CN118849098A - Automatic optical detection method, system, device and storage medium for die-cutting machine - Google Patents

Abstract

The application relates to the technical field of image recognition and discloses an automatic optical detection method, an automatic optical detection system, automatic optical detection equipment and a storage medium applied to a die cutting machine, wherein the method comprises the steps of determining workpiece shape and position parameters, raw material models and processing types of processing tasks, and setting an image cutting area to update an image cutting algorithm; before the workpiece is pressed and cut, a first optical detection image is acquired and input into an image preprocessing model, and correction cutting processing is carried out to generate a first correction image; intercepting a stage finished product area image from the first corrected image, comparing the stage finished product area image with a corresponding raw material reference image, generating feeding specification information, and generating a pressing and cutting starting instruction when the feeding specification information is qualified; after the workpiece is pressed and cut, a second optical detection image is obtained, a second correction image is generated through preprocessing, and the phase finished product pattern and the phase finished product boundary are intercepted from the second correction image and are compared with the corresponding finished product reference image and boundary reference image, so that finished product specification information is generated; the application has the effect of improving the detection timeliness of the die-cutting quality problem.

Description

Automatic optical detection method, system, device and storage medium for die-cutting machine

Technical Field

The present application relates to the technical field of image recognition, and in particular to an automatic optical detection method, system, device and storage medium applied to a die-cutting machine.

Background Art

Die-cutting machines are widely used for cutting workpieces such as plates and sheets, and can also be used for manufacturing processes such as indentation, hot stamping, and material bonding. During the use of die-cutting machines, common processing quality problems include inaccurate positioning of workpieces, deformation of workpieces, damage to tools, dirty plates and sheets to be processed, etc. At present, the detection method for such quality problems is usually determined after inspecting the finished workpieces, and the discovery of quality problems has a large lag, making it difficult to timely discover quality problems and maintain the equipment during the processing process. Therefore, the above-mentioned related technologies have the problem of poor timeliness in detecting die-cutting quality problems.

Summary of the invention

In order to improve the timeliness of the detection of die-cutting quality problems, the present application provides an automatic optical detection method, system, device and storage medium applied to a die-cutting machine.

The first invention objective of this application is achieved by adopting the following technical solution:

Automatic optical inspection methods applied to die-cutting machines include:

Determine the workpiece shape and position parameters, raw material model, and processing type of the current processing task, and set the image cropping area of the current processing task based on the workpiece shape and position parameters to update the image cropping algorithm;

Before the workpiece cutting begins, a first optical detection image of the cutting surface is obtained and input into a preset image preprocessing model, and correction and cropping are performed to generate a first corrected image;

Based on the workpiece shape and position parameters, a stage finished product area image is intercepted from the first corrected image, the stage finished product area image is compared with a raw material reference image corresponding to the raw material model, and feeding specification information is generated, and when the feeding specification information is qualified, a pressing and cutting start instruction is generated;

After the workpiece is pressed and cut, a second optical detection image is obtained, and a second corrected image is generated after preprocessing. A stage finished product pattern and a stage finished product boundary are intercepted from the second corrected image based on the workpiece shape and position parameters, and the stage finished product pattern and the stage finished product boundary are compared with the finished product reference image and the boundary reference image corresponding to the processing type to generate finished product specification information;

The die-cutting machine is provided with an imaging component for photographing the pressed and cut plate to obtain an optical detection image; the image preprocessing model is built with a perspective correction algorithm and an image cropping algorithm.

By adopting the above technical scheme, the workpiece shape and position parameters, raw material model, and processing type of the current die-cutting machine processing task are determined, so as to subsequently determine the image cropping area and reduce the computer resource consumption for image recognition of invalid areas; before the workpiece pressing and cutting begins, the first optical detection image of the pressing and cutting layout is obtained and input into the image preprocessing model, so as to perform perspective correction and image cropping on the first optical detection image to obtain a first corrected image, which is convenient for subsequent image comparison; according to the workpiece shape and position parameters, the stage finished product area image is intercepted from the first corrected image, and compared with the raw material reference image corresponding to the raw material model, so as to generate feeding specification information, and when the feeding specification information is qualified, the pressing and cutting start instruction is generated, so as to facilitate the suspension of the die-cutting machine in the case of unqualified raw materials to eliminate faults; after the workpiece is pressed and cut, the second optical detection image is obtained and preprocessed to obtain the second corrected image, and the stage finished product pattern and stage finished product boundary are intercepted from the second corrected image according to the workpiece shape and position parameters, and compared with the corresponding finished product reference image and boundary reference image, so as to generate finished product specification information, which is used to judge whether the finished product quality is qualified, thereby improving the timeliness of die-cutting quality problem detection.

In a preferred example of the present application, the determining of the workpiece shape and position parameters, material model, and processing type of the current processing task, and setting the image cropping area of the current processing task based on the workpiece shape and position parameters to update the image cropping algorithm includes:

Receive processing setting information to set the processing program of the target die-cutting machine, and determine the workpiece shape and position parameters, raw material model, and processing type of the current processing task based on the processing setting information;

The shape information, size information and position information relative to the cutting surface of the finished product of the workpiece stage of the current processing task are determined based on the workpiece shape and position parameters, and the image cropping area of the current processing task is set to update the image cropping algorithm.

