CN115713476A - Visual detection method and device based on laser welding and readable storage medium - Google Patents
Visual detection method and device based on laser welding and readable storage medium Download PDFInfo
- Publication number
- CN115713476A CN115713476A CN202110942117.1A CN202110942117A CN115713476A CN 115713476 A CN115713476 A CN 115713476A CN 202110942117 A CN202110942117 A CN 202110942117A CN 115713476 A CN115713476 A CN 115713476A
- Authority
- CN
- China
- Prior art keywords
- welding
- data
- image
- unqualified
- image sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Laser Beam Processing (AREA)
Abstract
The invention relates to the technical field of laser welding, and provides a visual detection method, a visual detection device and a readable storage medium based on laser welding, wherein the method comprises the following steps: controlling a galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples; comparing and analyzing each welding image sample with a standard welding image to obtain welding data of the welding image sample, and labeling unqualified welding sections in the welding data to obtain labeled welding data; acquiring welding process parameters of unqualified welding sections in the welding image sample, and associating the welding process parameters with welding data of the welding image sample to obtain training data; constructing and training a visual inspection model; and inputting the collected welding image to be detected into the trained visual detection model, and judging whether the welding image to be detected has an unqualified welding section by the visual detection model. The technical scheme provided by the embodiment of the invention can effectively improve the detection efficiency of visually detecting the welding effect.
Description
[ technical field ] A
The invention relates to the technical field of laser welding, in particular to a visual detection method and device based on laser welding and a readable storage medium.
[ background ] A method for producing a semiconductor device
Laser galvanometer welding is the leading welding technology in the current welding process. And detecting a welding pattern formed by welding the laser galvanometer to judge whether the welding process is qualified.
At present, a welding pattern is influenced by factors such as axial motion error, laser energy change, layer overlapping, uneven materials and the like, the welding pattern can have certain deviation with an expected pattern, particularly, the deviation is obvious at a joint of the welding pattern, the quality of the joint is judged mainly by observing under a microscope through human eyes in the industry at present, in an actual application scene, an inventor finds that the welding pattern is limited by the quality of detection personnel, and the detection efficiency of the welding effect is low.
[ summary of the invention ]
In view of this, embodiments of the present invention provide a visual inspection method and apparatus based on laser welding, and a readable storage medium, which can effectively improve the detection efficiency of visually detecting the welding effect.
In order to achieve the above object, in a first aspect, there is provided a visual inspection method based on laser welding, the method including:
controlling a galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples;
comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample, and marking unqualified welding sections in the welding data to obtain marked welding data;
acquiring welding process parameters of unqualified welding sections in the welding image sample, and associating the welding process parameters with welding data of the welding image sample to obtain training data;
constructing a visual detection model, and importing the training data into the visual detection model for training to obtain a trained visual detection model;
and inputting the acquired welding image to be detected into the trained visual detection model, judging whether the welding image to be detected has an unqualified welding section or not by the visual detection model, and outputting the welding process parameter range of the unqualified welding section, so that the parameter adjustment can be performed on the galvanometer welding joint according to the welding process parameter range of the unqualified welding section.
With reference to the first aspect, in a possible implementation manner, the comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample includes:
smoothing the welding image sample to obtain a denoised welding image sample;
and positioning a welding area in the welding image sample through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image sample, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
With reference to the first aspect, in a possible implementation manner, the labeling the failed welding segment in the welding data to obtain labeled welding data includes:
performing primary marking on the welding data, wherein a label of the primary marking comprises one of qualified label and unqualified label; and carrying out secondary labeling on the welding data, wherein the label of the secondary labeling comprises one of an overlapping defect, a dislocation defect and an offset defect, and the labeled welding data is obtained.
With reference to the first aspect, in one possible implementation manner, the obtaining welding process parameters of a failed welding section in the welding image sample and associating the welding process parameters with the welding data of the welding image sample includes the following steps:
acquiring a first welding process parameter of the galvanometer welding head when the unqualified welding section is welded, and associating the first welding process parameter with the unqualified welding section in the marked welding image sample;
and acquiring a second welding process parameter of the galvanometer welding head when the galvanometer welding head welds a qualified welding section corresponding to the position of the unqualified welding section, and associating the second welding process parameter with the qualified welding section in the marked welding image sample.
With reference to the first aspect, in a possible implementation manner, the welding process parameter includes at least one of a moving track of the galvanometer welding head, a speed of the galvanometer welding head, an acceleration of the galvanometer welding head, a laser power of the galvanometer welding head, and a laser switch delay of the galvanometer welding head.
With reference to the first aspect, in a possible implementation manner, the welding image to be detected and the welding image sample are acquired by an industrial camera, the welding image to be detected and the welding image sample are both 2D grayscale images or 3D grayscale images, and the imaging device is a CCD industrial camera, a CMOS industrial camera, a 2D line scan industrial camera or a 3D line scan industrial camera.
With reference to the first aspect, in one possible implementation, the visual inspection model includes at least one of a support vector machine, an artificial neural network, and a random forest decision.
In order to achieve the above object, in a third aspect, the present application provides a visual inspection apparatus, comprising:
the control unit is used for controlling the galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples;
the comparison unit is used for comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample, and labeling unqualified welding sections in the welding data to obtain labeled welding data;
the acquisition unit is used for acquiring welding process parameters of unqualified welding sections in the welding image sample, and associating the welding process parameters with welding data of the welding image sample to obtain training data;
the training unit is used for constructing a visual detection model and importing the training data into the visual detection model for training until network parameters of the visual detection model are converged to obtain a trained visual detection model;
and the output unit is used for inputting the acquired welding image to be detected into the trained visual detection model, judging whether the welding image to be detected has an unqualified welding section or not by the visual detection model, and outputting the welding process parameter range of the unqualified welding section so as to perform parameter adjustment on the galvanometer welding head according to the welding process parameter range of the unqualified welding section.
