CN114240837A - Welding seam positioning method, device, equipment and storage medium - Google Patents
Welding seam positioning method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114240837A CN114240837A CN202111358559.8A CN202111358559A CN114240837A CN 114240837 A CN114240837 A CN 114240837A CN 202111358559 A CN202111358559 A CN 202111358559A CN 114240837 A CN114240837 A CN 114240837A
- Authority
- CN
- China
- Prior art keywords
- weld
- image
- target
- welding seam
- preset
- 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.)
- Withdrawn
Links
- 238000003466 welding Methods 0.000 title claims abstract description 190
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000000605 extraction Methods 0.000 claims abstract description 61
- 238000012549 training Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 claims description 12
- 238000003709 image segmentation Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 10
- 238000002372 labelling Methods 0.000 claims description 8
- 238000005520 cutting process Methods 0.000 claims description 7
- 238000003062 neural network model Methods 0.000 claims description 5
- 241001016380 Reseda luteola Species 0.000 claims 53
- 230000008569 process Effects 0.000 description 16
- 230000007613 environmental effect Effects 0.000 description 10
- 230000006870 function Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 241000872198 Serjania polyphylla Species 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000003416 augmentation Effects 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000003999 initiator Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000003209 petroleum derivative Substances 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a welding seam positioning method, a welding seam positioning device, welding seam positioning equipment and a storage medium, and belongs to the technical field of welding. The method comprises the steps of extracting the characteristics of a welding image of the welding pipeline through a preset welding line characteristic extraction model to obtain target welding line characteristic information, effectively positioning the position of a welding line in the welding pipeline, scoring the target welding line characteristic information through a preset scoring model after positioning is finished, comparing a target welding line score with a preset scoring threshold, and outputting the target welding line characteristic information when the target welding line score is larger than the preset scoring threshold so as to ensure more accurate position determination of the welding line position information in the welding pipeline.
Description
Technical Field
The invention relates to the technical field of welding, in particular to a welding seam positioning method, a welding seam positioning device, welding seam positioning equipment and a storage medium.
Background
With the rapid development of social economy in China, the use of a large amount of resources such as petroleum and natural gas, the demand of transport pipelines and large storage tanks is rising day by day, the traditional manual welding has low efficiency, high difficulty and high cost, and the process requirements of welding large pipelines are difficult to meet, so that a machine welding technology is derived, but in the machine welding process, environmental factors have great influence on welding points, the welding points are difficult to accurately position, the welding quality is influenced, and the efficiency is not high.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a welding seam positioning method, a welding seam positioning device, welding seam positioning equipment and a storage medium, and aims to solve the technical problems that in the prior art, in the welding process of a machine, environmental factors have great influence on a welding point, the welding point is difficult to be accurately positioned, the welding quality is influenced, and the efficiency is low.
In order to achieve the above object, the present invention provides a weld positioning method, comprising the steps of:
acquiring a welding seam image of a welded pipeline;
performing feature extraction on the weld image through a preset weld feature extraction model to obtain target weld feature information;
grading the target weld joint characteristic information through a preset grading model to obtain a target weld joint grading value;
and when the score value of the target weld joint is greater than a preset score threshold value, outputting the characteristic information of the target weld joint to a user side for displaying.
Optionally, the performing feature extraction on the weld image through a preset weld feature extraction model further includes, before obtaining target weld feature information:
acquiring a weld sample image and corresponding weld characteristic information;
and performing model training on an initial neural network model according to the welding seam sample image and the welding seam sample image to obtain a preset welding seam feature extraction model.
Optionally, the acquiring the weld sample image and the corresponding weld characteristic information includes:
acquiring an initial welding seam sample image;
performing feature enhancement on the initial weld sample image through a preset image feature enhancement model to obtain a weld sample image;
carrying out image segmentation on the welding seam sample image to obtain a welding seam characteristic image;
and extracting weld characteristic information from the weld characteristic image.
