Visual inspection system and method capable of being used for multi-product inspection

Abstract

The invention provides a visual detection system and method for multi-product detection, and the system comprises a control terminal which comprises a transportation module, a data storage module, a product recognition module, a visual detection module, and a classification module. The positioning unit is additionally arranged in the visual detection module, passing products can be stopped on the transportation module, the detection unit can perform accurate detection, through the arrangement, manual operation is replaced by mechanical operation, the transportation module does not need to be operated to be frequently stopped and started, the detection process is smoother, and the detection efficiency is improved. And the classification module is used for receiving information transmitted by the detection control unit and classifying and storing incomplete products and qualified products, and can be suitable for detecting various products produced in batches.

Description

Technical Field

[0001] The present invention relates to the technical field of visual inspection equipment, and in particular to a visual inspection system and method that can be used for multiple product inspections. Background Art

[0002] With the development of society, people have higher and higher requirements for product quality, so it is necessary to test products before they leave the factory to ensure product quality. At present, the quality inspection of the appearance and shape of known batch products is mostly carried out by human eyes. Due to the fact that human eyes are difficult to quantify and execute, human eyes are easily fatigued, people are easily emotional, and some people cannot work for a long time in high temperature or toxic environments, the current product visual inspection generally has a high rate of misjudgment and wrong judgment.

[0003] As a result, machine vision inspection technology has been rapidly developed and popularized. However, the existing machine vision inspection technology has a single functionality, and a machine vision inspection device can generally only detect a single specific product. Since multiple components of a complete product often need to be visually inspected during the production process, it is necessary to configure a variety of visual inspection machines, which not only wastes resources, but also reduces the overall inspection efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a visual inspection system and method that can be used for multi-product inspection, so as to solve the problem that existing machine vision inspection devices can generally only detect a single specific product, and need to configure multiple visual inspection machines, which wastes resources and reduces the overall inspection efficiency.

[0005] According to one object of the present invention, the present invention provides a visual inspection system that can be used for multi-product inspection, including a control terminal, wherein the control terminal includes:

[0006] A transport module, wherein the transport module is used to transport the product;

[0007] A data storage module, the data storage module includes a visual distinguishing feature storage library for commonly tested products, a detection configuration parameter storage library, and a visual inspection tool storage library, which are respectively used to store visual distinguishing features of various commonly tested products, corresponding detection configuration parameters of various commonly tested products, and visual inspection tools of various commonly tested products;

[0008] A product identification module, the product identification module is used to identify which common test product the product to be tested is;

[0009] A visual inspection module, the visual inspection module comprising a detection control unit and a positioning unit. The detection control unit calls the corresponding visual inspection tool to load the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool according to the recognition result of the product recognition module, performs visual inspection on the product to be inspected, and transmits the information to the classification module; the positioning unit is used to fix the moving product in the inspection area of ​​the detection unit and release the product after the inspection is completed;

[0010] The classification module is used to receive information transmitted by the detection control unit, and classify and store defective products and qualified products, and is suitable for detecting a variety of products produced in batches.

[0011] Furthermore, the product recognition module includes an image acquisition unit, and the image acquisition unit includes a plurality of cameras for taking multi-angle shots of the product.

[0012] Furthermore, the visual distinguishing feature repository for commonly tested products is used to pre-store visual distinguishing features of various commonly tested products, the detection configuration parameter repository is used to pre-store corresponding detection configuration parameters of various commonly tested products, and the visual inspection tool repository is used to pre-store visual inspection tools of various commonly tested products.

[0013] Furthermore, the product identification module is used to perform feature matching between the collected image of the product to be tested and the pre-stored visual distinguishing features to identify the product to be tested.