With reference to the second aspect, in a possible implementation manner, the processing unit is further configured to:
smoothing the welding image sample to obtain a denoised welding image sample;
and positioning a welding area in the welding image sample through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image sample, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
In order to achieve the above object, in a third aspect, the present application provides a readable storage medium, where the storage medium includes a stored program, and when the program runs, the apparatus on which the storage medium is located is controlled to execute the above-mentioned visual detection method based on laser welding.
In the scheme, the laser welding process parameters are associated with the welding data and used as training data to train the visual inspection model, so that the visual inspection model can identify qualified welding images or unqualified welding images, and then the trained visual inspection model is used for detecting the welding images to be detected. The visual detection model can learn the characteristics of various unqualified defects and the characteristics of unqualified welding process parameters according to the labeled labels, so that the visual detection model can identify whether the welding image to be detected is qualified or not, the intelligent detection of the visual detection model is realized, the influence of inconsistent judgment due to personnel difference can be reduced, the detection efficiency of the welding effect is improved, an optimization scheme aiming at welding unqualified is improved according to the welding process parameter range output of the unqualified welding section, and the welding qualification rate is further improved.
[ description of the drawings ]
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a method for visual inspection based on laser welding according to an embodiment of the present invention;
FIG. 2-1 is a schematic structural diagram of a laser welding visual inspection system provided by an embodiment of the invention;
2-2 are schematic views of the working area of the laser galvanometer welding head in the laser welding visual detection system provided by the embodiment of the invention;
FIG. 3 is a schematic view of an alternative visual inspection device provided by an embodiment of the present invention;
fig. 4 is a schematic diagram of an alternative computer device provided by the embodiment of the present invention.
[ detailed description ] A
In order to better understand the technical scheme of the invention, the following detailed description of the embodiments of the invention is made with reference to the accompanying drawings.
It should be understood that the described embodiments are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terminology used in the embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the description of the invention and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be understood that the term "and/or" as used herein is merely one type of association that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter associated objects are in an "or" relationship.
It should be understood that although the terms first, second, third, etc. may be used to describe the terminals in the embodiments of the present invention, the terminals should not be limited by these terms. These terms are only used to distinguish one terminal from another. For example, a first terminal may also be referred to as a second terminal, and similarly, a second terminal may also be referred to as a first terminal, without departing from the scope of embodiments of the present invention.
The word "if" as used herein may be interpreted as "at 8230; \8230;" or "when 8230; \8230;" or "in response to a determination" or "in response to a detection", depending on the context. Similarly, the phrase "if determined" or "if detected (a stated condition or event)" may be interpreted as "upon determining" or "in response to determining" or "upon detecting (a stated condition or event)" or "in response to detecting (a stated condition or event)", depending on the context.
Fig. 1 is a flowchart of a method for visual inspection based on laser welding according to an embodiment of the present invention, as shown in fig. 1, the method comprising:
and S10, controlling a galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples.
And step S20, comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample, and labeling unqualified welding sections in the welding data to obtain labeled welding data.
And S30, acquiring welding process parameters of unqualified welding sections in the welding image sample, and associating the welding process parameters with the welding data of the welding image sample to obtain training data.
And S40, constructing a visual detection model, and importing training data into the visual detection model for training to obtain the trained visual detection model.
And S50, inputting the collected welding image to be detected into a trained visual detection model, judging whether the welding image to be detected has an unqualified welding section by the visual detection model, and outputting a welding process parameter range of the unqualified welding section so as to perform parameter adjustment on the galvanometer welding head according to the welding process parameter range of the unqualified welding section.
In the scheme, the laser welding process parameters are associated with the welding data and used as training data to train the visual inspection model, so that the visual inspection model can recognize qualified welding images or unqualified welding images, and the trained visual inspection model is used for detecting the welding images to be detected. The visual detection model can learn the characteristics of various unqualified defects and the characteristics of unqualified welding process parameters according to the labeled labels, so that the visual detection model can identify whether the welding image to be detected is qualified or not, the intelligent detection of the visual detection model is realized, the influence of inconsistent judgment due to personnel difference can be reduced, the detection efficiency of the welding effect is improved, an optimization scheme aiming at welding unqualified is improved according to the welding process parameter range output of the unqualified welding section, and the welding qualification rate is further improved.
The above-described laser welding-based visual inspection method is described in detail below with reference to specific examples:
and S10, controlling a galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples.
In some embodiments, the standard weld image is a weld image of a workpiece design that the customer or designer has machined as desired. Specifically, a standard welding image can be drawn by drawing software such as CAD. The welding path in the labeled welding image may be a circle, a square, a circular array, a square array, or other predetermined path, which is not limited herein. Specifically, the line width of the welded segment in the image produced by the CAD is within a preset range.
As shown in fig. 2-1, the laser galvanometer welding system comprises a support 1, a laser galvanometer welding head 2 and a moving platform 3. The laser galvanometer welding head 2 is used for emitting laser beams to weld workpieces, and the laser galvanometer welding head 2 is connected with the support 1 and can deflect the laser beams to change the emitting direction of the laser beams. As long as the workpiece is located in the welding range of the laser galvanometer welding head, the moving platform 3 is used for driving the laser galvanometer welding head to move so as to weld the workpiece along a preset welding path.