By adopting the above technical solution, processing setting information is received to set the processing program of the target die-cutting machine, and the workpiece shape and position parameters, raw material model, processing type and other information corresponding to the current processing task are determined according to the processing setting information; the shape, size and position of the workpiece stage finished product of the current processing task relative to the cutting surface are determined based on the workpiece shape and position parameters, and then the image cropping area of the current processing task is set to update the image cropping algorithm, so as to facilitate the subsequent use of the image cropping algorithm to crop the image taken from the cutting surface and optimize the image comparison efficiency.

In a preferred example of the present application, before the workpiece cutting begins, a first optical detection image of the cutting surface is obtained and input into a preset image preprocessing model, and correction and cropping are performed to generate a first corrected image, including:

When the raw material is delivered to the place and before the workpiece cutting begins, a first optical detection image taken by the imaging component is acquired and input into the image preprocessing model;

Obtaining the viewing angle information of the imaging component and the shooting orientation information of the lens axis relative to the pressed-cutting layout and loading them into the perspective correction algorithm, scaling the first optical detection image in different directions according to different scaling ratios through the perspective correction algorithm to generate a first scaled image;

An image cropping region is identified from the first zoomed image, and the first zoomed image is cropped based on the image cropping region to generate a first corrected image.

By adopting the above technical scheme, when the raw material is delivered to the place and before the workpiece cutting begins, the first optical detection image of the cutting surface taken by the imaging component is obtained and input into the image preprocessing model for subsequent preprocessing; the viewing angle information and shooting orientation information of the imaging component are obtained and loaded into the perspective correction algorithm to determine the image deformation caused by the perspective principle of the first optical detection image, and the first optical detection image is scaled in different directions at different scaling ratios by the perspective correction algorithm to generate a first scaled image, thereby reducing the image distortion caused by the perspective principle; the image cropping area is identified from the first scaled image to crop the first scaled image, thereby generating a first corrected image, which is convenient for subsequent image recognition and comparison.

In a preferred example of the present application, based on the workpiece shape and position parameters, the stage finished product area image is intercepted from the first corrected image, the stage finished product area image is compared with the raw material reference image corresponding to the raw material model, and the feed specification information is generated. When the feed specification information is qualified, a press-cut start instruction is generated, including:

Based on the shape and position parameters of the workpiece, the stage finished product area is determined according to the boundary position of the stage finished product of each workpiece, and the stage finished product area image is intercepted from the first corrected image;

Matching a corresponding raw material reference image from a preset comparison image library based on the raw material model;

Compare the stage finished product area image with the raw material reference image to determine whether there are abnormal image features in the stage finished product area image. If there are no abnormal image features, generate qualified feeding specification information, and generate a pressing and cutting start instruction and send it to the die-cutting controller;

The stage finished product area image completely covers the stage finished product pattern and the area where the stage finished product boundary is located.

By adopting the above technical scheme, the boundary position of each stage finished product of the workpiece is determined based on the shape and position parameters of the workpiece, and then the area of the stage finished product of the workpiece in the first corrected image is determined to intercept the stage finished product area image; the image of the sheet raw material that meets the raw material model is matched from the preset comparison image library as the raw material reference image; the stage finished product area image is compared with the raw material reference image to determine whether there are abnormal image features in the stage finished product area image that are inconsistent with the raw material reference image. When no abnormal image features are found, it is considered that the quality of the sheet raw material in the stage finished product area image is qualified, so as to generate qualified feeding specification information, and generate a pressing and cutting start instruction and send it to the die-cutting controller to control the die-cutting machine to start die-cutting work.

In a preferred example of the present application, the method of intercepting a stage finished product pattern and a stage finished product boundary from the second corrected image based on the workpiece shape and position parameters, comparing the stage finished product pattern and the stage finished product boundary with the finished product reference image and the boundary reference image corresponding to the processing type, and generating finished product specification information includes:

Based on the shape and position parameters of the workpiece, the area where the staged product and the die-cutting boundary of the workpiece are located in the second corrected image is determined, so as to intercept the image of the corresponding area from the second corrected image to obtain the staged product pattern and the staged product boundary;

Based on the known raw material model and processing type, the corresponding finished product reference image and boundary reference image are matched from the preset comparison image library;

The stage finished product pattern is compared with the finished product reference image, and the stage finished product boundary is compared with the boundary reference image. If there are no abnormal image features in both, qualified finished product specification information is generated.

By adopting the above technical scheme, based on the workpiece shape and position parameters, the area where the workpiece stage finished product and the die-cutting boundary are located in the second corrected image is determined, so as to intercept the stage finished product pattern according to the area where the workpiece stage finished product is located in the second corrected image, and intercept the stage finished product boundary according to the area where the die-cutting boundary is located; based on the known raw material model and processing type, the workpiece stage finished product image and the die-cutting boundary image of the successful sample of the die-cutting work under the same working conditions are matched from the comparison image library as the finished product reference image and the boundary reference image; the stage finished product pattern is compared with the finished product reference image, and the stage finished product boundary is compared with the boundary reference image to determine whether the quality of the finished product of this die-cutting work is qualified. If no abnormal image features appear in the comparison process of the two images, qualified finished product specification information is generated, so as to facilitate the understanding that the current workpiece stage finished product processing quality is qualified.