Optionally, the image segmentation is performed on the weld sample image to obtain a weld characteristic image, and the method includes:
carrying out image segmentation on the welding seam sample image to obtain a segmented image;
extracting weld contour information and background information in the segmented image;
and carrying out feature labeling on the welding seam outline information based on a preset first color, carrying out feature labeling on the background information based on a preset second color, and obtaining a welding seam feature image.
Optionally, the performing feature extraction on the weld image through a preset weld feature extraction model to obtain target weld feature information includes:
carrying out resolution adjustment on the welding seam image, and carrying out feature enhancement on the adjusted welding seam image to obtain a target welding seam image;
and performing feature extraction on the target weld image through a preset weld feature extraction model to obtain target weld information.
Optionally, the adjusting the resolution of the weld image and performing feature enhancement on the adjusted weld image to obtain a target weld image includes:
performing downsampling processing on the weld image to obtain a weld characteristic diagram;
acquiring a welding line gray-scale image, and determining area histogram information in the welding line gray-scale image;
and adjusting the brightness of the welding seam characteristic graph according to the regional histogram information to obtain a target welding seam image.
Optionally, the adjusting the brightness of the weld feature map according to the region histogram information to obtain a target weld image includes:
cutting the area histogram according to a preset cutting threshold value to obtain a target area histogram;
acquiring target area histogram information corresponding to the target area histogram;
and adjusting the brightness of the welding seam characteristic graph based on the target area histogram and the target area histogram information to obtain a target welding seam image.
In addition, in order to achieve the above object, the present invention further provides a weld positioning apparatus, including:
the image acquisition module is used for acquiring a welding seam image of the welded pipeline;
the characteristic extraction module is used for extracting the characteristics of the welding seam image through a preset welding seam characteristic extraction model to obtain target welding seam characteristic information;
the characteristic scoring module is used for scoring the target weld characteristic information through a preset scoring model to obtain a target weld scoring value;
and the information display module is used for outputting the target weld characteristic information to a user side for display when the score value of the target weld is greater than a preset score threshold value.
In addition, in order to achieve the above object, the present invention also provides a weld seam positioning apparatus, including: a memory, a processor, and a weld positioning program stored on the memory and executable on the processor, the weld positioning program configured to implement the steps of the weld positioning method as described above.
In addition, to achieve the above object, the present invention further provides a storage medium having a weld seam positioning program stored thereon, wherein the weld seam positioning program, when executed by a processor, implements the steps of the weld seam positioning method as described above.
The method comprises the steps of obtaining a weld image of a welded pipeline, carrying out feature extraction on the weld image through a preset weld feature extraction model to obtain target weld feature information, grading the target weld feature information through a preset grading model to obtain a target weld grade value, and outputting the target weld feature information to a user side for displaying when the target weld grade value is larger than a preset grading threshold value. Compared with the prior art, the invention extracts the characteristics of the welding image of the welding pipeline by the preset welding line characteristic extraction model to obtain the characteristic information of the target welding line, can effectively position the welding line position in the welding pipeline, and in addition, after the positioning is finished, the target weld joint feature information is graded through a preset grading model, the target weld joint grading value is compared with a preset grading threshold value, when the score value of the target weld joint is greater than the preset score threshold value, the characteristic information of the target weld joint is output to ensure that the position of the weld joint in the welded pipeline is more accurately determined, and the problems that in the machine welding process, the influence of environmental factors on welding points is great, the welding points are difficult to be accurately positioned, the welding quality is influenced, the efficiency is not high, and the welding quality and the working efficiency of machine welding pipelines are improved.
Drawings
FIG. 1 is a schematic diagram of a weld seam positioning apparatus for a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a first embodiment of a weld positioning method of the present invention;
FIG. 3 is a schematic view of a weld joint display image according to an embodiment of the weld joint positioning method of the present invention;
FIG. 4 is a schematic flow chart of a second embodiment of the weld seam positioning method of the present invention;
FIG. 5 is a schematic flow chart of a third embodiment of a weld locating method of the present invention;
FIG. 6 is a schematic diagram illustrating a change in a pixel gray value of a weld image according to an embodiment of the weld positioning method of the present invention;
fig. 7 is a block diagram showing the structure of the first embodiment of the weld seam positioning apparatus of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a weld joint positioning device in a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the weld seam positioning apparatus may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in FIG. 1 does not constitute a limitation of the weld positioning apparatus and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a storage medium, may include therein an operating system, a network communication module, a user interface module, and a weld positioning program.