The laser galvanometer welding head can enable light spots formed on the surface of a workpiece by laser beams to circularly swing along a preset welding path. That is, when the light spot moves, the light spot forms a circular ring first, and then moves along a preset welding path to form a plurality of circular rings.
The laser galvanometer welding system comprises an imaging device 4, the imaging device 4 is connected with the support 1, and the imaging device 4 is used for acquiring a welding image to be detected of a welded workpiece. The welding image to be detected is a 2D gray image or a 3D gray image, and the imaging device is a CCD industrial camera, a CMOS industrial camera, a 2D line scanning industrial camera or a 3D line scanning industrial camera. In the specific embodiment, taking a CCD industrial camera as an example for the following description, the CCD industrial camera may be dotted by laser to calibrate the object distance between the CCD industrial camera and the laser galvanometer welding head. Optionally, the mechanical coordinate system of the CCD industrial camera is determined by a 9-point calibration emulation matrix, and camera distortion calibration is performed by a calibration plate. The welding image taken by the CCD industrial camera is a grayscale image with 500 ten thousand pixels.
For workpieces of the same specification, the welding path is constant, i.e., the welding path fed into the laser galvanometer welding head in advance is not changed. However, different workpieces may have a slight deviation in the position of the portion to be welded due to processing errors, and in order to make welding more accurate, it is necessary to determine a welding start point for each workpiece again, that is, a start position of a light spot formed on the surface of the workpiece by the laser beam after the laser beam is applied to the laser galvanometer welding head. Of course, if the deviation is small, the untimely time can be ignored, and the imaging device can be omitted.
As shown in fig. 2-2, the rectangular area is a single welding area of the laser galvanometer welding head, and since the welding area is limited, after exceeding the range of the welding area, the laser galvanometer welding head can be driven to move by the moving platform, and then welding is performed again, and after a plurality of welding areas are spliced, a complete welding path is completed, so that the defects of overlapping, deviation, dislocation and the like at the interface are easily caused by mechanical displacement errors at the joint of the welding area.
In the present embodiment, the standard welding image includes a welding path. After the welding path is determined, a welding simulation test can be carried out. And obtaining a plurality of welding image samples through a plurality of times of welding. Specifically, a welding image to be detected and a welding image sample can be acquired through an imaging device, the welding image to be detected and the welding image sample are both 2D gray images or 3D gray images, and the imaging device is a CCD industrial camera, a CMOS industrial camera, a 2D line scanning industrial camera or a 3D line scanning industrial camera. Alternatively, each preset standard welding image may correspond to welding at least 50 welding image samples. The specifications of each weld image sample should be uniform so that there is good consistency of the weld image samples.
Furthermore, in order to increase the sample size, the welding image sample can be expanded, so that the sample is diversified. Specifically, the strategy of augmentation may be at least one of a rotation transformation, a flipping transformation, a scaling transformation, a region clipping, a noise addition, a contrast transformation, a color dithering, a composite overlay.
The rotation transformation means that the welding image is randomly rotated by a preset angle, and the orientation of the welding image is changed. The flip transformation is to flip the welding image in either the horizontal or vertical direction. The scaling transformation refers to enlarging or reducing an image in a preset scale. The scale transformation refers to enlarging or reducing an image according to a preset scale factor, or constructing a scale space by filtering the image by using the preset scale factor, so as to change the size or the fuzzy degree of the image content. Region cropping refers to cropping a region of interest in a welding image. Adding noise means that some noise is randomly superimposed on the original picture. The contrast transformation is to change the saturation S and V brightness components in the HSV color space of the image, keep the hue H unchanged, perform exponential operation on the S and V components of each pixel (the exponential factor is between 0.25 and 4), and increase the illumination change. The color dithering means that exposure, saturation and hue of an image are randomly changed to form pictures under different illumination and colors, and the model can be used in the situation that different illumination conditions are small as much as possible. The composite superposition refers to randomly extracting two pictures, respectively performing basic data augmentation operation processing, and superposing the two pictures into a new sample in a pixel averaging mode, wherein the label of the new sample is one of original sample labels.
Specifically, step S20 includes:
and S21, comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample.
And S22, labeling the unqualified welding sections in the welding data to obtain the labeled welding data.
Wherein, step S21 includes:
smoothing the welding image sample to obtain a denoised welding image sample;
and positioning a welding area in the welding image sample through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image sample, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
The smoothing process includes at least one of a gaussian filter process, a mean filter process, and a bilateral filter process. Preferably, the weld image samples are smoothed using a gaussian filtering process. Blob analysis is the analysis of connected components of the same pixel (pad, edge, etc.) in the pad image.
In other embodiments, the welding image may be analyzed by other analysis means as long as the welding data of the welding path in the welding image can be acquired. When the imaging device is a 2D industrial camera, the weld data may be the width of the weld segment. When the imaging device is a three-dimensional industrial camera, the weld data may be the width and height of the weld segment.
After the second order differential processing, the data acquisition is performed with the start point of the welding path as the acquisition start point of the welding data. In the present embodiment, the welding data is a set of shortest distances from a differential point at one end of the current welding segment to another adjacent welding segment. To improve the computational efficiency of the welding data, in some embodiments, the algorithm may be accelerated by high concurrent Processing based on an Open Computing Language (OpenCL) software library, which is matched with the increased computational power of a Graphics Processing Unit (GPU).