The second invention objective of this application is achieved by the following technical solution:

An automatic optical inspection system for a die-cutting machine, applied to any of the above-mentioned automatic optical inspection methods for a die-cutting machine, comprising:

An image cropping algorithm update module is used to determine the workpiece shape and position parameters, raw material model, and processing type of the current processing task, and to set the image cropping area of the current processing task based on the workpiece shape and position parameters to update the image cropping algorithm;

A first corrected image generation module is used to obtain a first optical detection image of the pressing and cutting surface before the workpiece pressing and cutting begins, and input it into a preset image preprocessing model to perform correction and cropping processing to generate a first corrected image;

A pressing and cutting start instruction generating module is used to intercept a stage finished product area image from the first corrected image based on the workpiece shape and position parameters, compare the stage finished product area image with a raw material reference image corresponding to the raw material model, generate feed specification information, and generate a pressing and cutting start instruction when the feed specification information is qualified;

The finished product specification information generating module is used to obtain the second optical detection image after the workpiece is pressed and cut, generate the second corrected image through preprocessing, intercept the stage finished product pattern and the stage finished product boundary from the second corrected image based on the workpiece shape and position parameters, compare the stage finished product pattern and the stage finished product boundary with the finished product reference image and the boundary reference image corresponding to the processing type, and generate the finished product specification information;

The die-cutting machine is provided with an imaging component for photographing the pressed and cut plate to obtain an optical detection image; the image preprocessing model is built with a perspective correction algorithm and an image cropping algorithm.

In a preferred example of the present application: the image cropping algorithm update module includes:

The processing task analysis submodule is used to receive processing setting information to set the processing program of the target die-cutting machine, and determine the workpiece shape and position parameters, raw material model, and processing type of the current processing task based on the processing setting information;

The image cropping area setting submodule is used to determine the shape information, size information and position information relative to the cutting surface of the workpiece stage finished product of the current processing task based on the workpiece shape and position parameters, set the image cropping area of the current processing task, and update the image cropping algorithm.

In a preferred example of the present application: the first corrected image generation module includes:

A first optical detection image acquisition submodule is used to acquire a first optical detection image taken by the imaging component and input it into an image preprocessing model when the raw material is delivered to the site and before the workpiece pressing and cutting begins;

A first zoomed image generation submodule is used to obtain the viewing angle information of the imaging component and the shooting orientation information of the lens axis relative to the pressing and cutting surface and load them into the perspective correction algorithm, and to perform zooming processing on the first optical detection image in different directions according to different zooming ratios through the perspective correction algorithm to generate a first zoomed image;

The scaled image cropping processing submodule is used to identify an image cropping area from the first scaled image, perform cropping processing on the first scaled image based on the image cropping area, and generate a first corrected image.

The third invention objective of this application is achieved by the following technical solution:

A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the automatic optical detection method applied to a die-cutting machine when executing the computer program.

The fourth invention objective of this application is achieved by the following technical solution:

A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the automatic optical detection method applied to a die-cutting machine are implemented.

In summary, the present application includes at least one of the following beneficial technical effects:

1. Determine the workpiece shape and position parameters, raw material model, and processing type of the current die-cutting machine processing task, so as to subsequently determine the image cropping area and reduce the computer resource consumption for image recognition of invalid areas; before the workpiece pressing and cutting begins, obtain the first optical detection image of the pressing and cutting layout and input it into the image preprocessing model, so as to perform perspective correction and image cropping on the first optical detection image to obtain a first corrected image, which is convenient for subsequent image comparison; intercept the stage finished product area image from the first corrected image according to the workpiece shape and position parameters, and compare it with the raw material reference image corresponding to the raw material model, so as to generate feeding specification information, and when the feeding specification information is qualified, generate a pressing and cutting start instruction, so as to facilitate the suspension of the die-cutting machine in the case of unqualified raw materials to eliminate faults; after the workpiece is pressed and cut, obtain the second optical detection image and perform preprocessing to obtain the second corrected image, intercept the stage finished product pattern and stage finished product boundary from the second corrected image according to the workpiece shape and position parameters, and compare it with the corresponding finished product reference image and boundary reference image, so as to generate finished product specification information, which is used to judge whether the finished product quality is qualified, thereby improving the timeliness of die-cutting quality problem detection.

2. Receive processing setting information to set the processing program of the target die-cutting machine, and determine the workpiece shape and position parameters, raw material model, processing type and other information corresponding to the current processing task based on the processing setting information; determine the shape, size and position of the workpiece stage finished product of the current processing task relative to the cutting surface based on the workpiece shape and position parameters, and then set the image cropping area of the current processing task to update the image cropping algorithm, so as to facilitate the subsequent use of the image cropping algorithm to crop the image taken from the cutting surface and optimize the image comparison efficiency.