In the weld seam positioning apparatus shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 of the weld positioning apparatus of the present invention may be provided in the weld positioning apparatus, and the weld positioning apparatus calls the weld positioning program stored in the memory 1005 through the processor 1001 and executes the weld positioning method provided by the embodiment of the present invention.
The embodiment of the invention provides a weld joint positioning method, and referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the weld joint positioning method.
In this embodiment, the weld positioning method includes the following steps:
step S10: and acquiring a welding seam image of the welded pipeline.
It should be noted that the execution main body of the embodiment is the weld positioning device, where the weld positioning device has functions of data processing, data communication, program operation, and the like, and the weld positioning device may be an integrated controller, a control computer, a palm computer, and the like, and of course, may also be other devices having similar functions, and the embodiment is not particularly limited.
It can be understood that the weld image is an image acquired by an image acquisition device in the welding pipe that needs to perform welding work, where the image acquisition device may be an electronic device such as a camera or an image acquisition card, or may be other electronic devices that have the function of acquiring the weld image in the weld pipe, and this embodiment is not particularly limited.
Step S20: and performing feature extraction on the welding seam image through a preset welding seam feature extraction model to obtain target welding seam feature information.
It should be noted that the preset weld feature extraction model is used for performing feature extraction on the weld image acquired by the image acquisition device, that is, extracting weld information in the weld image to obtain target weld feature information.
It can be understood that, since the interference of the environmental noise such as strong arc light or spatter may occur during the process of welding the pipeline, the target weld characteristic image may be a clear weld display image obtained after image segmentation and image enhancement are performed on the weld area in the image weld pipeline, and there is an obvious definition between a general weld and the background in the weld display image, referring to fig. 3, the display image is divided by different colors, for example: the position of a welding line in the image is marked and displayed in yellow; while the background portion is shown in black with a label.
In a specific implementation, the preset weld feature extraction model may be an improved U-Net model, and compared with an original U-Net network model, the improved U-Net model used in this embodiment is additionally provided with a feature enhancement module and a learnable adjustor, where the feature enhancement module is used to enhance the accuracy of model training and prevent overfitting due to too single data set during training; in addition, the learnable adjustor is used for searching for a proper image resolution ratio during training, and since the picture size adjustment is usually performed by using a fixed method during the training of the U-Net network model, the learnable adjustor has great limitation.
Step S30: and grading the characteristic information of the target weld joint through a preset grading model to obtain a target weld joint grading value.
It is worth to be noted that the preset scoring model is used for scoring the image subjected to the weld joint feature extraction to obtain a target weld joint score value, whether the feature extraction is accurate or not is judged according to the size of the target weld joint score value, if the target weld joint score value is too low, the feature extraction can be performed again, and the accuracy of the weld joint position positioning is improved.
It should be understood that before the target weld characteristic information is scored, a scoring rule needs to be determined, and in this embodiment, the target weld characteristic information may be analyzed according to one or more evaluation indexes of three indexes, namely, Accuracy (Accuracy, which is a ratio of a segmented correct weld characteristic image to the whole image), Mean Intersection over Union (mIou), and Mean Pixel Accuracy (mpa); the average intersection and union ratio is the ratio of the intersection and union of the two sets of the real value and the predicted value, and each IOU is calculated and then accumulated and averaged; average pixel accuracy refers to the proportion of the number of pixels in each class that are correctly classified.
In addition, the calculation formulas of the Accuracy (Accuracy, Acc), the average Intersection ratio (MIOU), and the average Pixel precision (MPA) are:
wherein TP is a weld characteristic pixel which is correctly segmented; TN is a welding seam background pixel with correct segmentation; FP is a welding seam background pixel with wrong segmentation; FN is weld seam characteristic pixel with wrong segmentation.
