CN114299297A

CN114299297A - Method, device, equipment and medium for identifying weld defects

Info

Publication number: CN114299297A
Application number: CN202111629245.7A
Authority: CN
Inventors: 王世宏; 车飞; 朱丽丽; 王一帆
Original assignee: BEIJING WEST TUBE INSPECTION TECHNOLOGY CO LTD
Current assignee: BEIJING WEST TUBE INSPECTION TECHNOLOGY CO LTD
Priority date: 2021-12-28
Filing date: 2021-12-28
Publication date: 2022-04-08

Abstract

The embodiment of the application provides a method, a device, equipment and a medium for identifying weld defects, wherein the method comprises the following steps: acquiring a weld map of a pipe to be identified; extracting the characteristics of the welding line atlas, and comparing the characteristics of the welding line atlas with the characteristics of a qualified atlas to obtain comparison characteristics, wherein the qualified atlas is an atlas corresponding to a qualified welding line; and confirming whether the welding seam of the pipe to be identified has defects or not through the comparison characteristics. Some embodiments of the application can carry out automatic identification to the welding seam defect to practice thrift a large amount of manpower resources, improve work efficiency, reduce uncertainty and the identification error that produces because of the human factor.

Description

Method, device, equipment and medium for identifying weld defects

Technical Field

The embodiment of the application relates to the field of material defect identification, in particular to a method, a device, equipment and a medium for identifying weld defects.

Background

In the related art, a weld is a portion of a pipe body of a pipe, where a defect is likely to exist, and the type of a typical defect is usually determined manually, but in the process of judging the type of the defect, if the characteristics of the two types of defects are relatively close, an error is easily identified, manual identification is used, so that the labor cost is increased, and the accuracy and the identification speed of identification cannot be guaranteed.

Therefore, how to improve the accuracy of the weld defect identification becomes an urgent problem to be solved.

Disclosure of Invention

The embodiment of the application provides a method, a device, equipment and a medium for identifying welding seam defects, and the welding seam defects can be automatically identified at least through some embodiments of the application, so that a large amount of manpower resources are saved, the working efficiency is improved, and the uncertainty and the identification error caused by human factors are reduced.

In a first aspect, the present application provides a method of weld defect identification, the method comprising: acquiring a weld map of a pipe to be identified; extracting the characteristics of the welding line atlas, and comparing the characteristics of the welding line atlas with the characteristics of a qualified atlas to obtain comparison characteristics, wherein the qualified atlas is an atlas corresponding to a qualified welding line; and confirming whether the welding seam of the pipe to be identified has defects or not through the comparison characteristics.

Therefore, the welding seam defect can be automatically identified through the welding seam defect identification method, so that a large amount of manpower resources are saved, the working efficiency is improved, and the uncertainty and the identification error caused by human factors are reduced.

With reference to the first aspect, in an embodiment of the application, the determining whether the weld has the defect through the comparison feature includes: and inquiring the comparison characteristics in a defect characteristic library, and confirming whether the welding seam has defects, wherein the defect characteristic library is used for representing the mapping relation between the characteristics and the welding seam defects.

Therefore, the embodiment of the application can obtain a more accurate defect type by comparing the comparison features in the defect features.

With reference to the first aspect, in an embodiment of the present application, before querying the comparison feature in the defect feature library to determine whether the weld has a defect, the method further includes: extracting positive features of the qualified map, wherein the positive features are features of a weld joint in the qualified map; extracting negative features of a defect map, wherein the negative features are features corresponding to various types of defect welding lines in the defect map, and the defect map is a map corresponding to the defect welding lines; and establishing the defect feature library through the positive features and the negative features.

Therefore, the method and the device can extract the features in the qualified map by acquiring the positive features of the qualified map, so that the follow-up comparison with the negative features is more accurate, accurate and detailed difference features are acquired, and an accurate defect feature library is established.

With reference to the first aspect, in an embodiment of the present application, the creating the defect feature library by using the positive features and the negative features includes: comparing the positive features with the negative features to obtain difference features; and establishing a mapping relation between the difference characteristics and the weld defects to obtain the defect characteristic library.

With reference to the first aspect, in one embodiment of the present application, the plurality of features includes a weld length and a weld width; the extracting of the forward characteristics of the qualified atlas comprises the following steps: taking the number of pixels meeting a gray threshold value in pixel points corresponding to the welding seam length region in the welding seam region of the qualified map as the forward characteristic of the welding seam length; and taking the number of pixels meeting the gray threshold value in the pixel points corresponding to the weld width area in the weld area as the forward characteristic of the weld width.