3. When the raw material is delivered to the site and before the workpiece cutting begins, the first optical detection image of the cutting surface taken by the imaging component is obtained and input into the image preprocessing model for subsequent preprocessing; the viewing angle information and shooting orientation information of the imaging component are obtained and loaded into the perspective correction algorithm to determine the image deformation caused by the perspective principle of the first optical detection image, and the first optical detection image is scaled in different directions at different scaling ratios through the perspective correction algorithm to generate a first scaled image to reduce the image distortion caused by the perspective principle; the image cropping area is identified from the first scaled image to crop the first scaled image, thereby generating a first corrected image to facilitate subsequent image recognition and comparison.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flow chart of an automatic optical detection method applied to a die-cutting machine in Embodiment 1 of the present application.

FIG. 2 is a principle block diagram of an automatic optical inspection system applied to a die-cutting machine in a second embodiment of the present application.

FIG3 is a schematic diagram of the equipment in Example 3 of the present application.

DETAILED DESCRIPTION

The present application is further described in detail below in conjunction with Figures 1 to 3.

Embodiment 1

Referring to FIG. 1 , the present application discloses an automatic optical detection method applied to a die-cutting machine, which specifically includes the following steps:

S10: Determine the workpiece shape and position parameters, material model, and processing type of the current processing task, and set the image cropping area of the current processing task based on the workpiece shape and position parameters to update the image cropping algorithm.

In this embodiment, a processing task refers to a production task of a workpiece stage finished product; the workpiece stage finished product refers to the finished product corresponding to the current processing task process; the workpiece shape and position parameters include the shape parameters, size parameters and position parameters of the workpiece stage finished product relative to the die-cutting machine processing surface.

Specifically, the workpiece shape and position parameters, raw material model, and processing type of the current die-cutting machine processing task are determined so as to subsequently determine the image cropping area and reduce the computer resource consumption for image recognition of invalid areas.

Wherein, in step S10, it includes:

S11: receiving processing setting information to set the processing program of the target die-cutting machine, and determining the workpiece shape and position parameters, raw material model, and processing type of the current processing task based on the processing setting information.

In this embodiment, processing setting information refers to information input by the operator of the die-cutting machine to the die-cutting machine for setting the die-cutting machine processing program to control the operation of the die-cutting machine, including workpiece shape and position parameters, raw material model, and processing type; raw material model refers to the model of the raw material required to correspond to the processing task, which may be specifically composed of characteristic information such as material, size, and appearance; processing type refers to the working mode of the die-cutting machine, including cutting, creasing, laminating, hot stamping, etc.

Specifically, the processing setting information is received to set the processing program of the target die-cutting machine, and the workpiece shape and position parameters, raw material model and processing type corresponding to the current processing task are determined according to the processing setting information, so as to facilitate the subsequent matching of the corresponding comparison images and quality judgment rules.

S12: determining the shape information, size information and position information relative to the cutting surface of the finished product of the workpiece stage of the current processing task based on the workpiece shape and position parameters, setting the image cropping area of the current processing task to update the image cropping algorithm.

Specifically, the shape, size and position of the finished workpiece stage of the current processing task relative to the cutting surface are determined based on the workpiece shape and position parameters, and then the image cropping area of the current processing task is set to update the image cropping algorithm, so as to facilitate the subsequent use of the image cropping algorithm to crop the images taken on the cutting surface and optimize the image comparison efficiency.

S20: Before the workpiece cutting begins, a first optical detection image of the cutting surface is obtained and input into a preset image preprocessing model, and correction and cropping are performed to generate a first corrected image.

In this embodiment, the die-cutting machine is provided with an imaging component for photographing the pressed and cut plate to obtain an optical detection image; the image preprocessing model has a built-in perspective correction algorithm and an image cropping algorithm for preprocessing the image photographed by the die-cutting machine imaging component.

Specifically, before the workpiece cutting begins, a first optical detection image of the cutting surface is acquired and input into an image preprocessing model so as to perform perspective correction and image cropping on the first optical detection image to obtain a first corrected image for subsequent image comparison.

Wherein, in step S20, it includes:

S21: When the raw material is delivered to the right place and before the workpiece cutting begins, a first optical detection image taken by the imaging component is acquired and input into the image preprocessing model.

Specifically, when the raw material is delivered to the desired position and before the workpiece cutting begins, a first optical detection image of the cutting surface captured by the imaging component is acquired and input into the image preprocessing model for subsequent preprocessing.

S22: Obtain the viewing angle information of the imaging component and the shooting orientation information of the lens axis relative to the cut surface and load them into the perspective correction algorithm, and use the perspective correction algorithm to scale the first optical detection image in different directions according to different scaling ratios to generate a first scaled image.

In this embodiment, the die-cutting surface refers to the plane on the die-cutting machine used to place raw materials, finished products and other sheet materials; the shooting orientation information specifically includes the imaging cone parameters of the imaging component, the relative position and relative direction parameters of the lens axis and the die-cutting surface; the first zoomed image refers to the image generated after the optical detection image is perspective corrected.