Step S40: and when the score value of the target weld joint is greater than a preset score threshold value, outputting the characteristic information of the target weld joint to a user side for displaying.
It should be noted that the preset scoring threshold may be a numerical value set by a user according to model training, or may be a numerical value automatically generated by the weld seam positioning device controller according to training data in the model training process, and when scoring is performed according to different evaluation rules, the corresponding preset scoring threshold is different, which is not limited in this embodiment.
In a specific implementation, if the score of the target weld is greater than the preset score threshold, the feature extraction is satisfactory, that is, the weld position is accurate, at this time, the target weld feature information may be output to a user side for display, and the user side may be an electronic device having functions of data transmission or image display, or other devices having the same or similar functions, which is not limited in this embodiment.
In addition, after positioning is completed, the target weld joint feature information is scored through a preset scoring model, the target weld joint score value is compared with a preset scoring threshold value, and when the target weld joint score value is larger than the preset scoring threshold value, the target weld joint feature information is output to ensure that more accurate position determination is carried out on the weld joint position information in the welded pipeline, the technical problem that in the machine welding process, environmental factors have great influence on a welding point, accurate positioning is difficult to carry out on the welding point, welding quality is affected, efficiency is low is solved, and the welding quality and the working efficiency of the machine welded pipeline are improved.
Referring to fig. 4, fig. 4 is a schematic flow chart of a weld positioning method according to a second embodiment of the present invention.
Based on the first embodiment, in this embodiment, before the step S20, the method further includes:
step S1: and acquiring a weld sample image and corresponding weld characteristic information.
It should be noted that, because the preset feature extraction model used in this embodiment is an improved U-Net model, and because there is interference of environmental factors such as arc interference in the welding seam positioning process, a large number of welding seam sample images and corresponding welding seam feature information need to be collected during model training, and in this embodiment, the collected data samples can be fully utilized through data enhancement.
Further, in order to obtain a weld sample image and corresponding weld characteristic information, step S1 includes:
acquiring an initial welding seam sample image;
performing feature enhancement on the initial weld sample image through a preset image feature enhancement model to obtain a weld sample image;
carrying out image segmentation on the welding seam sample image to obtain a welding seam characteristic image;
and extracting weld characteristic information from the weld characteristic image.
It can be understood that the initial weld sample image, i.e. the image captured by the camera, needs to be adjusted and enhanced because there may be interference of environmental factors such as arc interference in the image.
In this embodiment, data enhancement will be performed by two methods: 1. simulated occlusion (Cutout); 2. automatic data enhancement (AutoAutoAutoAutoAutoAutoAutoAutoAutoAutoAutomation).
The simulated shielding (Cutout) is to cut and expand images on the premise of not changing the image precision, namely the simulated shielding (Cutout) can fully utilize other contents in a scene through the scene in which the weld joint features are shielded in the simulated welding process, and prevent a network from only paying attention to the obvious feature region, so that the occurrence of the overfitting condition is avoided.
In addition, automatic data enhancement (automation) is an augmentation scheme for finding a proper weld image through a search algorithm in a search space of a series of image augmentation sub-strategies, and random rotation, mirror image, translation and other operations are performed on the weld image, so that the image is augmented.
In addition, in order to obtain a clear weld joint feature image, the weld joint sample image can be subjected to image segmentation to obtain a segmented image, weld joint outline information and background information in the segmented image are extracted, feature labeling is carried out on the weld joint outline information based on a preset first color, feature labeling is carried out on the background information based on a preset second color, and the weld joint feature image is obtained.
It can be understood that the preset first color and the preset second color can be set by a user, and the embodiment is not particularly limited.
In a specific implementation, in order to show the weld characteristics more clearly, it is necessary to clearly distinguish between the weld and the background in the weld sample image, which is represented by dividing in different colors in the display image, for example: the position of a welding line in the image is marked and displayed in yellow; while the background portion is shown in black with a label.
Step S2: and performing model training on an initial neural network model according to the welding seam sample image and the welding seam sample image to obtain a preset welding seam feature extraction model.