Further, step S22, labeling the unqualified welding section in the welding data to obtain the labeled welding data, specifically includes the following steps:
performing primary labeling on the welding data, wherein the label of the primary labeling comprises one of qualified label and unqualified label; and performing secondary labeling on the welding data, wherein the label of the secondary labeling comprises one of an overlapping defect, a dislocation defect and an offset defect, and obtaining the labeled welding data.
The welding data of each welding image sample is provided with a label, for example, the welding data with a positive label, namely the welding data marked as qualified (welding image sample), the welding data with a negative label, and the welding data marked as unqualified (welding image sample). The unqualified label also comprises a sub-label, and the sub-label specifically comprises one of an overlapping defect, a dislocation defect and an offset defect. When setting the label, the label of the welding data can be represented by (A1), wherein a is a label of the primary label, namely, represents pass or fail; 1 is a label of secondary labeling, which represents an overlapping defect; then the weld data for this weld image sample is a failure due to an overlay defect.
And S30, acquiring welding process parameters of unqualified welding sections in the welding image sample, and associating the welding process parameters with the welding data of the welding image sample to obtain training data.
Specifically, a first welding process parameter of a galvanometer welding head during welding of an unqualified welding section is obtained, and the first welding process parameter is associated with the unqualified welding section in the marked welding image sample;
and acquiring a second welding process parameter of the galvanometer welding head when the qualified welding section corresponding to the unqualified welding section is welded, and associating the second welding process parameter with the qualified welding section in the marked welding image sample.
In some embodiments, the welding process parameter comprises at least one of a galvanometer weld head movement trajectory, a velocity of the galvanometer weld head, an acceleration of the galvanometer weld head, a power of the galvanometer weld head. The welding process parameters are correlated with the marked welding data, so that the welding process parameters of unqualified welding sections in the welding image can be clearly recorded. In the welding process of workpieces with the same specification, the welding process parameters of the qualified welding section at the same position can be used as an optimization scheme for optimizing the unqualified welding section. For example, the speed of the galvanometer welding head in the welding section (L1) is between V1 and V2, and when the speed of the galvanometer welding head is greater than V2 or less than V1, dislocation defects or deviation defects can occur in the welding section.
The welding process parameters are associated with the welding data, so that the visual inspection model can extract the welding process parameters in the subsequent training process, and the welding process parameter range of the defects causing the type of unqualified defects can be given. For example, after acquiring welding process parameters associated with 50 welding data from 50 welding data of welding data with an overlap defect failure tag, analysis shows that the speed of the galvanometer welding head is between (V3, V4), and the speed of the galvanometer welding head is between (V1, V2) in the second welding process parameter of a qualified welding section corresponding to the position of the failed welding section, so that the speed of the galvanometer welding head deviates from a preset range and overlap defects are easy to generate.
And S40, constructing a visual detection model, importing training data into the visual detection model for training, and obtaining the trained visual detection model.
Optionally, the visual inspection model comprises at least one of a support vector machine, an artificial neural network, a random forest decision. Preferably, the visual inspection model is a support vector machine model.
In the specific training process, the training data is learned and classified by using a support vector machine algorithm, and model parameters of the trained visual detection model are obtained.
And S50, inputting the acquired welding image to be detected into the trained visual detection model, judging whether the welding image to be detected has an unqualified welding section or not by the visual detection model, and outputting the welding process parameter range of the unqualified welding section, so that the parameter of the galvanometer welding head can be adjusted according to the welding process parameter range of the unqualified welding section.
In some embodiments, the weld image to be detected may be acquired by an imaging device. Similarly, before the visual inspection model is input, the welding image to be detected also needs to be smoothed to obtain a de-noised welding image to be detected; and then, positioning a welding area in the welding image to be detected through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image to be detected, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
The visual inspection model can extract features of the welding data and determine whether the welding data is qualified or unqualified based on the features. In some embodiments, the data output by the visual inspection model includes the probability of being acceptable and the associated defect.
Step S50, further including:
acquiring all welding process parameters related to the welding data of the same type of label stored in the model according to the label type of the unqualified welding section;
and performing feature extraction according to all welding process parameters related to the welding data of the same type of label to obtain the welding process parameter range of the unqualified welding section.
Exemplarily, when the welding image to be detected is unqualified caused by the deviation defect, the visual detection model stores all welding process parameters of the deviation defect label according to the deviation defect label after training; therefore, a user can adaptively adjust specific welding process parameters according to the welding process parameter range output by the model so as to eliminate the deviation defect.
An embodiment of the present invention provides a visual inspection apparatus, which is configured to perform the above-mentioned visual inspection method based on laser welding, and as shown in fig. 3, the apparatus includes: control unit 10, processing unit 20, acquisition unit 30, training unit 40, output unit 50.
The control unit 10 is used for controlling the galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples;
the processing unit 20 is configured to compare and analyze each welding image sample with a standard welding image to obtain welding data of the welding image sample, and label an unqualified welding section in the welding data to obtain labeled welding data;
the acquiring unit 30 is configured to acquire welding process parameters of an unqualified welding section in a welding image sample, and associate the welding process parameters with welding data of the welding image sample to obtain training data;
the training unit 40 is used for constructing a visual detection model, and importing training data into the visual detection model for training until network parameters of the visual detection model are converged to obtain a trained visual detection model;
and the output unit 50 is used for inputting the acquired welding image to be detected into the trained visual detection model, judging whether the welding image to be detected has an unqualified welding section or not by the visual detection model, and outputting the welding process parameter range of the unqualified welding section, so that the parameter adjustment can be performed on the galvanometer welding head according to the welding process parameter range of the unqualified welding section.