Specifically, the viewing angle information and shooting orientation information of the imaging component are obtained and loaded into the perspective correction algorithm to determine the image deformation of the first optical detection image caused by the perspective principle. The first optical detection image is scaled in different directions at different scaling ratios through the perspective correction algorithm to generate a first scaled image, thereby reducing the image distortion caused by the perspective principle.

S23: identifying an image cropping region from the first zoomed image, and performing cropping processing on the first zoomed image based on the image cropping region to generate a first corrected image.

In this embodiment, the first corrected image refers to an image generated after performing image cropping processing on the first zoomed image.

Specifically, an image cropping region is identified from the first zoomed image to perform cropping processing on the first zoomed image, thereby generating a first corrected image to facilitate subsequent image recognition and comparison.

S30: based on the workpiece shape and position parameters, a stage finished product area image is intercepted from the first corrected image, the stage finished product area image is compared with the raw material reference image corresponding to the raw material model, and feeding specification information is generated. When the feeding specification information is qualified, a pressing and cutting start instruction is generated.

In this embodiment, the raw material reference image refers to the image obtained by photographing the qualified raw material on the die-cutting machine, which is used to subsequently determine whether the raw material supplied in the actual processing process is dirty, damaged, erroneous, etc.; the pressing and cutting start instruction refers to the instruction sent to the die-cutting machine to control the die-cutting machine to perform pressing and cutting.

Specifically, the finished product area image is captured from the first corrected image based on the workpiece shape and position parameters, and compared with the raw material reference image corresponding to the raw material model, so as to generate feeding specification information. When the feeding specification information is qualified, the pressing and cutting start instruction is generated, so as to suspend the operation of the die-cutting machine to troubleshoot the problem when the raw material is unqualified.

Wherein, in step S30, it includes:

S31: Based on the shape and position parameters of the workpiece, the stage finished product area is determined according to the boundary position of each stage finished product of the workpiece, and the stage finished product area image is intercepted from the first corrected image.

In this embodiment, the staged finished product area image completely covers the staged finished product pattern and the area where the staged finished product boundary is located.

Specifically, the boundary position of each stage product of the workpiece is determined based on the workpiece shape and position parameters, and then the area of the stage product of the workpiece in the first corrected image is determined to intercept the stage product area image.

S32: Matching a corresponding raw material reference image from a preset comparison image library based on the raw material model.

In this embodiment, the comparison image library refers to a database for storing various reference images as standard images for image comparison.

Specifically, an image of a sheet material that meets the material model is matched from a preset comparison image library as a material reference image.

S33: Compare the stage finished product area image with the raw material reference image to determine whether there are abnormal image features in the stage finished product area image. If there are no abnormal image features, generate qualified feeding specification information and generate a pressing and cutting start instruction and send it to the die-cutting controller.

Specifically, the stage finished product area image is compared with the raw material reference image to determine whether there are abnormal image features in the stage finished product area image that are inconsistent with the raw material reference image. When no abnormal image features are found, it is considered that the image quality of the sheet raw material in the stage finished product area is qualified, so as to generate qualified feeding specification information, and generate a pressing and cutting start instruction and send it to the die-cutting controller to control the die-cutting machine to start die-cutting work.

S40: After the workpiece is pressed and cut, a second optical detection image is obtained, which is preprocessed to generate a second corrected image. The stage finished product pattern and the stage finished product boundary are intercepted from the second corrected image based on the workpiece shape and position parameters. The stage finished product pattern and the stage finished product boundary are compared with the finished product reference image and the boundary reference image corresponding to the processing type to generate finished product specification information.

In this embodiment, the second optical detection image and the second corrected image are acquired in the same manner as the first optical detection image and the first corrected image.

Specifically, after the workpiece is press-cut, a second optical inspection image of the press-cut surface is obtained and input into a preset image preprocessing model for perspective correction, image cropping and other preprocessing to obtain a second corrected image. The stage finished product pattern and stage finished product boundary are intercepted from the second corrected image according to the workpiece shape and position parameters, and compared with the corresponding finished product reference image and boundary reference image to generate finished product specification information for judging whether the finished product quality is qualified, thereby improving the timeliness of die-cutting quality problem detection.

Wherein, in step S40, it includes:

S41: Based on the shape and position parameters of the workpiece, determine the area where the staged product and the die-cutting boundary of the workpiece are located in the second corrected image, so as to intercept the image of the corresponding area from the second corrected image to obtain the staged product pattern and the staged product boundary.

Specifically, based on the workpiece shape and position parameters, the areas where the workpiece stage finished product and the die-cutting boundary are located in the second corrected image are determined, so as to intercept the stage finished product pattern according to the area where the workpiece stage finished product is located in the second corrected image, and intercept the stage finished product boundary according to the area where the die-cutting boundary is located.

S42: Based on the known raw material model and processing type, the corresponding finished product reference image and boundary reference image are matched from a preset comparison image library.

In this embodiment, the finished product reference image and the boundary reference image may be specifically obtained by capturing an image of a successfully produced finished product at the workpiece stage.

Specifically, based on the known raw material model and processing type, the workpiece stage finished product image and the die-cutting boundary image of the successful sample die-cutting under the same working conditions are matched from the comparison image library as the finished product reference image and the boundary reference image.