It should be noted that a weld sample image and a weld sample image are obtained, model training is performed on the initial neural network model according to the weld sample image and the weld sample image, and a preset weld feature extraction model is obtained, where the preset weld feature extraction model may be an improved U-Net model, and the improved U-Net model includes a adjuster Module (RM), an encoding Module, and a decoding Module.
The adjuster module is used for adjusting the size of the input image, can better adapt to welding seam images of various sizes, and improves the adaptability of the model.
It is understood that the first convolution layer and the pooling layer exist in the encoding module; the decoding module has a second convolutional layer and a deconvolution layer, and after each convolutional layer, a Batch Normalization layer (BN) may be inserted, and the Batch Normalization layer uses an activation function as an initiator, in this embodiment, the activation function is a leak Relu function.
It should be noted that, in the encoding module and the decoding module, the detailed parameters of the size of each layer of convolution kernel, the moving step of convolution kernel, and the number of layers of zero padding are referred to table 1.
TABLE 1
In addition, compared with the original U-Net network model, the improved U-Net model used in this embodiment adds a feature enhancement module and a learnable adjustor, which can perform feature extraction on an image by using 4 convolution layers and 1 residual block, wherein a Bilinear adjustor (BR) can merge weld features calculated by the original resolution into the subsequent network model, and several skip connections used in the structure of the adjustor can make the learnable adjustor fuse more image features so as to perform learning more easily.
The invention extracts the characteristics of the welding image of the welding pipeline through the preset welding seam characteristic extraction model to obtain the target welding seam characteristic information, can effectively position the welding seam position in the welding pipeline, trains the initial neural network through the welding seam sample image and the corresponding welding seam characteristic information to obtain the proper preset characteristic extraction model, scores the target welding seam characteristic information through the preset scoring model after the positioning is finished, compares the target welding seam score value with the preset scoring threshold value, outputs the target welding seam characteristic information when the target welding seam score value is greater than the preset scoring threshold value to ensure that the more accurate position of the welding seam position information in the welding pipeline is determined, and avoids that environmental factors have great influence on the welding point and are difficult to accurately position the welding point in the machine welding process, the welding quality is influenced, the efficiency is not high, and the welding quality and the working efficiency of the machine for welding the pipeline are improved.
Referring to fig. 5, fig. 5 is a schematic flow chart of a weld positioning method according to a third embodiment of the present invention.
Based on the second embodiment, in this embodiment, the step S20 includes:
step S201: and adjusting the resolution of the welding seam image, and performing characteristic enhancement on the adjusted welding seam image to obtain a target welding seam image.
It should be noted that, because the picture size adjustment is usually performed by using a fixed method during the training of the U-Net network model, the method has great limitation, and the resolution adjustment in this embodiment can well solve the problem and improve the working efficiency.
Further, in order to adjust the resolution of the weld image, step S201 further includes:
performing downsampling processing on the weld image to obtain a weld characteristic diagram;
acquiring a welding line gray-scale image, and determining area histogram information in the welding line gray-scale image;
and adjusting the brightness of the welding seam characteristic graph according to the regional histogram information to obtain a target welding seam image.
The downsampling is to perform the operation of reducing the resolution of the weld image, and due to the existence of the feature enhancement module, the downsampled weld feature map can still be subjected to feature extraction through the preset weld feature extraction model, so as to obtain the position information of the weld.
It is understood that the region histogram information is information of the histogram of each region in the weld gray map.
In a specific implementation, the weld characteristic map is subjected to contrast-limited adaptive histogram equalization processing according to a first convolution layer, a pooling layer, a second convolution layer and a deconvolution layer which are arranged in an encoding module and a decoding module, and a histogram of each region of the weld gray-scale image subjected to downsampling is calculated through the histogram equalization processing, so that region histogram information in the weld gray-scale image is obtained.
In addition, in order to obtain accurate region histogram information, the region histogram can be cut according to a preset cutting threshold value to obtain a target region histogram, target region histogram information corresponding to the target region histogram is obtained, and a weld characteristic map is subjected to brightness adjustment based on the target region histogram and the target region histogram information to obtain a target weld image.