With reference to the first aspect, in an embodiment of the present application, the extracting negative features of the defect map includes: and extracting multiple groups of negative features corresponding to the multiple defect maps, wherein each group of negative features comprises a weld length negative feature and a weld width negative feature.

Therefore, various types of defect features can be obtained by extracting multiple sets of negative features corresponding to multiple defect maps, and therefore the accuracy of identification is guaranteed.

With reference to the first aspect, in an embodiment of the application, the comparing the positive features with the negative features to obtain difference features includes: comparing the multiple sets of negative features with the positive features to obtain multiple corresponding difference features; the obtaining the defect feature library according to the difference features comprises: identifying a defect type corresponding to the plurality of difference features; and binding the plurality of difference characteristics with the defect type to obtain the defect characteristic library.

Therefore, the embodiment of the application can obtain the difference characteristics corresponding to different defects by comparing the positive characteristics with the negative characteristics, so that the defect type can be accurately obtained in the identification process.

With reference to the first aspect, in an embodiment of the present application, the creating the defect feature library by using the positive features and the negative features includes: inputting the forward characteristic into a first model to obtain a forward standard characteristic; inputting the negative features and the positive standard features into a second model to obtain difference features between the negative features and the positive standard features; and establishing the defect feature library through the difference features.

With reference to the first aspect, in one embodiment of the present application, a part of the defect maps are generated by a countermeasure network.

Therefore, in the embodiment of the present application, part of the defect maps are generated by the countermeasure network. The defect maps are generated through the countermeasure network, so that the number of samples can be increased, negative features are enriched, a defect sample library is enriched, and the accuracy of defect identification is improved.

With reference to the first aspect, in an embodiment of the present application, after the obtaining the weld map of the pipe to be identified, the method further includes: preprocessing the welding line map of the pipe to be identified to obtain the processed welding line map of the pipe to be identified; the method comprises the following steps of extracting the characteristics of the welding line atlas of the pipe to be identified, comparing the characteristics of the welding line atlas of the pipe to be identified with the characteristics of a qualified atlas to obtain comparison characteristics, and comprises the following steps: extracting the characteristics of the weld map of the processed pipe to be identified, and comparing the weld map of the processed pipe to be identified with the qualified map to obtain comparison characteristics.

Therefore, the embodiment of the application can remove the noise in the collected map by preprocessing the welding line map of the pipe to be identified, so that the feature extraction is more accurate after the influence of the noise is removed, and the identification accuracy can be increased.

With reference to the first aspect, in an embodiment of the application, the determining whether the weld has a defect through the comparison feature includes: confirming that the welding seam has defects through the comparison characteristics; and confirming the position of the defect, wherein the position is determined by the X-axis size and the Y-axis size.

With reference to the first aspect, in one embodiment of the present application, after confirming the weld is defective through the comparison feature, the method further includes: and confirming the defect grade corresponding to the defect type according to the defect type and the preset grade so that maintenance personnel can repair the damage to the pipe according to the defect grade and the defect type.

Therefore, according to the embodiment of the application, after the defect types are obtained according to the comparison characteristics, the defect types are corresponding to the corresponding defect grades, so that maintenance personnel can determine the severity of the defects, the damage repairing time and the like are determined according to the severity, meanwhile, a damage repairing method is obtained according to the determined defect types, and the pipe is repaired by using the damage repairing method.

In a second aspect, the present application provides an apparatus for atlas defect identification, the apparatus comprising: the map acquisition module is configured to acquire a weld map of the pipe to be identified; the map comparison module is configured to extract the features of the welding seam map and compare the features of the welding seam map with the features of a qualified map to obtain comparison features, wherein the qualified map is a map corresponding to a qualified welding seam; and the result confirmation module is configured to confirm whether the welding seam of the pipe to be identified has defects or not through the comparison characteristics.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: and inquiring the comparison characteristics in a defect characteristic library, and confirming whether the welding seam has defects, wherein the defect characteristic library is used for representing the mapping relation between the characteristics and the welding seam defects.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: extracting positive features of the qualified map, wherein the positive features are features of a weld joint in the qualified map; extracting negative features of a defect map, wherein the negative features are features corresponding to various types of defect welding lines in the defect map, and the defect map is a map corresponding to the defect welding lines; and establishing the defect feature library through the positive features and the negative features.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: comparing the positive features with the negative features to obtain difference features; and establishing a mapping relation between the difference characteristics and the weld defects to obtain the defect characteristic library.