S43: Compare the stage finished product pattern with the finished product reference image, and compare the stage finished product boundary with the boundary reference image. If there are no abnormal image features in both, generate qualified finished product specification information.

Specifically, the stage finished product pattern is compared with the finished product reference image, and the stage finished product boundary is compared with the boundary reference image to determine whether the quality of the finished product of this die-cutting work is qualified. If no abnormal image features appear during the comparison of the two images, qualified finished product specification information is generated, which makes it easy to know whether the current workpiece stage finished product processing quality is qualified.

It should be understood that the serial numbers of the steps in the above embodiments do not imply a sequence of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

Embodiment 2

An automatic optical detection system applied to a die-cutting machine corresponds to the automatic optical detection method applied to the die-cutting machine in the above embodiment.

As shown in Figure 2, the automatic optical inspection system applied to the die-cutting machine includes an image cropping algorithm update module, a first correction image generation module, a pressing and cutting start instruction generation module, and a finished product specification information generation module. The detailed description of each functional module is as follows:

An image cropping algorithm update module is used to determine the workpiece shape and position parameters, raw material model, and processing type of the current processing task, and to set the image cropping area of the current processing task based on the workpiece shape and position parameters to update the image cropping algorithm;

A first corrected image generation module is used to obtain a first optical detection image of the pressing and cutting surface before the workpiece pressing and cutting begins, and input it into a preset image preprocessing model to perform correction and cropping processing to generate a first corrected image;

A pressing and cutting start instruction generating module is used to intercept a stage finished product area image from the first corrected image based on the workpiece shape and position parameters, compare the stage finished product area image with a raw material reference image corresponding to the raw material model, generate feed specification information, and generate a pressing and cutting start instruction when the feed specification information is qualified;

The module for generating finished product specification information is used to obtain a second optical detection image after the workpiece is pressed and cut, generate a second corrected image through preprocessing, extract the stage finished product pattern and stage finished product boundary from the second corrected image based on the workpiece shape and position parameters, compare the stage finished product pattern and stage finished product boundary with the finished product reference image and boundary reference image corresponding to the processing type, and generate finished product specification information.

Among them, the image cropping algorithm update module also includes:

The processing task analysis submodule is used to receive processing setting information to set the processing program of the target die-cutting machine, and determine the workpiece shape and position parameters, raw material model, and processing type of the current processing task based on the processing setting information;

The image cropping area setting submodule is used to determine the shape information, size information and position information relative to the cutting surface of the workpiece stage finished product of the current processing task based on the workpiece shape and position parameters, set the image cropping area of the current processing task, and update the image cropping algorithm.

Wherein, the first corrected image generation module further includes:

A first optical detection image acquisition submodule is used to acquire a first optical detection image taken by the imaging component and input it into an image preprocessing model when the raw material is delivered to the site and before the workpiece pressing and cutting begins;

A first zoomed image generation submodule is used to obtain the viewing angle information of the imaging component and the shooting orientation information of the lens axis relative to the pressing and cutting surface and load them into the perspective correction algorithm, and to perform zooming processing on the first optical detection image in different directions according to different zooming ratios through the perspective correction algorithm to generate a first zoomed image;

The scaled image cropping processing submodule is used to identify an image cropping area from the first scaled image, perform cropping processing on the first scaled image based on the image cropping area, and generate a first corrected image.

The pressure cutting start instruction generation module also includes:

The submodule for capturing the image of the stage finished product area is used to determine the stage finished product area according to the boundary position of each stage finished product of each workpiece based on the shape and position parameters of the workpiece, and to capture the image of the stage finished product area from the first corrected image;

A raw material reference image matching submodule, used to match the corresponding raw material reference image from a preset comparison image library based on the raw material model;

The feed specification information generation submodule is used to compare the stage finished product area image with the raw material reference image to determine whether there are abnormal image features in the stage finished product area image. If there are no abnormal image features, qualified feed specification information is generated, and a pressing and cutting start instruction is generated and sent to the die-cutting controller.

Among them, the finished product specification information generation module also includes:

The second corrected image processing submodule is used to determine the area where the staged product and the die-cutting boundary of the workpiece are located in the second corrected image based on the shape and position parameters of the workpiece, so as to intercept the image of the corresponding area from the second corrected image to obtain the staged product pattern and the staged product boundary;

A finished product reference image matching submodule is used to match the corresponding finished product reference image and boundary reference image from a preset comparison image library based on the known raw material model and processing type;

The finished product image comparison submodule is used to compare the stage finished product pattern with the finished product reference image, and the stage finished product boundary with the boundary reference image. If there are no abnormal image features in both, qualified finished product specification information is generated.

For the specific limitations on the automatic optical inspection system applied to the die-cutting machine, please refer to the limitations on the automatic optical inspection method applied to the die-cutting machine in the above text, which will not be repeated here; the various modules in the above-mentioned automatic optical inspection system applied to the die-cutting machine can be fully or partially implemented by software, hardware and their combination; the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of the above modules.