In the scheme, the laser welding process parameters are associated with the welding data and used as training data to train the visual inspection model, so that the visual inspection model can identify qualified welding images or unqualified welding images, and then the trained visual inspection model is used for detecting the welding images to be detected. The visual detection model can learn the characteristics of various unqualified defects and the characteristics of unqualified welding process parameters according to the labeled labels by performing deep learning training according to the labeled welding image samples, so that the visual detection model can identify whether the welding image to be detected is qualified or not, intelligent detection of the visual detection model is realized, the influence of inconsistent judgment due to personnel difference can be reduced, the detection efficiency of the welding effect is improved, and an optimization scheme for welding unqualified can be output and improved according to the welding process parameter range of the unqualified welding section, so that the welding qualification rate is further improved.
In some embodiments, the standard weld image is a weld image of a workpiece design that the customer or designer has machined as desired. Specifically, a standard welding image can be drawn by drawing software such as CAD. The welding path in the labeled welding image may be a circle, a square, a circular array, a square array, or other predetermined path, which is not limited herein. Specifically, the line width of the welded segment in the image produced by the CAD is within a preset range.
In the present embodiment, the standard welding image includes a welding path. After the welding path is determined, a welding simulation test can be performed. And obtaining a plurality of welding image samples through a plurality of times of welding. Specifically, a welding image to be detected and a welding image sample can be acquired through an imaging device, the welding image to be detected and the welding image sample are both 2D gray images or 3D gray images, and the imaging device is a CCD industrial camera, a CMOS industrial camera, a 2D line scanning industrial camera or a 3D line scanning industrial camera. Optionally, each preset standard welding image corresponds to welding at least 50 welding image samples. The specifications of each weld image sample should be uniform.
Further, in order to increase the sample size, the welding image sample can be expanded, so that the sample is diversified. Specifically, the strategy of augmentation may be at least one of rotation transformation, flip transformation, scaling transformation, scale transformation, region clipping, noise addition, contrast transformation, color dithering, composite superposition.
The rotation transformation means that the welding image is randomly rotated by a preset angle, and the orientation of the welding image is changed. The flip transformation is to flip the welding image in either the horizontal or vertical direction. The scaling transformation refers to enlarging or reducing an image in a preset scale. The scale transformation refers to enlarging or reducing an image according to a preset scale factor, or constructing a scale space by filtering the image by using the preset scale factor, so as to change the size or the fuzzy degree of the image content. Region cropping refers to cropping a region of interest in a welding image. Adding noise means randomly superimposing some noise on the original picture. The contrast transformation is to change the saturation S and V brightness components in the HSV color space of the image, keep the hue H unchanged, perform exponential operation on the S and V components of each pixel (the exponential factor is between 0.25 and 4), and increase the illumination change. The color dithering means that exposure, saturation and hue of an image are randomly changed to form pictures under different illumination and colors, and the model can be used in the situation that different illumination conditions are small as much as possible. The composite superposition refers to randomly extracting two pictures, respectively performing basic data augmentation operation processing, and superposing the two pictures into a new sample in a pixel averaging mode, wherein the label of the new sample is one of original sample labels.
Specifically, the processing unit 20 includes an analyzing subunit and a labeling subunit;
the analysis subunit is used for comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample;
and the marking subunit is used for marking the unqualified welding sections in the welding data to obtain the marked welding data.
The analysis subunit is further configured to smooth the welding image sample to obtain a denoised welding image sample; and positioning the welding area in the welding image sample through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image sample. In some embodiments, the weld data includes at least one of a height of the weld segment, a width of the weld segment.
The smoothing process includes at least one of a gaussian filter process, a mean filter process, and a bilateral filter process. Preferably, the weld image samples are smoothed using a gaussian filtering process. Blob analysis is the analysis of connected components of the same pixel (pad, edge, etc.) in the pad image.
In other embodiments, the welding image may be analyzed by other analysis means as long as the welding data of the welding path in the welding image can be acquired. When the imaging device is a 2D industrial camera, the weld data may be the width of the weld segment. When the imaging device is a three-dimensional industrial camera, the weld data may be the width and height of the weld segment.
After the second order differential processing, the data acquisition is performed with the start point of the welding path as the acquisition start point of the welding data. In the present embodiment, the welding data is a set of shortest distances from a differential point at one end of the current welding segment to another adjacent welding segment. To improve the computational efficiency of the welding data, in some embodiments, the algorithm may be accelerated by high-concurrency processing based on an OpenCL software library, and the computation power may be increased by matching with a high-performance GPU.
Further, the labeling unit 22 is specifically configured to perform primary labeling on the welding data, where a label of the primary labeling includes one of a qualified label and an unqualified label; and performing secondary labeling on the welding data, wherein the label of the secondary labeling comprises one of an overlapping defect, a dislocation defect and an offset defect, and obtaining the labeled welding data.
The welding data of each welding image sample is provided with a label, for example, the welding data with a positive label is the welding data marked as qualified (welding image sample), the welding data with a negative label is the welding data marked as unqualified (welding image sample). The unqualified label also comprises a sub-label, and the sub-label specifically comprises one of an overlapping defect, a dislocation defect and an offset defect. When setting the label, the label of the welding data can be represented by (A1), wherein a is a label of the primary label, namely, represents pass or fail; 1 is a label of secondary labeling, which represents an overlapping defect; the weld data for this weld image sample is a failure due to an overlay defect.