[0014] According to another object of the present invention, the present invention provides a visual inspection method that can be used for multi-product inspection, comprising the following steps:

[0015] S1: Information Storage

[0016] The visual distinguishing features of various commonly tested products are stored in a commonly tested product visual distinguishing feature storage library, the corresponding detection configuration parameters of various commonly tested products are stored in a detection configuration parameter storage library, and the visual detection tools of various commonly tested products are stored in a visual detection tool storage library;

[0017] S2: Product Identification

[0018] The product to be tested is transported to the product recognition module at a uniform speed through the transport module, and the image of the product to be tested is collected by the image acquisition unit in the product recognition module, and the collected image of the product to be tested is matched with the pre-stored visual distinguishing features by the product recognition module to identify which common test product the product to be tested is;

[0019] S3: Matching parameters

[0020] After the product to be tested is identified, the detection configuration parameters corresponding to the commonly tested product are retrieved from the detection configuration parameter repository;

[0021] S4: Detection

[0022] The transport module transports the product to be tested to the visual inspection module at a uniform speed, and the positioning unit in the visual inspection module fixes the moving product in the inspection area of ​​the inspection unit. The inspection control unit calls the corresponding visual inspection tool to load the inspection configuration parameters from the inspection configuration parameter repository to the visual inspection tool according to the recognition result of the product recognition module, and performs visual inspection on the product to be tested;

[0023] S5: Classification

[0024] After the inspection of the product to be tested is completed, the positioning unit will release the product. At this time, the product will be transported to the classification module through the transportation module. The classification module receives the information transmitted by the detection control unit, and determines whether the product is defective. The defective products and qualified products are classified and stored. The automatic intelligent identification of product style and the identification of the product are carried out, and then the automatic positioning mechanism is controlled by the program to adjust the camera position for image collection, and then the inspection is carried out. At the same time, the defective products are eliminated, realizing automation without manual operation.

[0025] Furthermore, when the transport module transports the products to be tested, a distance of 80-90 cm is set between the products to be tested.

[0026] Furthermore, the corresponding detection configuration parameters of various commonly tested products are visual detection content and camera parameter settings. The visual detection content includes the appearance and color of various commonly tested products, and the camera parameter settings specifically include camera resolution and RGB gain.

[0027] Furthermore, in S2, the industrial camera and the object to be detected move relative to each other when taking pictures, and the industrial camera takes pictures at fixed intervals. The actual deviation between the pictures taken by the industrial camera and the object to be detected is Δα pixels.

[0028] Furthermore, in S4, during detection, the overlapping area of ​​the two adjacent photos in S2 is used as the calculation area of ​​the stitching algorithm, the Sobel method is used to calculate the edge strength of the two photos, and the edges are thickened and highlighted, N matching areas are taken in the calculation area of ​​the front photo to form a matching template, and the template is matched and positioned with the calculation area of ​​the adjacent rear photo to find the best matching position, where N is a natural number greater than or equal to 1; the best matching position is used as the stitching point to complete the edge stitching of the two adjacent photos; the above steps are repeated until the stitching forms the overall photo of the object to be detected.

[0029] Furthermore, in S4, for two adjacent photos, the overlapping area plus 2Δα pixels is taken as the calculation area of ​​the stitching algorithm, that is, the edges of the front photo and the back photo are each added with Δα pixels as the calculation area of ​​the stitching algorithm, and N matching areas are respectively selected in the calculation areas of the adjacent edges of the adjacent photos to form two groups of matching templates. The matching degrees of the two groups of matching templates are compared, and the best matching position is selected as the stitching point. The distribution direction of the N matching areas in the matching template is perpendicular to the overall stitching direction of the photos of the object to be tested.

[0030] The technical solution of the present invention adds a positioning unit to the visual inspection module, which can stop the passing products on the transport module so that the inspection unit can perform accurate inspection. Through this setting, not only is manual operation replaced by machinery, but there is also no need to frequently stop and start the transport module, allowing the inspection process to run more smoothly. The classification module is used to receive information transmitted by the inspection control unit, and classify and store defective products and qualified products, and can be applied to the inspection of a variety of products in batch production. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 is a flowchart of an embodiment of the present invention;

[0033] Figure 2 4 is a system structure diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0036] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0037] Example 1

[0038] like Figure 1 and Figure 2 As shown:

[0039] A visual inspection system that can be used for multiple product inspections includes a control terminal, which includes a transportation module, a data storage module, a product identification module, a visual inspection module, and a classification module.

[0040] in:

[0041] The transport module is used to transport products.

[0042] The data storage module includes a visual distinguishing feature repository for commonly tested products, a detection configuration parameter repository, and a visual detection tool repository.

[0043] The product identification module includes an image acquisition unit, and the product identification module is used to identify which common test product the product to be tested is.

