaged pictures and record them as marked products. Step 3: Obtain the position information of the electronic components in the qualified packaged picture, and blow the qualified electronic components in the packaged picture off the conveyor belt through the pneumatic rejection valve and collect them.

4. The product yield monitoring system for a high-speed package testing machine according to claim 1, characterized in that: In the reflection judgment module, the re-inspection image and the packaged image are used to determine whether there is any reflection misjudgment in the packaged image of the marked product. Specifically: Step 1: Obtain a post-packaging image and a re-inspection image, grayscale-process the post-packaging image and the re-inspection image, respectively capture reflective lines in the grayscale-processed post-packaging image and the re-inspection image, obtain the extension direction of the reflective lines in the re-inspection image and the extension direction of the reflective lines in the post-packaging image, respectively capture the extension direction of the conveyor belt edge in the grayscale-processed post-packaging image and the re-inspection image, obtain a virtual angle of the re-inspection image by the extension direction of the reflective lines in the re-inspection image and the extension direction of the conveyor belt edge in the re-inspection image, and obtain a virtual angle of the post-packaging image by the extension direction of the reflective lines in the post-packaging image and the extension direction of the conveyor belt edge in the post-packaging image; Step 2: Measure the virtual angle between the re-inspection image and the packaged image respectively, and determine whether a virtual angle exists. If no virtual angle exists, define the packaged image or the re-inspection image as if the reflective line is parallel to the conveyor belt edge. If a virtual angle exists, proceed to step 3: Step 3: Determine the relationship between the virtual angle between the re-inspected image and the packaged image and 45°. If the virtual angle between the re-inspected image or the packaged image is less than or equal to 45°, define the re-inspected image or the packaged image as having its reflective line parallel to the conveyor belt edge. If the virtual angle between the re-inspected image or the packaged image is greater than 45°, define the re-inspected image or the packaged image as having its reflective line perpendicular to the conveyor belt edge. Step 4: Determine whether the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is consistent. If the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is consistent, the marked product is defined as having no misjudgment of reflection. If the positional relationship between the reflective line in the re-inspected image and the image after packaging and the edge of the conveyor belt is inconsistent, the marked product is defined as having misjudgment of reflection.

5. The product yield monitoring system for a high-speed package testing machine according to claim 1 is characterized in that: In the reflection judgment module, unqualified electronic components in the marked product are separated from the conveyor belt according to the judgment result, and the remaining electronic components on the conveyor belt are recorded as products to be identified. Specifically, the judgment result is obtained. If the judgment result is a reflection misjudgment, the marked product is defined as a product to be identified. If the judgment result is no reflection misjudgment, the marked product is defined as an unqualified electronic component, and the unqualified electronic component is blown off the conveyor belt and collected through the pneumatic rejection valve.

6. The product yield monitoring system for a high-speed package testing machine according to claim 1, characterized in that: In the edge recognition module, specifically: Step 1: Obtain a re-inspection image and a pre-packaging image of the product to be identified, and grayscale process the re-inspection image and the pre-packaging image of the product to be identified to obtain a re-inspection grayscale image and a pre-packaging grayscale image; Step 2: Obtain the position of the reflective line in the re-inspection grayscale image, capture the two uncovered edge positions of the electronic component parallel to the reflective line position in the re-inspection grayscale image, and identify whether there are scratches on the two uncovered edge positions of the electronic component based on brightness. If the identification result shows scratches, the product to be identified is defined as unqualified, and the pneumatic rejection valve is used to blow the unqualified product to be identified off the conveyor belt and collect it. If the identification result shows no scratches, proceed to step 3: Step three: obtain the position of the reflective lines in the grayscale image before the package, capture the two uncovered edge positions of the electronic components parallel to the reflective line positions in the grayscale image before the package, and identify whether the two uncovered edge positions of the electronic components have scratches by brightness. If the identification result is that there are scratches, the product to be identified is defined as unqualified, and the unqualified product to be identified is blown off the conveyor belt and collected by the pneumatic rejection valve. If the identification result is that there are no scratches, the product to be identified is defined as a secondary qualified product.

7. The product yield monitoring system for a high-speed package testing machine according to claim 1, characterized in that: In the yield statistics module, the yield percentage is obtained by the total number of products, first-pass products, and second-pass products. Specifically: Step 1: Obtain the total number of electrical components delivered to the test charter aircraft during the statistical period and record it as the total number of products. Obtain the location information of the first-pass products and count the first-pass products. Obtain the location information of the second-pass products and count the second-pass products. Step 2: Sum the first qualified product counting result and the second qualified product counting result to obtain the total qualified number, divide the total qualified number by the total number of products, and multiply the product by 100% to obtain the yield percentage.

8. A product yield monitoring method for a high-speed package inspection machine, applied to a product yield monitoring system for a high-speed package inspection machine according to any one of claims 1 to 7, characterized in that: The steps include: S1: A conveyor belt is set up outside the main console to transport unpacked electronic components to be tested. An indoor light source is set up directly above the main console. A high-definition camera is used on the upper side of the conveyor belt to capture pre-pack images of the electronic components. The pre-pack images are then used to determine whether the electronic components are qualified, and unqualified electronic components are removed from the conveyor belt. S2: The qualified electronic components are packaged by a robot, and the packaged electronic components are photographed by a high-definition camera to obtain a packaged image. The packaged image is used to determine whether the electronic components are qualified. Unqualified packaged electronic components are marked to obtain marked products. Qualified packaged electronic components are removed from the conveyor belt and recorded as qualified products. S3: The marked product is diverted and transported through the main control station via a conveyor belt. A high-definition camera is used to capture a re-inspection image of the marked product. The re-inspection image and the packaged image are used to determine whether there is any reflection misjudgment in the packaged image of the marked product. Based on the judgment result, unqualified electronic components in the marked product are removed from the conveyor belt, and the remaining electronic components on the conveyor belt are recorded as products to be identified; S4: Determine whether there are scratches on the edge of the product to be identified based on the re-inspection image and the pre-packaging image. Remove the scratched product from the conveyor belt and record the remaining electronic components on the conveyor belt as secondary qualified products. S5: Set a statistical period, obtain the total number of electrical components delivered to the test package machine during the statistical period and record it as the total number of products, obtain the yield percentage through the total number of products, the first qualified products and the second qualified products, establish a yield database and store the yield percentage in the yield database.

Images