In one embodiment of the present application, in combination with the second aspect, the plurality of features includes a weld length and a weld width; the result validation module is further configured to: taking the number of pixels meeting a gray threshold value in pixel points corresponding to the welding seam length region in the welding seam region of the qualified map as the forward characteristic of the welding seam length; and taking the number of pixels meeting the gray threshold value in the pixel points corresponding to the weld width area in the weld area as the forward characteristic of the weld width.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: and extracting multiple groups of negative features corresponding to the multiple defect maps, wherein each group of negative features comprises a weld length negative feature and a weld width negative feature.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: comparing the multiple sets of negative features with the positive features to obtain multiple corresponding difference features; the obtaining the defect feature library according to the difference features comprises: identifying a defect type corresponding to the plurality of difference features; and binding the plurality of difference characteristics with the defect type to obtain the defect characteristic library.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: inputting the forward characteristic into a first model to obtain a forward standard characteristic; inputting the negative features and the positive standard features into a second model to obtain difference features between the negative features and the positive standard features; and establishing the defect feature library through the difference features.

With reference to the second aspect, in one embodiment of the present application, a portion of the defect maps are generated by a countermeasure network.

With reference to the second aspect, in one embodiment of the present application, the atlas acquisition module is further configured to: preprocessing the welding line map of the pipe to be identified to obtain the processed welding line map of the pipe to be identified; the extracting features of the weld map of the pipe to be identified, and the map comparing module is further configured to: extracting the characteristics of the weld map of the processed pipe to be identified, and comparing the weld map of the processed pipe to be identified with the qualified map to obtain comparison characteristics.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: confirming that the welding seam has defects through the comparison characteristics; and confirming the position of the defect, wherein the position is determined by the X-axis size and the Y-axis size.

With reference to the second aspect, in an embodiment of the application, the result confirmation module is further configured to: and confirming the defect grade corresponding to the defect type according to the defect type and the preset grade so that maintenance personnel can repair the damage to the pipe according to the defect grade and the defect type.

In a third aspect, the present application provides an electronic device, comprising: a processor, a memory, and a bus; the processor is connected to the memory via the bus, the memory storing computer readable instructions for implementing the method as described in the first aspect and any embodiments thereof when the computer readable instructions are executed by the processor.

In a fourth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which, when executed, implements the method as described in the first aspect and any implementation thereof.

Drawings

FIG. 1 is a schematic diagram illustrating a system for identifying weld defects according to an embodiment of the present disclosure;

FIG. 2 is a flowchart of a method for identifying weld defects according to an embodiment of the present disclosure;

FIG. 3 is a schematic illustration of a qualifying map as shown in an embodiment of the present application;

FIG. 4 is a schematic diagram of a defect map shown in an embodiment of the present application;

FIG. 5 is a flowchart illustrating a method for creating a defect feature library according to an embodiment of the present application;

FIG. 6 is a flowchart illustrating one embodiment of weld defect identification according to an embodiment of the present disclosure;

FIG. 7 is a block diagram of an apparatus for identifying weld defects according to an embodiment of the present disclosure;

fig. 8 is an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as presented in the figures, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

The method and the device for identifying the weld joint defects can be applied to scenes for identifying the weld joint defects of the pipes, and in order to solve the problems in the background art, in some embodiments of the method and the device for identifying the weld joint defects, the weld joint defects are determined by extracting comparison characteristics between the weld joint and a qualified map. For example, in some embodiments of the present application, a library of defective features that includes defective features is first created or directly obtained. In the process of identifying the welding line atlas, firstly, extracting the characteristics of the welding line atlas, and comparing the characteristics of the welding line atlas with the characteristics of a qualified atlas to obtain comparison characteristics; and then, searching whether the welding seam corresponding to the comparison characteristic has defects in a defect characteristic library.

The method steps in the embodiments of the present application are described in detail below with reference to the accompanying drawings.

Fig. 1 provides a weld defect identification system in an embodiment of the present application, such as the system in fig. 1, which includes an atlas acquisition apparatus 110 and a weld defect identification apparatus 120. Specifically, the spectrum acquisition device 110 acquires microwave signals through a plurality of detection probes, and converts the microwave signals into a weld spectrum of the pipe to be identified. After the weld joint atlas of the pipe to be identified is obtained, the weld joint defect identification equipment 120 extracts features in the weld joint atlas, compares the features in the weld joint atlas with the features of a qualified atlas to obtain comparison features, and then confirms whether the weld joint has defects or not through the comparison features. According to the embodiment of the application, the weld defects can be automatically identified through the system, so that a large amount of manpower resources are saved, the working efficiency is improved, and the uncertainty and the identification error caused by human factors are reduced.

It should be noted that the pipe to be identified is a material acquired by the spectrum acquisition device 110 for the microwave corresponding to the weld joint.