[0044] The visual inspection module includes a detection control unit and a positioning unit. The detection control unit calls the corresponding visual inspection tool according to the recognition result of the product recognition module, loads the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool, performs visual inspection on the product to be tested, and passes the information to the classification module; the positioning unit is used to fix the moving product in the detection area of ​​the detection unit and release the product after the detection is completed. By adding a positioning unit to the visual inspection module, the passing product can be stopped on the transport module, so that the detection unit can perform accurate inspection. Through this setting, not only is manual operation replaced by machinery, but there is also no need to frequently stop and start the transport module, making the inspection process run more smoothly.

[0045] The classification module is used to receive information transmitted by the detection control unit, and to classify and store defective products and qualified products. It can be applied to the detection of a variety of products in batch production. The use of this system is fully flexible and suitable for manufacturers of all sizes. This system can detect large and small parts on the same production line according to different inspection requirements. The system can be designed with up to 10 professional cameras or more to automatically detect different stations and moving parts with automatic rejection devices, saving resources and improving detection efficiency.

[0046] In the present invention, a visual distinguishing feature repository for commonly tested products is used to pre-store visual distinguishing features of various commonly tested products, a detection configuration parameter repository is used to pre-store corresponding detection configuration parameters of various commonly tested products, a visual inspection tool repository is used to pre-store visual inspection tools of various commonly tested products, an image acquisition unit includes multiple cameras for multi-angle photography of products, and a product recognition module is used to perform feature matching between the acquired image of the product to be tested and the pre-stored visual distinguishing features to identify the product to be tested.

[0047] The present invention also proposes a visual inspection method that can be used for multiple product inspections, comprising the following steps:

[0048] S1: Information Storage

[0049] The visual distinguishing features of various commonly tested products are stored in a commonly tested product visual distinguishing feature storage library, the corresponding detection configuration parameters of various commonly tested products are stored in a detection configuration parameter storage library, and the visual detection tools of various commonly tested products are stored in a visual detection tool storage library;

[0050] S2: Product Identification

[0051] The product to be tested is transported to the product recognition module at a uniform speed through the transport module, and the image of the product to be tested is collected by the image acquisition unit in the product recognition module, and the collected image of the product to be tested is matched with the pre-stored visual distinguishing features by the product recognition module to identify which common test product the product to be tested is;

[0052] S3: Matching parameters

[0053] After the product to be tested is identified, the detection configuration parameters corresponding to the commonly tested product are retrieved from the detection configuration parameter repository;

[0054] S4: Detection

[0055] Then, the transport module transports the product to be tested to the visual inspection module at a uniform speed, and the positioning unit in the visual inspection module fixes the moving product in the inspection area of ​​the inspection unit. Then, the inspection control unit calls the corresponding visual inspection tool according to the recognition result of the product recognition module, loads the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool, and performs visual inspection on the product to be tested.

[0056] S5: Classification

[0057] After the inspection of the product to be tested is completed, the positioning unit will release the product. At this time, the product will be transported to the classification module through the transportation module. The classification module receives the information transmitted by the detection control unit and determines whether the product is defective. It then classifies and stores the defective products and qualified products. It can automatically and intelligently identify the product style and identify which product it is. Then, it controls the automatic positioning mechanism through the program to adjust the camera position for collecting images and then conduct inspection. At the same time, it can eliminate defective products, realizing automation without manual operation.

[0058] In particular, when the transport module transports the products to be tested, a distance of 80 cm is set between the products to be tested. The corresponding detection configuration parameters of various commonly tested products are visual detection content and camera parameter settings. The visual detection content specifically includes the appearance and main features of various commonly tested products, and the camera parameter settings specifically include camera resolution and RGB gain.

[0059] Example 2

[0060] like Figure 1 and Figure 2 As shown,

[0061] A visual inspection system that can be used for multiple product inspections includes a control terminal, which includes a transportation module, a data storage module, a product identification module, a visual inspection module, and a classification module.

[0062] in:

[0063] The transport module is used to transport products.

[0064] The data storage module includes a visual distinguishing feature repository for commonly tested products, a detection configuration parameter repository, and a visual detection tool repository.