Embodiment 3

A computer device, which may be a server, and its internal structure diagram may be shown in FIG3. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as workpiece shape and position parameters, raw material model, processing type, image clipping algorithm, first optical detection image, image preprocessing model, first correction image, stage finished product area image, raw material reference image, feeding specification information, pressing and cutting start instruction, second optical detection image, second correction image, stage finished product pattern, stage finished product boundary, finished product reference image, boundary reference image and finished product specification information. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an automatic optical detection method applied to a die-cutting machine is implemented.

In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

S10: determining the workpiece shape and position parameters, raw material model, and processing type of the current processing task, and setting the image cropping area of the current processing task based on the workpiece shape and position parameters to update the image cropping algorithm;

S20: before the workpiece cutting starts, a first optical detection image of the cutting surface is obtained and input into a preset image preprocessing model, and correction and cropping are performed to generate a first corrected image;

S30: intercepting a stage finished product area image from the first corrected image based on the workpiece shape and position parameters, comparing the stage finished product area image with a raw material reference image corresponding to the raw material model, generating feed specification information, and generating a press-cut start instruction when the feed specification information is qualified;

S40: After the workpiece is pressed and cut, a second optical detection image is obtained, which is preprocessed to generate a second corrected image. The stage finished product pattern and the stage finished product boundary are intercepted from the second corrected image based on the workpiece shape and position parameters. The stage finished product pattern and the stage finished product boundary are compared with the finished product reference image and the boundary reference image corresponding to the processing type to generate finished product specification information.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

S10: determining the workpiece shape and position parameters, raw material model, and processing type of the current processing task, and setting the image cropping area of the current processing task based on the workpiece shape and position parameters to update the image cropping algorithm;

S20: before the workpiece cutting starts, a first optical detection image of the cutting surface is obtained and input into a preset image preprocessing model, and correction and cropping are performed to generate a first corrected image;

S30: intercepting a stage finished product area image from the first corrected image based on the workpiece shape and position parameters, comparing the stage finished product area image with a raw material reference image corresponding to the raw material model, generating feed specification information, and generating a press-cut start instruction when the feed specification information is qualified;

S40: After the workpiece is pressed and cut, a second optical detection image is obtained, which is preprocessed to generate a second corrected image. The stage finished product pattern and the stage finished product boundary are intercepted from the second corrected image based on the workpiece shape and position parameters. The stage finished product pattern and the stage finished product boundary are compared with the finished product reference image and the boundary reference image corresponding to the processing type to generate finished product specification information.

Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink), DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments may still be modified, or some of the features thereof may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims (10)

1. The automatic optical detection method applied to the die cutting machine is characterized by comprising the following steps of:

Determining workpiece shape and position parameters, raw material model and processing type of a current processing task, and setting an image cutting area of the current processing task based on the workpiece shape and position parameters so as to update an image cutting algorithm;

Before the workpiece is pressed and cut, a first optical detection image of a pressing and cutting plate surface is obtained and input into a preset image preprocessing model, and correction and cutting processing are carried out to generate a first correction image;

Intercepting a stage finished product area image from the first corrected image based on the workpiece shape and position parameters, comparing the stage finished product area image with a raw material reference image corresponding to the raw material model, generating feeding specification information, and generating a press cutting starting instruction when the feeding specification information is qualified;

After the workpiece is pressed and cut, a second optical detection image is obtained, a second correction image is generated through preprocessing, a stage finished product pattern and a stage finished product boundary are intercepted from the second correction image based on the workpiece shape and position parameters, and the stage finished product pattern and the stage finished product boundary are compared with a finished product reference image and a boundary reference image corresponding to the processing type to generate finished product specification information;

The die cutting machine is provided with an imaging component for shooting the press-cut surface to obtain an optical detection image; the image preprocessing model is internally provided with a perspective correction algorithm and an image clipping algorithm.

2. The automatic optical inspection method applied to a die cutting machine according to claim 1, wherein: the determining of the workpiece shape and position parameter, the raw material model and the processing type of the current processing task, setting the image clipping area of the current processing task based on the workpiece shape and position parameter so as to update the image clipping algorithm, and the method comprises the following steps:

receiving processing setting information to set a processing program of a target die-cutting machine, and determining workpiece shape and position parameters, raw material models and processing types of a current processing task based on the processing setting information;

And determining the shape information, the size information and the position information of a finished product at the workpiece stage of the current processing task relative to the press layout on the basis of the workpiece shape and position parameters, and setting an image cutting area of the current processing task so as to update an image cutting algorithm.

3. The automatic optical inspection method applied to a die cutting machine according to claim 1, wherein: before the workpiece is pressed and cut, a first optical detection image of a pressing and cutting plate surface is acquired and input into a preset image preprocessing model, and correction and cutting processing are carried out to generate a first correction image, wherein the method comprises the following steps of:

When the raw materials are conveyed in place, a first optical detection image shot by an imaging assembly is acquired and input into an image preprocessing model before the workpiece is pressed and cut;

Acquiring visual angle information of an imaging assembly and shooting azimuth information of a lens axis relative to a press-cut panel, loading the visual angle information and the shooting azimuth information into a perspective correction algorithm, and performing scaling processing on a first optical detection image in different directions according to different scaling ratios through the perspective correction algorithm to generate a first scaled image;

An image cropping area is identified from the first scaled image, cropping processing is performed on the first scaled image based on the image cropping area, and a first corrected image is generated.