Further, an obtaining unit 30 is configured to:
acquiring a first welding process parameter of a galvanometer welding head when the galvanometer welding head is welding an unqualified welding section, and associating the first welding process parameter with the unqualified welding section in the marked welding image sample;
and acquiring a second welding process parameter of the galvanometer welding head when the galvanometer welding head welds a qualified welding section corresponding to the position of the unqualified welding section, and associating the second welding process parameter with the qualified welding section in the marked welding image sample.
In some embodiments, the welding process parameter comprises at least one of a galvanometer weld head movement trajectory, a speed of the galvanometer weld head, an acceleration of the galvanometer weld head, a power of the galvanometer weld head. The welding process parameters are associated with the marked welding data, so that the welding process parameters of unqualified welding sections in the welding image can be clearly recorded. In the welding process of workpieces with the same specification, the welding process parameters of the qualified welding section at the same position can be used as an optimization scheme for optimizing the unqualified welding section. For example, the speed of the galvanometer welding head in the welding section (L1) is between V1 and V2, and when the speed of the galvanometer welding head is greater than V2 or less than V1, dislocation defects or deviation defects can occur in the welding section.
As can be appreciated, by associating the welding process parameters with the welding data, it is beneficial for the visual inspection model to extract the welding process parameters during the subsequent training process, so that the welding process parameter range that causes such unqualified defects can be given. For example, after acquiring welding process parameters associated with 50 welding data from 50 welding data of welding data with an overlap defect failure tag, analysis shows that the speed of the galvanometer welding head is between (V3, V4), and the speed of the galvanometer welding head is between (V1, V2) in the second welding process parameter of a qualified welding section corresponding to the position of the failed welding section, so that the speed of the galvanometer welding head deviates from a preset range and overlap defects are easy to generate.
Optionally, the visual inspection model comprises at least one of a support vector machine, an artificial neural network, a random forest decision. Preferably, the visual detection model is a support vector machine model.
In the specific training process, the training data is learned and classified by using a support vector machine algorithm, and model parameters of the trained visual detection model are obtained.
In some embodiments, the weld image to be detected may be acquired by an imaging device. Similarly, before inputting the visual inspection model, smoothing the welding image to be inspected to obtain a de-noised welding image to be inspected; and positioning a welding area in the welding image to be detected through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image to be detected, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
The visual inspection model can extract features of the welding data and determine whether the welding data is qualified or unqualified based on the features. In some embodiments, the data output by the visual inspection model includes the probability of being acceptable and the associated defect.
An output unit 50, further configured to:
acquiring all welding process parameters related to the welding data of the same type of label stored in the model according to the label type of the unqualified welding section;
and performing feature extraction according to all welding process parameters related to the welding data of the same type of label to obtain the welding process parameter range of the unqualified welding section.
Exemplarily, when the welding image to be detected is unqualified caused by the deviation defect, the visual detection model stores all welding process parameters of the deviation defect label according to the deviation defect label after training; therefore, a user can adaptively adjust specific welding process parameters according to the welding process parameter range output by the model so as to eliminate the deviation defect.
The embodiment of the invention provides a readable storage medium, which comprises a stored program, wherein when the program runs, a device where the storage medium is located is controlled to execute the following steps:
controlling a galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples; comparing and analyzing each welding image sample with a standard welding image to obtain welding data of the welding image sample, and labeling unqualified welding sections in the welding data to obtain labeled welding data; acquiring welding process parameters of unqualified welding sections in the welding image sample, and associating the welding process parameters with welding data of the welding image sample to obtain training data; constructing a visual detection model, and importing training data into the visual detection model for training to obtain a trained visual detection model; and inputting the collected welding image to be detected into the trained visual detection model, judging whether the welding image to be detected has an unqualified welding section by the visual detection model, and outputting the welding process parameter range of the unqualified welding section, so that the parameter adjustment can be performed on the galvanometer welding head according to the welding process parameter range of the unqualified welding section.
Optionally, the program controls the apparatus in which the storage medium is located to perform the following steps when running: smoothing the welding image sample to obtain a denoised welding image sample; and positioning a welding area in the welding image sample through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image sample, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
Optionally, the program controls the apparatus in which the storage medium is located to perform the following steps when running: performing primary labeling on the welding data, wherein the label of the primary labeling comprises one of qualified label and unqualified label; and carrying out secondary labeling on the welding data, wherein a label of the secondary labeling comprises one of an overlapping defect, a dislocation defect and an offset defect, and obtaining the labeled welding data.
Optionally, the program controls the apparatus in which the storage medium is located to perform the following steps when running: acquiring a first welding process parameter of a galvanometer welding head when the galvanometer welding head is welding an unqualified welding section, and associating the first welding process parameter with the unqualified welding section in the marked welding image sample;
and acquiring a second welding process parameter of the galvanometer welding head when the qualified welding section corresponding to the unqualified welding section is welded, and associating the second welding process parameter with the qualified welding section in the marked welding image sample.
Fig. 4 is a schematic diagram of a computer device according to an embodiment of the present invention. As shown in fig. 4, the computer apparatus 100 of this embodiment includes: the processor 101, the memory 102, and the computer program 103 stored in the memory 102 and capable of running on the processor 101, wherein the processor 101 implements the live pig weight measurement method in the embodiment when executing the computer program 103, and therefore, for avoiding repetition, details are not repeated herein. Alternatively, the computer program is executed by the processor 101 to implement the functions of each model/unit in the live pig weight measuring device in the embodiment, which is not repeated herein to avoid repetition.