[0065] The product identification module includes an image acquisition unit, and the product identification module is used to identify which common test product the product to be tested is.

[0066] The visual inspection module includes a detection control unit and a positioning unit. The detection control unit calls the corresponding visual inspection tool according to the recognition result of the product recognition module, loads the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool, performs visual inspection on the product to be tested, and passes the information to the classification module; the positioning unit is used to fix the moving product in the detection area of ​​the detection unit and release the product after the detection is completed. By adding a positioning unit to the visual inspection module, the passing product can be stopped on the transport module, so that the detection unit can perform accurate inspection. Through this setting, not only is manual operation replaced by machinery, but there is also no need to frequently stop and start the transport module, making the inspection process run more smoothly.

[0067] The classification module is used to receive information transmitted by the detection control unit, and to classify and store defective products and qualified products. It can be applied to the detection of a variety of products in batch production. The use of this system is fully flexible and suitable for manufacturers of all sizes. This system can detect large and small parts on the same production line according to different inspection requirements. The system can be designed with up to 10 professional cameras or more to automatically detect different stations and moving parts with automatic rejection devices, saving resources and improving detection efficiency.

[0068] In the present invention, a visual distinguishing feature repository for commonly tested products is used to pre-store visual distinguishing features of various commonly tested products, a detection configuration parameter repository is used to pre-store corresponding detection configuration parameters of various commonly tested products, a visual inspection tool repository is used to pre-store visual inspection tools of various commonly tested products, an image acquisition unit includes multiple cameras for multi-angle photography of products, and a product recognition module is used to perform feature matching between the acquired image of the product to be tested and the pre-stored visual distinguishing features to identify the product to be tested.

[0069] The present invention also proposes a visual inspection method that can be used for multiple product inspections, comprising the following steps:

[0070] S1: Information Storage

[0071] The visual distinguishing features of various commonly tested products are stored in a commonly tested product visual distinguishing feature storage library, the corresponding detection configuration parameters of various commonly tested products are stored in a detection configuration parameter storage library, and the visual detection tools of various commonly tested products are stored in a visual detection tool storage library;

[0072] S2: Product Identification

[0073] The product to be tested is transported to the product identification module at a uniform speed through the transport module, and the image of the product to be tested is collected by the image acquisition unit in the product identification module, and the collected image of the product to be tested is matched with the pre-stored visual distinguishing features by the product identification module to identify which common test product the product to be tested is; the image acquisition unit takes a photo of the item to be tested, and ensures that two adjacent photos have overlapping areas;

[0074] When taking photos in S2, the industrial camera and the object to be detected move relative to each other. The industrial camera takes photos at fixed intervals. The actual deviation between the photos taken by the industrial camera and the object to be detected is Δα pixels.

[0075] S3: Matching parameters

[0076] After the product to be tested is identified, the detection configuration parameters corresponding to the commonly tested product are retrieved from the detection configuration parameter repository;

[0077] S4: Detection

[0078] Then, the transport module transports the product to be tested to the visual inspection module at a uniform speed, and the positioning unit in the visual inspection module fixes the moving product in the inspection area of ​​the inspection unit. Then, the inspection control unit calls the corresponding visual inspection tool according to the recognition result of the product recognition module, loads the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool, and performs visual inspection on the product to be tested.

[0079] In S4, during detection, the overlapping area of ​​the two adjacent photos in S2 is used as the calculation area of ​​the stitching algorithm, the edge strength of the two photos is calculated using the Sobel method, and the edges are bolded and highlighted. N matching areas are taken from the calculation area of ​​the front photo to form a matching template, which is matched and positioned with the calculation area of ​​the adjacent rear photo to find the best matching position, where N is a natural number greater than or equal to 1; the best matching position is used as the stitching point to complete the edge stitching of the two adjacent photos; the above steps are repeated until the stitching forms the overall photo of the object to be detected;

[0080] For two adjacent photos in S4, the overlapping area plus 2Δα pixels is taken as the calculation area of ​​the stitching algorithm, that is, the edges of the front photo and the back photo are each added with Δα pixels as the calculation area of ​​the stitching algorithm, and N matching areas are respectively selected in the calculation areas of the adjacent edges of the adjacent photos to form two groups of matching templates, and the matching degrees of the two groups of matching templates are compared, and the best matching position is selected as the stitching point. The distribution direction of the N matching areas in the matching template is perpendicular to the overall stitching direction of the photos of the object to be tested;