In specific implementation, referring to fig. 6, a dotted line is a pixel gray value of the weld image after the histogram equalization processing diagram is processed, a solid line is a pixel gray value of the weld image after the histogram equalization processing diagram is not processed, and the histogram is cut by presetting a cutting threshold value through the histogram equalization processing diagram, so that the distribution of the pixel gray value of the weld image is more balanced, the edge information of the weld characteristic diagram is enhanced, and the weld segmentation efficiency is higher.
Step S202: and performing feature extraction on the target weld image through a preset weld feature extraction model to obtain target weld information.
The invention extracts the characteristics of the welding image of the welding pipeline through the preset welding seam characteristic extraction model to obtain the target welding seam characteristic information, can effectively position the welding seam position in the welding pipeline, trains the initial neural network through the welding seam sample image and the corresponding welding seam characteristic information to obtain a proper preset characteristic extraction model, obtains the target welding seam image with proper resolution by adjusting the resolution of the welding seam image to adapt to the characteristic extraction of different image qualities, scores the target welding seam characteristic information through the preset scoring model after the positioning is finished, compares the target welding seam score value with the preset scoring threshold value, outputs the target welding seam characteristic information when the target welding seam score value is larger than the preset scoring threshold value to ensure more accurate position determination of the welding seam position information in the welding pipeline, the welding machine has the advantages that the technical problems that in the welding process of the machine, environmental factors greatly affect welding points, the welding points are difficult to be accurately positioned, the welding quality is affected, and the efficiency is low are solved, and the welding quality and the working efficiency of a machine welding pipeline are improved.
In addition, an embodiment of the present invention further provides a storage medium, where the storage medium stores a weld positioning program, and the weld positioning program, when executed by a processor, implements the steps of the weld positioning method described above.
Since the storage medium adopts all the technical solutions of all the embodiments, at least all the advantages brought by the technical solutions of the embodiments are available, and are not described in detail herein.
Referring to fig. 7, fig. 7 is a block diagram illustrating a first embodiment of a weld seam positioning device according to the present invention.
As shown in fig. 7, a weld positioning apparatus according to an embodiment of the present invention includes:
the image acquisition module 10 is used for acquiring a welding seam image of a welded pipeline;
the feature extraction module 20 is configured to perform feature extraction on the weld image through a preset weld feature extraction model to obtain target weld feature information;
the feature scoring module 30 is configured to score the target weld feature information through a preset scoring model to obtain a target weld score value;
and the information display module 40 is used for outputting the target weld characteristic information to a user side for displaying when the score value of the target weld is greater than a preset score threshold value.
In addition, after positioning is completed, the target weld joint feature information is scored through a preset scoring model, the target weld joint score value is compared with a preset scoring threshold value, and when the target weld joint score value is larger than the preset scoring threshold value, the target weld joint feature information is output to ensure that more accurate position determination is carried out on the weld joint position information in the welded pipeline, the technical problem that in the machine welding process, environmental factors have great influence on a welding point, accurate positioning is difficult to carry out on the welding point, welding quality is affected, efficiency is low is solved, and the welding quality and the working efficiency of the machine welded pipeline are improved.
In an embodiment, the feature extraction module 20 is further configured to obtain a weld sample image and corresponding weld feature information; and performing model training on an initial neural network model according to the welding seam sample image and the welding seam sample image to obtain a preset welding seam feature extraction model.
In an embodiment, the feature extraction module 20 is further configured to obtain an initial weld sample image; performing feature enhancement on the initial weld sample image through a preset image feature enhancement model to obtain a weld sample image; carrying out image segmentation on the welding seam sample image to obtain a welding seam characteristic image; and extracting weld characteristic information from the weld characteristic image.
In an embodiment, the feature extraction module 20 is further configured to perform image segmentation on the weld sample image to obtain a segmented image; extracting weld contour information and background information in the segmented image; and carrying out feature labeling on the welding seam outline information based on a preset first color, carrying out feature labeling on the background information based on a preset second color, and obtaining a welding seam feature image.