It is to be understood that the weld defect identifying apparatus 120 is any electronic apparatus capable of performing weld defect identification. As a specific embodiment of the present application, the weld defect identifying apparatus may be a computer, or the like; as another specific example of the present application, the weld defect identifying apparatus may also be a server or the like. The embodiments of the present application are not limited thereto.

Unlike the embodiment of the present application, in the related art, the type of a typical defect is usually determined manually, but in the process of determining the type of a defect, if the characteristics of the two types of defects are relatively close, errors are easily identified, manual identification is used, the labor cost is increased, and the accuracy and the identification speed of identification cannot be guaranteed.

In order to solve at least the above problem, as shown in fig. 2, an embodiment of the present application provides a method for performing weld defect identification by a weld defect identification apparatus 120, including:

and S210, acquiring a welding line map of the pipe to be identified.

It should be noted that, in the embodiment of the present application, the weld map may be a weld map of a pipe, a weld map of a plate, or a weld map of other different shapes or materials. The embodiments of the present application are not limited thereto.

S220, extracting the characteristics of the welding line atlas, and comparing the characteristics of the welding line atlas with the characteristics of a qualified atlas to obtain comparison characteristics.

In one embodiment of the present application, after S210, the method includes: preprocessing the welding line map of the pipe to be identified to obtain the processed welding line map of the pipe to be identified; extracting the characteristics of the weld map of the processed pipe to be identified, and comparing the weld map of the processed pipe to be identified with the qualified map to obtain comparison characteristics.

Namely, after the welding line atlas of the pipe to be identified is obtained, the welding line atlas of the pipe to be identified is subjected to preprocessing processes of noise reduction, surface influence removal and the like, and the welding line atlas of the pipe to be identified after processing is obtained. And in the process of extracting the characteristics, extracting the characteristics by using the processed weld map of the pipe to be identified.

In an embodiment of the present application, S220 specifically includes: after the welding seam map is obtained, firstly, the area where the welding seam is located in the welding seam map of the pipe to be identified is extracted, and in addition, the number of pixels meeting the gray threshold value in the pixel points corresponding to the welding seam length area and the number of pixels meeting the gray threshold value in the pixel points corresponding to the welding seam width area are extracted. And then, acquiring the number of pixels meeting the gray threshold value in the pixel points corresponding to the weld length region in the extracted qualified atlas and the number of pixels meeting the gray threshold value in the pixel points corresponding to the weld width region in the qualified atlas. And then comparing the length and the width of the welding seam in the welding seam map of the pipe to be identified, the corresponding number of pixels meeting the gray threshold value, the length and the width of the welding seam in the qualified map, and the corresponding number of pixels meeting the gray threshold value to obtain a comparison characteristic.

For example, the number of pixels corresponding to a weld length region in a weld map to be identified is 1000, and the number of pixels meeting a gray threshold is 500; the number of pixel points corresponding to the weld width region in the weld map to be identified is 200, and the number of pixel points meeting the gray threshold is 50. The number of pixels, meeting the gray level threshold, of pixel points corresponding to the welding seam length region in the qualified atlas is 900, and the number of pixels, meeting the gray level threshold, of pixel points corresponding to the welding seam width region in the qualified atlas is 150. Therefore, the comparison characteristics between the welding line spectrum and the qualified spectrum of the pipe to be identified are as follows: the difference of the welding seam length region is 400 pixel points, and the difference of the welding seam width region is 100 pixel points.

It should be understood that the gray value corresponding to each pixel point in the weld map is used to represent whether the weld exists at the position and the welding condition. The smaller the gray value is, the darker the color representing the pixel point is, and the better the corresponding welding condition is; the larger the gray value is, the lighter the color representing the pixel point is, the worse the corresponding welding condition is, and even no welding seam exists. The gray threshold is used for measuring the welding condition of the welding seam position corresponding to the pixel point, the gray value of the pixel point meets the gray threshold, and the welding condition of the corresponding welding seam position is represented to be poor.

It can be understood that the grayscale threshold is a value range, and can be adjusted according to the actual production situation, and may be greater than or equal to 50, or greater than or equal to 80.

It can be understood that the qualified map is a weld map corresponding to the qualified pipe.

And S230, confirming whether the welding seam of the pipe to be identified has defects or not through the comparison characteristics.

In one embodiment of the present application, before S230, the method further includes establishing a defect feature library. The defect feature library is established by the following method:

the method comprises the following steps: and extracting the forward characteristics of the qualified map, wherein the forward characteristics are a plurality of characteristics of the welding line in the qualified map.