[0081] S5: Classification

[0082] After the inspection of the product to be tested is completed, the positioning unit will release the product. At this time, the product will be transported to the classification module through the transportation module. The classification module receives the information transmitted by the detection control unit and determines whether the product is defective. It then classifies and stores the defective products and qualified products. It can automatically and intelligently identify the product style and identify which product it is. Then, it controls the automatic positioning mechanism through the program to adjust the camera position for collecting images and then conduct inspection. At the same time, it can eliminate defective products, realizing automation without manual operation.

[0083] In particular, when the transport module transports the products to be tested, a distance of 90 cm is set between the products to be tested. The corresponding detection configuration parameters of various commonly tested products are visual detection content and camera parameter settings. The visual detection content specifically includes the appearance and color of various commonly tested products, and the camera parameter settings specifically include camera resolution and exposure time.

[0084] Example 3

[0085] like Figure 1 and Figure 2 As shown,

[0086] A visual inspection system that can be used for multiple product inspections includes a control terminal, which includes a transportation module, a data storage module, a product identification module, a visual inspection module, and a classification module.

[0087] in:

[0088] The transport module is used to transport products.

[0089] The data storage module includes a visual distinguishing feature repository for commonly tested products, a detection configuration parameter repository, and a visual detection tool repository.

[0090] The product identification module includes an image acquisition unit, and the product identification module is used to identify which common test product the product to be tested is.

[0091] The visual inspection module includes a detection control unit and a positioning unit. The detection control unit calls the corresponding visual inspection tool according to the recognition result of the product recognition module, loads the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool, performs visual inspection on the product to be tested, and passes the information to the classification module; the positioning unit is used to fix the moving product in the detection area of ​​the detection unit and release the product after the detection is completed. By adding a positioning unit to the visual inspection module, the passing product can be stopped on the transport module, so that the detection unit can perform accurate inspection. Through this setting, not only is manual operation replaced by machinery, but there is also no need to frequently stop and start the transport module, making the inspection process run more smoothly.

[0092] The classification module is used to receive information transmitted by the detection control unit, and classify and store defective products and qualified products. It can be applied to the detection of a variety of products in batch production. The use of this system is fully flexible and suitable for manufacturers of various sizes. This system can detect large and small parts on the same production line at the same time according to different inspection requirements. The system can be designed with up to 10 professional cameras or more to automatically detect different workstations and moving parts and is equipped with an automatic rejection device, which saves resources and improves detection efficiency.

[0093] In the present invention, a visual distinguishing feature repository for commonly tested products is used to pre-store visual distinguishing features of various commonly tested products, a detection configuration parameter repository is used to pre-store corresponding detection configuration parameters of various commonly tested products, a visual inspection tool repository is used to pre-store visual inspection tools of various commonly tested products, an image acquisition unit includes multiple cameras for multi-angle photography of products, and a product recognition module is used to perform feature matching between the acquired image of the product to be tested and the pre-stored visual distinguishing features to identify the product to be tested.

[0094] The present invention also proposes a visual inspection method that can be used for multiple product inspections, comprising the following steps:

[0095] S1: Information Storage

[0096] The visual distinguishing features of various commonly tested products are stored in a commonly tested product visual distinguishing feature storage library, the corresponding detection configuration parameters of various commonly tested products are stored in a detection configuration parameter storage library, and the visual detection tools of various commonly tested products are stored in a visual detection tool storage library;

[0097] S2: Product Identification

[0098] The product to be tested is transported to the product recognition module at a uniform speed through the transport module, and the image of the product to be tested is collected by the image acquisition unit in the product recognition module, and the collected image of the product to be tested is matched with the pre-stored visual distinguishing features by the product recognition module to identify which common test product the product to be tested is;

[0099] S3: Matching parameters

[0100] After the product to be tested is identified, the detection configuration parameters corresponding to the commonly tested product are retrieved from the detection configuration parameter repository;

[0101] S4: Detection

[0102] Then, the transport module transports the product to be tested to the visual inspection module at a uniform speed, and the positioning unit in the visual inspection module fixes the moving product in the inspection area of ​​the inspection unit. Then, the inspection control unit calls the corresponding visual inspection tool according to the recognition result of the product recognition module, loads the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool, and performs visual inspection on the product to be tested.