In an embodiment, the feature extraction module 20 is further configured to perform resolution adjustment on the weld image, and perform feature enhancement on the adjusted weld image to obtain a target weld image; and performing feature extraction on the target weld image through a preset weld feature extraction model to obtain target weld information.
In an embodiment, the feature extraction module 20 is further configured to perform downsampling on the weld image to obtain a weld feature map; acquiring a welding line gray-scale image, and determining area histogram information in the welding line gray-scale image; and adjusting the brightness of the welding seam characteristic graph according to the regional histogram information to obtain a target welding seam image.
In an embodiment, the feature extraction module 20 is further configured to cut the region histogram according to a preset cutting threshold, so as to obtain a target region histogram; acquiring target area histogram information corresponding to the target area histogram; and adjusting the brightness of the welding seam characteristic graph based on the target area histogram and the target area histogram information to obtain a target welding seam image.
It should be understood that the above is only an example, and the technical solution of the present invention is not limited in any way, and in a specific application, a person skilled in the art may set the technical solution as needed, and the present invention is not limited thereto.
It should be noted that the above-described work flows are only exemplary, and do not limit the scope of the present invention, and in practical applications, a person skilled in the art may select some or all of them to achieve the purpose of the solution of the embodiment according to actual needs, and the present invention is not limited herein.
In addition, the technical details that are not described in detail in this embodiment may refer to the weld positioning method provided in any embodiment of the present invention, and are not described herein again.
Further, it is to be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention or portions thereof that contribute to the prior art may be embodied in the form of a software product, where the computer software product is stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk), and includes several instructions for enabling a terminal device (e.g. a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A weld positioning method, comprising:
acquiring a welding seam image of a welded pipeline;
performing feature extraction on the weld image through a preset weld feature extraction model to obtain target weld feature information;
grading the target weld joint characteristic information through a preset grading model to obtain a target weld joint grading value;
and when the score value of the target weld joint is greater than a preset score threshold value, outputting the characteristic information of the target weld joint to a user side for displaying.
2. The weld positioning method according to claim 1, wherein before the feature extraction of the weld image by using a preset weld feature extraction model to obtain target weld feature information, the method further comprises:
acquiring a weld sample image and corresponding weld characteristic information;
and performing model training on an initial neural network model according to the welding seam sample image and the welding seam sample image to obtain a preset welding seam feature extraction model.
3. The weld positioning method according to claim 2, wherein the acquiring of the weld specimen image and the corresponding weld characteristic information comprises:
acquiring an initial welding seam sample image;
performing feature enhancement on the initial weld sample image through a preset image feature enhancement model to obtain a weld sample image;
carrying out image segmentation on the welding seam sample image to obtain a welding seam characteristic image;
and extracting weld characteristic information from the weld characteristic image.
4. The weld seam positioning method according to claim 3, wherein the image segmentation of the weld seam sample image to obtain a weld seam feature image comprises:
carrying out image segmentation on the welding seam sample image to obtain a segmented image;
extracting weld contour information and background information in the segmented image;
and carrying out feature labeling on the welding seam outline information based on a preset first color, carrying out feature labeling on the background information based on a preset second color, and obtaining a welding seam feature image.
5. The weld positioning method according to claim 4, wherein the performing feature extraction on the weld image through a preset weld feature extraction model to obtain target weld feature information comprises:
carrying out resolution adjustment on the welding seam image, and carrying out feature enhancement on the adjusted welding seam image to obtain a target welding seam image;
and performing feature extraction on the target weld image through a preset weld feature extraction model to obtain target weld information.
6. The weld joint positioning method according to claim 5, wherein the adjusting the resolution of the weld joint image and performing feature enhancement on the adjusted weld joint image to obtain a target weld joint image comprises:
performing downsampling processing on the weld image to obtain a weld characteristic diagram;
acquiring a welding line gray-scale image, and determining area histogram information in the welding line gray-scale image;
and adjusting the brightness of the welding seam characteristic graph according to the regional histogram information to obtain a target welding seam image.