4. The automatic optical inspection method applied to a die cutting machine according to claim 1, wherein: intercepting a stage finished product area image from the first corrected image based on workpiece shape and position parameters, comparing the stage finished product area image with a raw material reference image corresponding to a raw material model, generating feeding specification information, and generating a pressing and cutting starting instruction when the feeding specification information is qualified, wherein the method comprises the following steps:

based on the workpiece shape and position parameters, determining a stage finished product area according to the boundary position of each workpiece stage finished product, and intercepting a stage finished product area image from the first corrected image;

matching corresponding raw material reference images from a preset comparison image library based on the raw material model;

Comparing the stage finished product area image with the raw material reference image, judging whether abnormal image features exist in the stage finished product area image, if the abnormal image features do not exist, generating qualified feeding standard information, generating a press cutting starting instruction at the same time, and sending the press cutting starting instruction to a die cutting controller;

and the stage finished product area image completely covers the stage finished product pattern and the area where the stage finished product boundary is located.

5. The automatic optical inspection method applied to a die cutting machine according to claim 1, wherein: intercepting a stage finished product pattern and a stage finished product boundary from a second corrected image based on workpiece shape and position parameters, comparing the stage finished product pattern and the stage finished product boundary with a finished product reference image and a boundary reference image corresponding to a processing type, and generating finished product specification information, wherein the method comprises the following steps of:

Determining the areas where the workpiece stage finished product and the die cutting boundary are located in the second corrected image based on the workpiece shape and position parameters, so as to intercept the image of the corresponding area from the second corrected image, and obtaining the stage finished product pattern and the stage finished product boundary;

based on the known raw material model and processing type, matching corresponding finished product reference images and boundary reference images from a preset comparison image library;

And comparing the stage finished product pattern with a finished product reference image, and comparing the stage finished product boundary with a boundary reference image, and if the stage finished product pattern and the boundary reference image have no abnormal image characteristics, generating qualified finished product specification information.

6. An automatic optical detection system applied to a die cutting machine, characterized in that it is applied to the automatic optical detection method applied to a die cutting machine according to any one of claims 1 to 5, comprising:

The image clipping algorithm updating module is used for determining workpiece shape and position parameters, raw material model and processing type of the current processing task, and setting an image clipping area of the current processing task based on the workpiece shape and position parameters so as to update the image clipping algorithm;

The first correction image generation module is used for acquiring a first optical detection image of the pressing and cutting plate before the workpiece is pressed and cut, inputting the first optical detection image into a preset image preprocessing model, and carrying out correction and cutting processing to generate a first correction image;

The pressing and cutting start instruction generation module is used for intercepting a stage finished product area image from the first corrected image based on the workpiece shape and position parameters, comparing the stage finished product area image with a raw material reference image corresponding to the raw material model, generating feeding standard information, and generating a pressing and cutting start instruction when the feeding standard information is qualified;

The finished product standard information generation module is used for acquiring a second optical detection image after the workpiece is pressed and cut, generating a second corrected image through preprocessing, intercepting a stage finished product pattern and a stage finished product boundary from the second corrected image based on the workpiece shape and position parameters, and comparing the stage finished product pattern and the stage finished product boundary with a finished product reference image and a boundary reference image corresponding to the processing type to generate finished product standard information;

The die cutting machine is provided with an imaging component for shooting the press-cut surface to obtain an optical detection image; the image preprocessing model is internally provided with a perspective correction algorithm and an image clipping algorithm.

7. The automated optical inspection system for a die cutting machine of claim 6 wherein: the image clipping algorithm updating module comprises:

The processing task analysis submodule is used for receiving processing setting information to set a processing program of the target die-cutting machine, and determining workpiece shape and position parameters, raw material models and processing types of the current processing task based on the processing setting information;

The image cutting area setting sub-module is used for determining the shape information, the size information and the position information relative to the press cutting plate surface of a finished product at the workpiece stage of the current processing task based on the workpiece shape and position parameters, and setting the image cutting area of the current processing task so as to update an image cutting algorithm.

8. The automated optical inspection system for a die cutting machine of claim 6 wherein: the first corrected image generation module includes:

the first optical detection image acquisition sub-module is used for acquiring a first optical detection image shot by the imaging component and inputting the first optical detection image into the image preprocessing model before the workpiece is pressed and cut when the raw material is conveyed in place;

The first zoom image generation sub-module is used for acquiring visual angle information of the imaging component and shooting azimuth information of the lens axis relative to the press-cut surface, loading the visual angle information and the shooting azimuth information into a perspective correction algorithm, and performing zoom processing on the first optical detection image in different directions according to different zoom scales through the perspective correction algorithm to generate a first zoom image;

And the scaled image clipping processing submodule is used for identifying an image clipping region from the first scaled image, clipping the first scaled image based on the image clipping region and generating a first corrected image.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, realizes the steps of the automatic optical inspection method applied to a die cutting machine according to any one of claims 1 to 5.

10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the automatic optical inspection method according to any one of claims 1 to 5 applied to a die cutting machine.