[0103] S5: Classification

[0104] After the inspection of the product to be tested is completed, the positioning unit will release the product. At this time, the product will be transported to the classification module through the transportation module. The classification module receives the information transmitted by the detection control unit and determines whether the product is defective. It then classifies and stores the defective products and qualified products. It can automatically and intelligently identify the product style and identify which product it is. Then, it controls the automatic positioning mechanism through the program to adjust the camera position for collecting images and then conduct inspection. At the same time, it can eliminate defective products, realizing automation without manual operation.

[0105] In particular, when the transport module transports the products to be tested, a distance of 80 cm is set between the products to be tested. The corresponding detection configuration parameters of various commonly tested products are visual detection content and camera parameter settings. The visual detection content specifically includes the appearance, color and main features of various commonly tested products. The camera parameter settings specifically include camera resolution, exposure time and RGB gain.

[0106] The present invention adds a positioning unit to the visual inspection module, which can stop the passing products on the transport module so that the inspection unit can perform accurate inspection. Through this arrangement, not only is manual operation replaced by machinery, but there is also no need to frequently stop and start the transport module, allowing the inspection process to run more smoothly. The classification module is used to receive information transmitted by the inspection control unit, and classify and store defective products and qualified products. It can be applied to the inspection of a variety of products produced in batches. The use of this system is fully flexible and suitable for manufacturers of various sizes. This system can inspect large and small parts on the same production line at the same time according to different inspection requirements. The system can be designed with up to 10 professional cameras or more to automatically inspect different workstations and moving parts and is equipped with an automatic rejection device, saving resources and improving inspection efficiency.

[0107] The present invention can be applied to the detection of a variety of products produced in batches. The use of the system is fully flexible and suitable for manufacturers of various sizes. The system can detect large and small parts according to different inspection requirements on the same production line at the same time. The system can be designed with up to 10 professional cameras or more to automatically detect different workstations and moving parts, saving resources and improving detection efficiency.

[0108] The present invention can automatically and intelligently identify the style of a product, and after identifying which product it is, it can control the automatic positioning mechanism through a program to adjust the camera position for collecting images, and then perform detection. At the same time, defective products can be eliminated, thus achieving automation without manual operation.

[0109] The present invention adds a positioning unit to the visual inspection module, which can stop the passing products on the transport module so that the inspection unit can perform accurate inspection. Through this arrangement, not only is manual operation replaced by machinery, but there is also no need to frequently stop and start the transport module, allowing the inspection process to run more smoothly.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A visual inspection system that can be used for multiple product inspections, characterized in that: A control terminal is included, and the control terminal includes: A transport module, wherein the transport module is used to transport the product; A data storage module, the data storage module includes a visual distinguishing feature storage library for commonly tested products, a detection configuration parameter storage library, and a visual inspection tool storage library, which are respectively used to store visual distinguishing features of various commonly tested products, corresponding detection configuration parameters of various commonly tested products, and visual inspection tools of various commonly tested products; A product identification module, the product identification module is used to identify which common test product the product to be tested is; A visual inspection module, the visual inspection module comprising a detection control unit and a positioning unit. The detection control unit calls the corresponding visual inspection tool to load the detection configuration parameters from the detection configuration parameter repository to the visual inspection tool according to the recognition result of the product recognition module, performs visual inspection on the product to be inspected, and transmits the information to the classification module; the positioning unit is used to fix the moving product in the inspection area of ​​the detection unit and release the product after the inspection is completed; The classification module is used to receive information transmitted by the detection control unit, and classify and store defective products and qualified products, and is suitable for detecting a variety of products produced in batches.

2. The visual inspection system for multiple product inspection according to claim 1, characterized in that: The product recognition module includes an image acquisition unit, and the image acquisition unit includes a plurality of cameras for taking multi-angle images of the product.

3. The visual inspection system for multiple product inspection according to claim 1, characterized in that: The commonly tested product visual distinguishing feature repository is used to pre-store visual distinguishing features of various commonly tested products, the detection configuration parameter repository is used to pre-store corresponding detection configuration parameters of various commonly tested products, and the visual detection tool repository is used to pre-store visual detection tools of various commonly tested products.