7. The weld joint positioning method according to claim 6, wherein the brightness adjustment of the weld joint feature map according to the region histogram information to obtain a target weld joint image comprises:
cutting the area histogram according to a preset cutting threshold value to obtain a target area histogram;
acquiring target area histogram information corresponding to the target area histogram;
and adjusting the brightness of the welding seam characteristic graph based on the target area histogram and the target area histogram information to obtain a target welding seam image.
8. A weld seam positioning device, comprising:
the image acquisition module is used for acquiring a welding seam image of the welded pipeline;
the characteristic extraction module is used for extracting the characteristics of the welding seam image through a preset welding seam characteristic extraction model to obtain target welding seam characteristic information;
the characteristic scoring module is used for scoring the target weld characteristic information through a preset scoring model to obtain a target weld scoring value;
and the information display module is used for outputting the target weld characteristic information to a user side for display when the score value of the target weld is greater than a preset score threshold value.
9. A weld seam positioning apparatus, comprising: a memory, a processor, and a weld positioning program stored on the memory and executable on the processor, the weld positioning program configured to implement the weld positioning method of any one of claims 1 to 7.
10. A storage medium having a weld positioning program stored thereon, the weld positioning program, when executed by a processor, implementing the weld positioning method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111358559.8A CN114240837A (en) | 2021-11-16 | 2021-11-16 | Welding seam positioning method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111358559.8A CN114240837A (en) | 2021-11-16 | 2021-11-16 | Welding seam positioning method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114240837A true CN114240837A (en) | 2022-03-25 |
Family
ID=80749689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111358559.8A Withdrawn CN114240837A (en) | 2021-11-16 | 2021-11-16 | Welding seam positioning method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114240837A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115122005A (en) * | 2022-07-27 | 2022-09-30 | 广东省源天工程有限公司 | Ultra-large type miter gate door body welding device |
-
2021
- 2021-11-16 CN CN202111358559.8A patent/CN114240837A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115122005A (en) * | 2022-07-27 | 2022-09-30 | 广东省源天工程有限公司 | Ultra-large type miter gate door body welding device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8019164B2 (en) | Apparatus, method and program product for matching with a template | |
| CN109671058B (en) | Defect detection method and system for large-resolution image | |
| CN114266773B (en) | Display panel defect positioning method, device, equipment and storage medium | |
| CN106202086B (en) | Picture processing and obtaining method, device and system | |
| CN114155244B (en) | Defect detection method, device, equipment and storage medium | |
| CN110245747B (en) | Image processing method and device based on full convolution neural network | |
| CN110737785B (en) | Picture labeling method and device | |
| CN111144215B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN114170227B (en) | Product surface defect detection method, device, equipment and storage medium | |
| CN112819748B (en) | Training method and device for strip steel surface defect recognition model | |
| CN113657409A (en) | Vehicle loss detection method, device, electronic device and storage medium | |
| CN114782286A (en) | Defect repairing method, optical repairing device, electronic device and storage medium | |
| CN111275051A (en) | Character recognition method, character recognition device, computer equipment and computer-readable storage medium | |
| CN114663904A (en) | PDF document layout detection method, device, equipment and medium | |
| CN115713476A (en) | Visual detection method and device based on laser welding and readable storage medium | |
| CN113705673A (en) | Character detection method, device, equipment and storage medium | |
| CN113505781A (en) | Target detection method and device, electronic equipment and readable storage medium | |
| CN114240837A (en) | Welding seam positioning method, device, equipment and storage medium | |
| CN111340025A (en) | Character recognition method, character recognition device, computer equipment and computer-readable storage medium | |
| CA3135111A1 (en) | Character positioning method and system for certificate document | |
| CN113537037A (en) | Pavement disease identification method, system, electronic device and storage medium | |
| CN113221897B (en) | Image correction method, image text recognition method, identity verification method and device | |
| CN115880288B (en) | Detection method, system and computer equipment for electronic element welding | |
| CN116596903A (en) | Defect identification method, device, electronic equipment and readable storage medium | |
| CN112927321B (en) | Intelligent image design method, device, equipment and storage medium based on neural network |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220325 |