The plurality of feature expression maps are features that reflect the weld joint welding conditions. The plurality of characteristics can be the length of the welding seam and the width of the welding seam, and can also be the length of the welding seam, the width of the welding seam and the pixel gray value of the welding seam.

In one embodiment of the present application, the first step specifically includes: taking the number of pixels meeting the gray threshold value in pixel points corresponding to the welding seam length region in the welding seam region of the qualified map as the forward characteristic of the welding seam length; and taking the number of pixels meeting the gray threshold value in the pixel points corresponding to the weld width area in the weld area as the forward characteristic of the weld width.

Specifically, as shown in the diagram of the qualified map shown in fig. 3, the abscissa in fig. 3 represents the position coordinate of the circumferential direction of the pipe, the ordinate represents the width value of the weld joint, the region pointed by the arrow is the weld joint 503, the region defined by the two vertical dashed lines is the length 501 of the weld joint, wherein the length of the weld joint can represent the weld joint of one circle of the circumferential direction of the pipe, and the width 502 of the weld joint is a corresponding value of the ordinate in the diagram.

And calculating the number of pixels corresponding to the weld width meeting the gray threshold value in the qualified map by taking the pixels in the qualified map as a basic unit, obtaining the weld width value of the qualified map according to the number of pixels, calculating the number of pixels corresponding to the weld length meeting the gray threshold value in the qualified map, and obtaining the weld length value of the qualified map according to the number of pixels.

For example, the length of the weld joint of which the qualified map meets the grayscale threshold is 10000 pixels, and the corresponding length value of the weld joint is 50cm, where the length value of the qualified weld joint should be equal to the circumference value of the pipe, which indicates that the pipe has weld joints in the circumferential direction. The width of the welding line of which the qualified atlas meets the gray threshold is 400 pixels, and the corresponding width value of the welding line is 2 cm.

As another embodiment of the application, the gray value of each pixel in the welding line pixel point is obtained, and the gray value of each pixel is averaged to obtain the average gray value of the qualified welding line.

For example, the corresponding gray values of each pixel point in the welding seam pixel point are respectively: 50. 0, 30, 10, 50, 10, 0, 20, the average gray value of the qualified weld is 20, so that the gray value of the weld among the plurality of features of the qualified map is 20.

Step two: and extracting negative features of the defect map, wherein the negative features are a plurality of features corresponding to each type of defect weld in the defect map.

As shown in the defect map diagram of fig. 4, the abscissa in fig. 4 represents the position coordinate of the circumferential direction of the pipe, the ordinate represents the width value of the weld, the region pointed by the arrow is the weld 503, the region defined by the two vertical dashed lines is the weld length 501, and the region defined by the two vertical straight lines is the defect weld 504.

It should be noted that the defect map is a map of defects in the weld accumulated during the production process. In the production process, the defect maps corresponding to different types of defects may have a small number of samples.

In an embodiment of the application, the second step specifically includes extracting multiple sets of negative features corresponding to the multiple defect maps, where each set of negative features includes a negative weld length feature and a negative weld width feature.

That is, the defect maps include a plurality of types of defect maps, and negative characteristics of the length of the weld joint and negative characteristics of the width of the weld joint in each type of defect map are extracted. The specific extraction process is the same as the extraction of forward features.

For example, the weld seam region of the defect map is obtained first, and the gray value and the number of pixels of the weld seam width satisfying the gray threshold are read out. For example: the width of the welding seam reaches 2cm at least under the qualified condition, but the color of the map is lightened at a certain position, which indicates that the corresponding position has defects. Specifically, the degree of lightening is different, the corresponding defects are different, and the most serious condition is that no welding seam exists at a certain position.

Step three: and establishing a defect feature library through the positive features and the negative features.

In one embodiment of the present application, the positive features and the negative features are compared to obtain difference features, and a mapping relationship between the difference features and the weld defects is established to obtain the defect feature library.

That is, comparing multiple sets of negative features with positive features to obtain multiple corresponding difference features, and determining defect types corresponding to the multiple difference features; and binding the plurality of difference characteristics with the defect types to obtain a defect characteristic library.

For example: the acceptable weld width value is 2cm, while the unacceptable weld width of a certain class is 0, and the difference in width between them is 2 cm.

In another embodiment of the present application, another method for creating a defect feature library is as follows:

the method comprises the following steps: and inputting the forward features into the first model to obtain forward standard features.

That is, as shown in fig. 5, the qualified atlas 301 is subjected to preprocessing at S310, and then the preprocessed qualified atlas 301 is subjected to feature extraction at S320, so that forward features are obtained. Inputting forward features into the first model 302 integrates sets of forward features of multiple qualified atlases, for example, the process of integrating includes averaging each type of feature to obtain forward standard features.