4. The visual inspection system for multiple product inspection according to claim 1, characterized in that: The product identification module is used to perform feature matching between the collected image of the product to be tested and the pre-stored visual distinguishing features to identify the product to be tested.

5. The method for using the visual inspection system for multiple product inspection according to claim 1, characterized in that: The following steps are involved: S1: Information Storage The visual distinguishing features of various commonly tested products are stored in a commonly tested product visual distinguishing feature storage library, the corresponding detection configuration parameters of various commonly tested products are stored in a detection configuration parameter storage library, and the visual detection tools of various commonly tested products are stored in a visual detection tool storage library; S2: Product Identification The product to be tested is transported to the product recognition module at a uniform speed through the transport module, and the image of the product to be tested is collected by the image acquisition unit in the product recognition module, and the collected image of the product to be tested is matched with the pre-stored visual distinguishing features by the product recognition module to identify which common test product the product to be tested is; S3: Matching parameters After the product to be tested is identified, the detection configuration parameters corresponding to the commonly tested product are retrieved from the detection configuration parameter repository; S4: Detection The transport module transports the product to be tested to the visual inspection module at a uniform speed, and the positioning unit in the visual inspection module fixes the moving product in the inspection area of ​​the inspection unit. The inspection control unit calls the corresponding visual inspection tool to load the inspection configuration parameters from the inspection configuration parameter repository to the visual inspection tool according to the recognition result of the product recognition module, and performs visual inspection on the product to be tested; S5: Classification After the inspection of the product to be tested is completed, the positioning unit will release the product. At this time, the product will be transported to the classification module through the transportation module. The classification module receives the information transmitted by the detection control unit, and determines whether the product is defective. The defective products and qualified products are classified and stored. The automatic intelligent identification of product style and the identification of the product are carried out, and then the automatic positioning mechanism is controlled by the program to adjust the camera position for image collection, and then the inspection is carried out. At the same time, the defective products are eliminated, realizing automation without manual operation.

6. The method for using the visual inspection system for multiple product inspection according to claim 5, characterized in that: When the transport module transports the products to be tested, a distance of 80-90 cm is set between the products to be tested.

7. The method for using the visual inspection system for multiple product inspection according to claim 5, characterized in that: The corresponding detection configuration parameters of various commonly tested products are visual inspection content and camera parameter settings. The visual inspection content includes the appearance and color of various commonly tested products, and the camera parameter settings specifically include camera resolution and RGB gain.

8. The method for using the visual inspection system for multiple product inspection according to claim 5, characterized in that: In S2, the industrial camera and the object to be detected move relative to each other when taking pictures. The industrial camera takes pictures at fixed intervals. The actual deviation between the pictures taken by the industrial camera and the object to be detected is Δα pixels.

9. The method for using the visual inspection system for multiple product inspection according to claim 5, characterized in that: In S4, during detection, the overlapping area of ​​the two adjacent photos in S2 is used as the calculation area of ​​the stitching algorithm, the Sobel method is used to calculate the edge strength of the two photos, and the edges are thickened and highlighted. N matching areas are taken in the calculation area of ​​the front photo to form a matching template, which is matched and positioned with the calculation area of ​​the adjacent rear photo to find the best matching position, where N is a natural number greater than or equal to 1; the best matching position is used as the stitching point to complete the edge stitching of the two adjacent photos; the above steps are repeated until the stitching forms the overall photo of the object to be detected.

10. The method for using the visual inspection system for multiple product inspection according to claim 5, characterized in that: In S4, for two adjacent photos, the overlapping area plus 2Δα pixels is taken as the calculation area of ​​the stitching algorithm, that is, the edges of the front photo and the back photo are each added with Δα pixels as the calculation area of ​​the stitching algorithm, and N matching areas are respectively selected in the calculation areas of the adjacent edges of the adjacent photos to form two groups of matching templates. The matching degrees of the two groups of matching templates are compared, and the best matching position is selected as the stitching point. The distribution direction of the N matching areas in the matching template is perpendicular to the overall stitching direction of the photos of the object to be tested.

Images