The computing device 100 may be a desktop computer, a notebook, a palmtop, a cloud server, or other computing device. The computer device may include, but is not limited to, a processor 101, a memory 102. Those skilled in the art will appreciate that fig. 3 is merely an example of a computing device 100 and is not intended to limit the computing device 100 and that it may include more or less components than those shown, or some of the components may be combined, or different components, e.g., the computing device may also include input output devices, network access devices, buses, etc.
The Processor 101 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 102 may be an internal storage unit of the computer device 100, such as a hard disk or a memory of the computer device 100. The memory 102 may also be an external storage device of the computer device 100, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc., provided on the computer device 100. Further, the memory 102 may also include both internal storage units and external storage devices of the computer device 100. The memory 102 is used for storing computer programs and other programs and data required by the computer device. The memory 102 may also be used to temporarily store data that has been output or is to be output.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one type of logical functional division, and other divisions may be realized in practice, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or in the form of hardware plus a software functional unit.
The integrated unit implemented in the form of a software functional unit may be stored in a computer-readable storage medium. The software functional unit is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) or a Processor (Processor) to execute some steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may 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. By way of illustration and not limitation, RAM is available in a variety of 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), rambus (Rambus) direct RAM (RDRAM), direct bused dynamic RAM (DRDRAM), and bused dynamic RAM (RDRAM), among others.
It will be clear to those skilled in the art that, for convenience and simplicity of description, the foregoing functional units and modules are merely illustrated as being divided, and in practical applications, the foregoing functional allocation may be performed by different functional modules, sub-modules and units according to needs, that is, the internal structure of the device is divided into different functional units or modules to perform all or part of the above-described functions.
The above-mentioned embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A method of visual inspection based on laser welding, the method comprising:
controlling a galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples;
comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample, and marking unqualified welding sections in the welding data to obtain marked welding data;
acquiring welding process parameters of unqualified welding sections in the welding image sample, and associating the welding process parameters with welding data of the welding image sample to obtain training data;
constructing a visual detection model, and importing the training data into the visual detection model for training to obtain a trained visual detection model;
and inputting the collected welding image to be detected into the trained visual detection model, judging whether the welding image to be detected has an unqualified welding section by the visual detection model, and outputting the welding process parameter range of the unqualified welding section so as to perform parameter adjustment on the galvanometer welding head according to the welding process parameter range of the unqualified welding section.
2. The method of claim 1, wherein comparing each of the welding image samples with the standard welding image to obtain welding data of the welding image samples comprises:
smoothing the welding image sample to obtain a denoised welding image sample;
and positioning a welding area in the welding image sample through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image sample, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
3. The method of claim 1 or 2, wherein the labeling of the failed weld segment in the weld data to obtain labeled weld data comprises:
performing primary marking on the welding data, wherein the label of the primary marking comprises one of qualified label and unqualified label;
and carrying out secondary labeling on the welding data, wherein the label of the secondary labeling comprises one of an overlapping defect, a dislocation defect and an offset defect, and obtaining the labeled welding data.
4. The method of claim 1, wherein the obtaining welding process parameters for a failed weld segment in the weld image sample and correlating the welding process parameters with the weld data for the weld image sample comprises:
acquiring a first welding process parameter of the galvanometer welding head when the unqualified welding section is welded, and associating the first welding process parameter with the unqualified welding section in the marked welding image sample;
and acquiring a second welding process parameter of the galvanometer welding head when the qualified welding section corresponding to the unqualified welding section position is welded, and associating the second welding process parameter with the qualified welding section in the marked welding image sample.
5. The method of claim 1 or 4, wherein the welding process parameter comprises at least one of a galvanometer weld head movement trajectory, a speed of the galvanometer weld head, an acceleration of the galvanometer weld head, a power of the galvanometer weld head.
6. The method according to claim 1, characterized in that the welding image to be detected and the welding image sample are acquired by an imaging device, both the welding image to be detected and the welding image sample are 2D gray scale images or 3D gray scale images, and the imaging device is a CCD industrial camera, a CMOS industrial camera, a 2D line scan industrial camera or a 3D line scan industrial camera.
7. The method of any one of claims 1 to 6, wherein the visual inspection model comprises at least one of a support vector machine, an artificial neural network, and a random forest decision.
8. A visual inspection apparatus, the apparatus comprising:
the control unit is used for controlling the galvanometer welding head to weld according to a preset standard welding image to obtain a plurality of welding image samples;
the processing unit is used for comparing and analyzing each welding image sample with the standard welding image to obtain welding data of the welding image sample, and labeling unqualified welding sections in the welding data to obtain labeled welding data;
the acquisition unit is used for acquiring welding process parameters of unqualified welding sections in the welding image sample and associating the welding process parameters with welding data of the welding image sample to obtain training data;
the training unit is used for constructing a visual detection model, importing the training data into the visual detection model for training until network parameters of the visual detection model are converged to obtain a trained visual detection model;
and the output unit is used for inputting the acquired welding image to be detected into the trained visual detection model, judging whether the welding image to be detected has an unqualified welding section or not by the visual detection model, and outputting the welding process parameter range of the unqualified welding section, so that the parameter adjustment can be performed on the galvanometer welding joint according to the welding process parameter range of the unqualified welding section.