It is understood that the forward standard features are obtained from the integration of sets of forward features from multiple qualified atlases. Since the features corresponding to each qualifying atlas are numerically related, the features of each type are averaged.

Step two: inputting the negative features and the positive standard features into a second model, obtaining difference features between the negative features and the positive standard features, and establishing a defect feature library through the difference features.

That is, as shown in fig. 5, the preprocessing is performed S330 on the defect map 303 and the defect map 303 generated by the countermeasure network 305, then the feature extraction is performed S340 on the defect map 303 after the preprocessing to obtain negative features, the negative features and the positive standard features obtained in the first step are input into the second model 304 to obtain difference features 306, and thus the defect feature library 307 is built based on the difference features 306.

Specifically, feature extraction is performed by using a qualified map, and forward standard features are calculated through a UBM (universal background model). Inputting the negative characteristic and the positive standard characteristic into a GMM model, calculating the probability in the corresponding pixel by mixing one or more Gaussian functions based on the algorithm of the Gaussian functions, and selecting the characteristic with high probability as the difference characteristic between the negative characteristic and the positive standard characteristic.

In one embodiment of the present application, S230 includes: and inquiring the comparison characteristics in a defect characteristic library and confirming whether the welding seam has defects, wherein the defect characteristic library is used for representing the mapping relation between the comparison characteristics and the existence of the welding seam, and the defect map is the welding seam map corresponding to the defective pipe.

That is, after obtaining the defect feature library, as shown in fig. 6, the preprocessing is performed S410 on the detection map 401, and the feature extraction is performed S420 on the preprocessed detection map 401 to obtain features, then the comparison with the qualified map is performed S430 to obtain difference features, and then the query of the difference features in the defect feature library is performed S440 to obtain the recognition result 402.

It should be noted that, creating and acquiring the defect feature library is completed before identifying features according to defects. The defect feature library can be established in the application, or can be established and completed in advance before other execution processes, and the defect feature library is directly acquired when the defect feature library is used.

In one embodiment of the present application, S230 includes: confirming that the welding seam has defects through the comparison characteristics; and confirming the position of the defect, wherein the position is determined by the X-axis size and the Y-axis size.

In the prior art, defects are identified by naked eyes, including manual measurement of X-axis size, Y-axis size and gray value; the defects can be determined qualitatively through the X-axis size and the Y-axis size; the defects can be quantified through the X-axis size, the Y-axis size and the gray value. Manually quantifying the quality of defects can take a significant amount of time.

In the application, under the condition that the welding seam defect identification equipment confirms that the welding seam has defects, the position where the defects are located can be confirmed, namely the X-axis size and the Y-axis size corresponding to the defects are obtained, so that the positions of the defects can be found according to the X-axis size and the Y-axis size, and damage is compensated.

In one embodiment of the present application, S230 includes: and confirming the defect grade corresponding to the defect type according to the defect type and the preset grade so that maintenance personnel can repair the pipe according to the defect grade and the defect type.

That is to say, after the defect type is obtained according to the comparison characteristics, the defect type is corresponding to the corresponding defect grade, so that a maintenance worker can determine the severity of the defect, determine the damage repairing time and the like according to the severity, and simultaneously obtain a damage repairing method according to the determined defect type, so that the pipe is repaired by using the damage repairing method.

The defect level corresponding to the defect is preset, for example: presetting the defect grade corresponding to the weld defects as one grade, after identifying the type of the defects, corresponding the type of the defects to the defect grade, and also performing early warning according to the defect grade and allocating different repairing processes for repairing.

It will be appreciated that the name of the defect type differs for different contrast features. For example, if the contrast characteristic of the length of the weld is 100 pixels, the corresponding defect characteristic is the type of cold welding.

The above describes a specific implementation process of the method for identifying weld defects of the present application, and the following describes an apparatus for identifying weld defects.

As shown in fig. 7, an apparatus 500 for weld defect identification includes: an atlas acquisition module 510, an atlas comparison module 520, and a result confirmation module 530.

In one embodiment of the present application, the present application provides an apparatus 500 for atlas defect identification, the apparatus comprising: the map acquisition module 510 is configured to acquire a weld map of the pipe to be identified; an atlas comparison module 520 configured to extract features of the weld atlas and compare the features of the weld atlas with features of a qualified atlas to obtain comparison features, wherein the qualified atlas is an atlas corresponding to a qualified weld; a result confirmation module 530 configured to confirm whether the weld of the pipe to be identified has a defect through the comparison features.

In one embodiment of the present application, the result confirmation module is further configured to: and inquiring the comparison characteristics in a defect characteristic library, and confirming whether the welding seam has defects, wherein the defect characteristic library is used for representing the mapping relation between the characteristics and the welding seam defects.

In one embodiment of the present application, the result confirmation module is further configured to: extracting positive features of the qualified map, wherein the positive features are features of a weld joint in the qualified map; extracting negative features of a defect map, wherein the negative features are features corresponding to various types of defect welding lines in the defect map, and the defect map is a map corresponding to the defect welding lines; and establishing the defect feature library through the positive features and the negative features.

In one embodiment of the present application, the result confirmation module is further configured to: comparing the positive features with the negative features to obtain difference features; and establishing a mapping relation between the difference characteristics and the weld defects to obtain the defect characteristic library.

In one embodiment of the present application, the plurality of characteristics includes a weld length and a weld width; the result validation module is further configured to: taking the number of pixels meeting a gray threshold value in pixel points corresponding to the welding seam length region in the welding seam region of the qualified map as the forward characteristic of the welding seam length; and taking the number of pixels meeting the gray threshold value in the pixel points corresponding to the weld width area in the weld area as the forward characteristic of the weld width.

In one embodiment of the present application, the result confirmation module is further configured to: and extracting multiple groups of negative features corresponding to the multiple defect maps, wherein each group of negative features comprises a weld length negative feature and a weld width negative feature.

In one embodiment of the present application, the result confirmation module is further configured to: comparing the multiple sets of negative features with the positive features to obtain multiple corresponding difference features; the obtaining the defect feature library according to the difference features comprises: identifying a defect type corresponding to the plurality of difference features; and binding the plurality of difference characteristics with the defect type to obtain the defect characteristic library.

In one embodiment of the present application, the result confirmation module is further configured to: inputting the forward characteristic into a first model to obtain a forward standard characteristic; inputting the negative features and the positive standard features into a second model to obtain difference features between the negative features and the positive standard features; and establishing the defect feature library through the difference features.

In one embodiment of the present application, a portion of the defect maps are generated by a countermeasure network.

In one embodiment of the present application, the atlas acquisition module is further configured to: preprocessing the welding line map of the pipe to be identified to obtain the processed welding line map of the pipe to be identified; the extracting features of the weld map of the pipe to be identified, and the map comparing module is further configured to: extracting the characteristics of the weld map of the processed pipe to be identified, and comparing the weld map of the processed pipe to be identified with the qualified map to obtain comparison characteristics.

In one embodiment of the present application, the result confirmation module is further configured to: confirming that the welding seam has defects through the comparison characteristics; and confirming the position of the defect, wherein the position is determined by the X-axis size and the Y-axis size.

In one embodiment of the present application, the result confirmation module is further configured to: and confirming the defect grade corresponding to the defect type according to the defect type and the preset grade so that maintenance personnel can repair the damage to the pipe according to the defect grade and the defect type.

In the embodiment of the present application, the module shown in fig. 7 can implement each process in the method embodiments of fig. 1 to 6. The operations and/or functions of the respective modules in fig. 7 are respectively for implementing the corresponding flows in the method embodiments in fig. 1 to 6. Reference may be made specifically to the description of the above method embodiments, and a detailed description is appropriately omitted herein to avoid redundancy.

As shown in fig. 8, an embodiment of the present application provides an electronic device 600, including: a processor 610, a memory 620 and a bus 630, wherein the processor is connected to the memory through the bus, the memory stores computer readable instructions, when the computer readable instructions are executed by the processor, for implementing the method according to any one of the above embodiments, specifically, the description of the above embodiments of the method can be referred to, and the detailed description is omitted here to avoid repetition.

Wherein the bus is used for realizing direct connection communication of the components. The processor in the embodiment of the present application may be an integrated circuit chip having signal processing capability. The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The Memory may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Read Only Memory (EPROM), an electrically Erasable Read Only Memory (EEPROM), and the like. The memory stores computer readable instructions that, when executed by the processor, perform the methods described in the embodiments above.

It will be appreciated that the configuration shown in fig. 8 is merely illustrative and may include more or fewer components than shown in fig. 8 or have a different configuration than shown in fig. 8. The components shown in fig. 8 may be implemented in hardware, software, or a combination thereof.

Embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a server, the method in any of the above-mentioned all embodiments is implemented, which may specifically refer to the description in the above-mentioned method embodiments, and in order to avoid repetition, detailed description is appropriately omitted here.

The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method of weld defect identification, the method comprising:

acquiring a weld map of a pipe to be identified;

extracting the characteristics of the welding line atlas, and comparing the characteristics of the welding line atlas with the characteristics of a qualified atlas to obtain comparison characteristics, wherein the qualified atlas is an atlas corresponding to a qualified welding line;

and confirming whether the welding seam of the pipe to be identified has defects or not through the comparison characteristics.

2. The method according to claim 1, wherein the confirming whether the weld of the pipe to be identified has defects through the comparison features comprises the following steps:

and inquiring the comparison characteristics in a defect characteristic library, and confirming whether the welding seam has defects, wherein the defect characteristic library is used for representing the mapping relation between the characteristics and the welding seam defects.

3. The method of claim 2, wherein prior to said querying said comparison feature in a library of defect features to determine if said weld is defective, said method further comprises:

extracting positive features of the qualified map, wherein the positive features are features of a weld joint in the qualified map;

extracting negative features of a defect map, wherein the negative features are features corresponding to various types of defect welding lines in the defect map, and the defect map is a map corresponding to the defect welding lines;

and establishing the defect feature library through the positive features and the negative features.

4. The method of claim 3, wherein the creating the defect feature library from the positive-going features and the negative-going features comprises:

comparing the positive features with the negative features to obtain difference features;

and establishing a mapping relation between the difference characteristics and the weld defects to obtain the defect characteristic library.

5. The method of claim 4, wherein the characteristics include a weld length and a weld width;

the extracting of the forward characteristics of the qualified atlas comprises the following steps:

taking the number of pixels meeting a gray threshold value in pixel points corresponding to the welding seam length region in the welding seam region of the qualified map as the forward characteristic of the welding seam length; and

and taking the number of pixels meeting the gray threshold value in the pixel points corresponding to the weld width area in the weld area as the forward characteristic of the weld width.

6. The method of claim 5, wherein the extracting negative features of the defect map comprises:

and extracting multiple groups of negative features corresponding to the multiple defect maps, wherein each group of negative features comprises a weld length negative feature and a weld width negative feature.

7. The method of claim 6, wherein comparing the positive features to the negative features to obtain difference features comprises:

comparing the multiple sets of negative features with the positive features to obtain multiple corresponding difference features;

the obtaining the defect feature library according to the difference features comprises:

identifying a defect type corresponding to the plurality of difference features;

and binding the plurality of difference characteristics with the defect type to obtain the defect characteristic library.

8. The method of claim 3, wherein the creating the defect feature library from the positive-going features and the negative-going features comprises:

inputting the forward characteristic into a first model to obtain a forward standard characteristic;

inputting the negative features and the positive standard features into a second model to obtain difference features between the negative features and the positive standard features;

and establishing the defect feature library through the difference features.

9. The method according to any one of claims 1 to 8, wherein a portion of the defect maps are generated by a challenge network.

10. The method according to any one of claims 1-8, wherein after the obtaining the weld map of the pipe to be identified, the method further comprises:

preprocessing the welding line map of the pipe to be identified to obtain the processed welding line map of the pipe to be identified;

the method comprises the following steps of extracting the characteristics of the welding line atlas of the pipe to be identified, comparing the characteristics of the welding line atlas of the pipe to be identified with the characteristics of a qualified atlas to obtain comparison characteristics, and comprises the following steps:

extracting the characteristics of the weld map of the processed pipe to be identified, and comparing the weld map of the processed pipe to be identified with the qualified map to obtain comparison characteristics.

11. The method according to any one of claims 1-8, wherein said confirming whether the weld is defective by the comparison feature comprises:

confirming that the welding seam has defects through the comparison characteristics;

and confirming the position of the defect, wherein the position is determined by the X-axis size and the Y-axis size.

12. The method of claim 11, wherein after confirming the weld is defective by the comparison feature, the method further comprises:

and confirming the defect grade corresponding to the defect type according to the defect type and the preset grade so that maintenance personnel can repair the damage to the pipe according to the defect grade and the defect type.

13. An apparatus for atlas defect identification, the apparatus comprising:

the map acquisition module is configured to acquire a weld map of the pipe to be identified;

the map comparison module is configured to extract the features of the welding seam map and compare the features of the welding seam map with the features of a qualified map to obtain comparison features, wherein the qualified map is a map corresponding to a qualified welding seam;

and the result confirmation module is configured to confirm whether the welding seam of the pipe to be identified has defects or not through the comparison characteristics.

14. An electronic device, comprising: a processor, a memory, and a bus;

the processor is connected to the memory via the bus, the memory storing computer readable instructions for implementing the method of any one of claims 1-12 when the computer readable instructions are executed by the processor.

15. A computer-readable storage medium, having stored thereon a computer program which, when executed, implements the method of any one of claims 1-12.