9. The apparatus of claim 8, wherein the processing unit is further configured to:
smoothing the welding image sample to obtain a denoised welding image sample;
and positioning a welding area in the welding image sample through Blob analysis, and performing second-order differential processing and Blob analysis on the welding area to obtain welding data of the welding image sample, wherein the welding data comprises at least one of the height of a welding section and the width of the welding section.
10. A readable storage medium comprising a stored program, wherein the program when executed controls a device on which the storage medium is located to perform the method for visual inspection based on laser welding of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110942117.1A CN115713476A (en) | 2021-08-17 | 2021-08-17 | Visual detection method and device based on laser welding and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110942117.1A CN115713476A (en) | 2021-08-17 | 2021-08-17 | Visual detection method and device based on laser welding and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115713476A true CN115713476A (en) | 2023-02-24 |
Family
ID=85229906
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110942117.1A Pending CN115713476A (en) | 2021-08-17 | 2021-08-17 | Visual detection method and device based on laser welding and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115713476A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115988866A (en) * | 2023-03-21 | 2023-04-18 | 深圳市利和兴股份有限公司 | NFC LAMI processing control method and system based on machine vision |
| CN116460495A (en) * | 2023-05-06 | 2023-07-21 | 深圳市万顺兴科技有限公司 | Welding processing method |
| CN116551263A (en) * | 2023-07-11 | 2023-08-08 | 苏州松德激光科技有限公司 | Visual control method and system for welding position selection |
| CN116818780A (en) * | 2023-05-26 | 2023-09-29 | 深圳市大德激光技术有限公司 | Visual 2D and 3D detection system for button cell shell after laser welding |
| CN118417724A (en) * | 2024-07-04 | 2024-08-02 | 宝鸡市永盛泰钛业有限公司 | Optimized production and processing method of zirconium alloy plate with constant temperature superplasticity |
-
2021
- 2021-08-17 CN CN202110942117.1A patent/CN115713476A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115988866A (en) * | 2023-03-21 | 2023-04-18 | 深圳市利和兴股份有限公司 | NFC LAMI processing control method and system based on machine vision |
| CN115988866B (en) * | 2023-03-21 | 2023-06-20 | 深圳市利和兴股份有限公司 | NFC LAMI processing control method and system based on machine vision |
| CN116460495A (en) * | 2023-05-06 | 2023-07-21 | 深圳市万顺兴科技有限公司 | Welding processing method |
| CN116818780A (en) * | 2023-05-26 | 2023-09-29 | 深圳市大德激光技术有限公司 | Visual 2D and 3D detection system for button cell shell after laser welding |
| CN116818780B (en) * | 2023-05-26 | 2024-03-26 | 深圳市大德激光技术有限公司 | Visual 2D and 3D detection system for button cell shell after laser welding |
| CN116551263A (en) * | 2023-07-11 | 2023-08-08 | 苏州松德激光科技有限公司 | Visual control method and system for welding position selection |
| CN116551263B (en) * | 2023-07-11 | 2023-10-31 | 苏州松德激光科技有限公司 | Visual control method and system for welding position selection |
| CN118417724A (en) * | 2024-07-04 | 2024-08-02 | 宝鸡市永盛泰钛业有限公司 | Optimized production and processing method of zirconium alloy plate with constant temperature superplasticity |
| CN118417724B (en) * | 2024-07-04 | 2024-10-01 | 宝鸡市永盛泰钛业有限公司 | Optimized production and processing method of zirconium alloy plate with constant temperature superplasticity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115713476A (en) | Visual detection method and device based on laser welding and readable storage medium | |
| US11915408B2 (en) | Methods of artificial intelligence-assisted infrastructure assessment using mixed reality systems | |
| CN111507976B (en) | Defect detection method and system based on multi-angle imaging | |
| US11536669B2 (en) | Detecting machining errors of a laser machining system using deep convolutional neural networks | |
| CN106803244B (en) | Defect identification method and system | |
| US20180052120A1 (en) | Image processing device, image processing method, and program | |
| CN110596120A (en) | Glass boundary defect detection method, device, terminal and storage medium | |
| US20230053085A1 (en) | Part inspection system having generative training model | |
| JP2021021669A (en) | Inspection assisting method, inspection assisting system, and inspection assisting program | |
| Guo et al. | Machine vision-based intelligent manufacturing using a novel dual-template matching: a case study for lithium battery positioning | |
| CN113781393A (en) | Screen defect detection method, device, equipment and storage medium | |
| ES2989011T3 (en) | Physics-based anomaly detection in formed metal parts | |
| JP7616891B2 (en) | Method and system for change detection using AR overlay | |
| CN118135179B (en) | Cascade identification method and device for cutting plate, computer equipment and storage medium | |
| JP4814116B2 (en) | Mounting board appearance inspection method | |
| KR20230036650A (en) | Defect detection method and system based on image patch | |
| Pham et al. | Automatic welding seam tracking and real-world coordinates identification with machine learning method | |
| CN118506338A (en) | Electronic device printed character recognition and detection method based on deep learning | |
| CN109559294B (en) | Method and device for detecting quality of circular hole of drop | |
| CN117495846B (en) | Image detection method, device, electronic equipment and storage medium | |
| CN118762019B (en) | Denture blank material taking method and device based on visual detection and computer equipment | |
| WO2024176319A1 (en) | Region-of-interest extraction device, method, and system | |
| US20230316492A1 (en) | Teacher data generating method and generating device | |
| US20240296661A1 (en) | Method of analyzing a component, method of training a system, apparatus, computer program, and computer-readable storage medium | |
| CN114648583B (en) | Method and system for dynamically positioning detection area of long-period sequence image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |