CN111257422A - Wheel axle defect identification model construction method and defect identification method based on machine vision - Google Patents
Wheel axle defect identification model construction method and defect identification method based on machine vision Download PDFInfo
- Publication number
- CN111257422A CN111257422A CN202010128206.8A CN202010128206A CN111257422A CN 111257422 A CN111257422 A CN 111257422A CN 202010128206 A CN202010128206 A CN 202010128206A CN 111257422 A CN111257422 A CN 111257422A
- Authority
- CN
- China
- Prior art keywords
- defect
- defect identification
- image
- wheel axle
- machine vision
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
- G01N29/048—Marking the faulty objects
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention relates to a wheel axle defect identification model construction method and a defect identification method based on machine vision, wherein the method constructs a defect identification model based on a machine vision technology, and can continuously improve the accuracy of defect identification through continuous autonomous learning and training; meanwhile, the machine vision is not limited by the experience and quality of personnel, so that the defect identification accuracy, the operation efficiency and the operation quality are greatly improved, and the misjudgment and the missed judgment of the defects are effectively reduced. The identification method comprises the following steps: utilizing a defect identification model to identify defects; screening the identified defects; displaying the screened defects and alarming; and continuously perfecting the defect identification model by using new defect data.
Description
Technical Field
The invention is suitable for the field of ultrasonic nondestructive inspection, and particularly relates to a wheel axle defect identification model construction method and a defect identification method based on machine vision.
Background
At present, the flaw detection of the wheel shaft adopts ultrasonic detection and takes manual analysis of detection data as a main means. The axle ultrasonic detection results are displayed in the forms of A display, B display and C display, and then the displayed images are observed by naked eyes to judge whether the axle has defects. The defect judgment result depends on the quality and experience of detection personnel to a great extent, so that the missed judgment and the erroneous judgment are easily caused, and the efficiency is low; the method is limited by the quality and experience of detection personnel, and different detection personnel may not have consistent analysis results on the same detection data, so that accurate basis cannot be provided for decision; in addition, as the number of rail transit rolling stocks increases year by year and the types of vehicle axles are various, the contradiction between the increase of axle detection tasks and the shortage of detection personnel is increasingly sharp; the efficiency of nondestructive test operation has been restricted to the not enough efficiency of testing personnel quantity, is unfavorable for carrying out short-term test to a large amount of shaft, and then influences production efficiency. Meanwhile, the existing automatic wheel axle flaw detection analysis method cannot meet the requirements of a wheel axle flaw detection operation process and the accuracy and consistency of flaw identification. There is a need in the art for a method for automated analysis of ultrasonic test data that can be performed quickly and accurately to solve the above problems.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a vehicle defect identification model construction method and a defect identification method based on machine vision, wherein the method constructs a defect identification model based on a machine vision technology, and can continuously improve the accuracy of defect identification through continuous autonomous learning and training; meanwhile, the machine vision is not limited by the experience and quality of personnel, so that the defect identification accuracy, the operation efficiency and the operation quality are greatly improved, and the misjudgment and the missed judgment of the defects are effectively reduced.
The invention provides a method for constructing a wheel axle defect identification model based on machine vision, which comprises the following steps:
s100, acquiring detection data of the wheel axle by using ultrasonic flaw detection equipment, and storing the detection data in a binary file;
s200, acquiring sample data through the binary file, and constructing a defect positive sample library and a defect negative sample library;
step S300, preprocessing a defect positive sample library to generate a defect positive sample template file;
and S400, constructing a defect identification model through a defect positive sample template file, a sample matching algorithm and a defect negative sample library.
Further, the ultrasonic flaw detection equipment acquires detection data according to a shaft type or a wheel type, a corresponding channel configuration and a scanning mode.
Further, the step S200 further includes the steps of:
step S210, performing cluster analysis on the binary files;
step S220, extracting an image from the binary file after the clustering analysis;
step S230, capturing a defect position image according to the image characteristics of the defect to acquire sample data;
step S240, the defective sample data forms a defective positive sample library, and the non-defective sample data forms a defective negative sample library.
Further, the defect positive sample library comprises a defect image; the defect negative sample library comprises a non-defect image, a noise image and a transition arc image.
Further, the defect images include flat-bottom hole images, transverse defect images, longitudinal defect images, outer surface defect images, inner surface defect images.
Further, the preprocessing includes a normalization process, i.e., unifying the size and format of the defect image.
The invention also provides a wheel axle defect identification method based on machine vision, which comprises the following steps:
step Z100, acquiring detection data of the wheel axle by using ultrasonic flaw detection equipment, storing the detection data in a binary file, and extracting a detection image from the binary file;
step Z200, utilizing the wheel axle defect identification model to identify the defects of the detected image;
step Z300, defect screening is carried out on the defect identification result according to the specific parameters;
and step Z400, displaying the screened defects and alarming.
Further, the step Z300 further includes: the specific parameter at least comprises one of defect type, threshold value and position.
Further, the axle defect identification method further comprises the following steps: and step Z500, utilizing the images contained in the defect identification result to supplement and construct the wheel axle defect identification model.
Further, step Z500 further comprises: the image contained in the identification result can be manually added to a defect positive sample library or a defect negative sample library; the computer can also automatically add the images contained in the defect identification result to a defect positive sample library or a defect negative sample library according to the comparison result of the defect image matching degree and the matching degree threshold.
The invention has the following beneficial effects: firstly, the invention introduces the artificial intelligence technology into the nondestructive testing field, promotes the development of the nondestructive testing technology to the intelligent direction, and greatly improves the quality and efficiency of the nondestructive testing. Based on a machine vision technology, utilizing ultrasonic flaw detection equipment to obtain detection data of a wheel axle, obtaining sample data according to image characteristics of wheel axle defects, and further constructing a defect identification model; by adopting a machine learning method, the characteristic information of the defects is obtained, and the requirements on the quality and experience of detection personnel are reduced; by adopting the computer analysis method, the problems of misjudgment, missed judgment and the like caused by artificial subjective factors during manual data analysis are solved, the detection quality is improved, and the automatic analysis of the ultrasonic detection data is quickly and accurately realized.
Secondly, defect recognition is carried out by using the wheel axle defect recognition model, and a large amount of data analysis work originally born by manpower is processed by a computer, so that the data processing speed is greatly improved, the rapid nondestructive detection of a large amount of workpieces is realized, the defect recognition efficiency is greatly improved, and the production efficiency of the wheel axle is improved; meanwhile, the sample data of the defects are continuously supplemented and perfected along with the increase of detection samples and the richness of defect types, and the defect identification model of the wheel axle is continuously supplemented and optimized through the defect identification result, so that the accuracy of data analysis is further improved.
Drawings
FIG. 1 is a flow chart of a method for constructing a defect identification model based on machine vision according to the present invention.
FIG. 2 is a flow chart of the defect identification method based on machine vision of the present invention.
Fig. 3 is a schematic diagram of the rim spoke defect identification result.
Fig. 4 is a schematic view of the hollow axle defect recognition result before screening.
Fig. 5 is a schematic diagram of the result of hollow axle defect identification after screening.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
Referring to fig. 1: a method for constructing a wheel axle defect identification model based on machine vision comprises the following steps:
and S100, acquiring detection data of the wheel shaft by using ultrasonic flaw detection equipment, and storing the detection data in a binary file.
Due to the fact that the wheel rim, the wheel disc, the wheel rim and the hollow shaft are different in geometric characteristics, flaw detection modes are different, and detection images are different. The ultrasonic wheel axle flaw detection equipment acquires detection data of wheel axle components according to different axle types or wheel types, corresponding channels and scanning modes, and stores the detection data in an upper computer of the flaw detection equipment in a binary file form, wherein the binary file comprises information such as positions, wave amplitudes, detection parameters and images.
The channels are probes, one channel is a probe, each probe is scanned according to a certain direction and angle, a plurality of probes are scanned together in one scanning process, and each probe has an independent scanning result.
Scanning modes comprise A scanning, B scanning and C scanning, and output results are also divided into A scanning display, B scanning display and C scanning display; a scanning is displayed in an X-axis time representing mode, and a Y-axis amplitude representing mode, namely the X-axis represents time and the Y-axis represents amplitude; b, scanning and displaying a cross-sectional view of the detected piece, which is drawn by the relationship between the sound path length of the echo signal with the amplitude within the preset range and the position of the sound beam axis when the probe is scanned along only one direction, wherein for a hollow shaft, the X axis is represented as the axial direction, and the sound path length of the Y axis is the radius; c scanning is displayed as a two-dimensional plane display of the detected piece, the existence of echo signals with the amplitude or the sound path within a preset range is drawn according to the scanning position of the probe, and for a hollow shaft, the X axis is represented as an axial direction, and the Y axis is represented as a circumferential angle.
The A scanning display, the B scanning display and the C scanning display are stored in an upper computer of the flaw detection equipment in a binary file mode, and if the A scanning display is stored as the mark.a ', the B scanning display is stored as the mark.b ', and the C scanning display is stored as the mark.c '.
And S200, acquiring sample data through the binary file, and constructing a defect positive sample library and a defect negative sample library.
And step S210, clustering and analyzing the binary files.
The clustering analysis adopts a clustering algorithm to obtain a clustering result corresponding to a preset dimension, wherein the preset dimension refers to a clustering mode according to one or more of an axis type or a wheel type, a channel and a scanning mode in the embodiment, and the clustering algorithm comprises at least one of a k-means clustering algorithm, a mean shift clustering algorithm, a pixel clustering algorithm, a hierarchical clustering algorithm, a spectral clustering algorithm and a deep embedded clustering algorithm based on a deep convolutional neural network. The binary file may be shared and transmitted over a computer network. And step S220, extracting an image from the binary file after the clustering analysis.
And step S230, intercepting the image of the defect position according to the image characteristics of the defect to acquire sample data.
Each defect has its own unique image characteristics, such as a C-scan display, where the defect is characterized by a progressive decrease in amplitude from center point to edge, and the corresponding image characteristics are a progressive decrease in table of center point to edge colors. B, scanning and displaying, wherein for the axle flaw detection, as the probe spirally advances along a certain direction, the same flaw can be continuously found by the same probe for multiple times, and the corresponding image characteristic is a point set which is continuously discontinuous along a certain direction; for wheel flaw detection, the probe fixes the wheel to rotate, the defects can also be continuously found for multiple times by the same probe, but the corresponding image features are point sets which are continuously interrupted along a certain direction, each point set presents a parabola (dovetail shape), and the color table gradually decreases from the top point to the two sides.
Sample data is acquired from the defect image using a feature extraction algorithm, which may include, but is not limited to, at least one of: RPN (Region pro-possible Network, candidate area generation Network) algorithm, FPN (feature pyramid Networks) algorithm, and DN (Deep Net) algorithm, and the like.
Step S240, the defective sample data forms a defective positive sample library, and the non-defective sample data forms a defective negative sample library.
Determining a defect position according to the image characteristics of the defect, and intercepting an image of the defect position to form a defect positive sample library and a defect negative sample library, wherein the defect positive sample library comprises defect images such as a flat-bottom hole image, a transverse defect image, a longitudinal defect image, an outer surface defect image, an inner surface defect image and the like; the defect negative sample library comprises non-defect images, noise images, transition arc images and the like.
And step S300, preprocessing the defect positive sample library to generate a defect positive sample template file.
Preprocessing a defect positive sample, wherein the preprocessing method comprises at least one of the following steps: adjusting at least one of brightness, gray scale and contrast; adjusting at least one of a size and an offset; and performing adaptive graph cutting and the like, and performing normalization processing on the size and the format of the defect image, for example, the processed size of the defect image is 20 pixels by 20 pixels, the format is stored as a BMP, and the defect positive sample template file is generated from the defect positive sample library after the normalization processing.
And S400, constructing a defect identification model through a defect positive sample template file, a sample matching algorithm and a defect negative sample library.
Inputting parameters such as a defect positive sample template file, a sample matching algorithm, a defect negative sample library, a training phase and the like to perform machine learning, and constructing a defect identification model after learning is finished, wherein the defect identification model is stored in an XML file form. The sample matching algorithm may be a gray value-based matching algorithm, a shape-based matching algorithm, a feature point-based matching algorithm, or the like.
Example 2
Referring to fig. 2: a wheel axle defect identification method based on machine vision comprises the following steps:
and step Z100, acquiring detection data of the wheel axle by using ultrasonic flaw detection equipment, storing the detection data in a binary file, and extracting a detection image from the binary file.
This step is the same as in embodiment 1 and will not be described herein.
And step Z200, utilizing a defect identification model to identify the defects of the detected image.
After the processing program loads the defect identification model, parameters such as a detection image, a defect lower limit (a target image pixel size lower limit), a defect upper limit (a target image pixel size upper limit), stepping (image division is scaled according to a certain size), neighborhood (hit frequency) and the like are input, the computer automatically matches and identifies the defect and outputs a defect identification result, the identification result comprises an image, a position, a size, a matching degree (the matching degree of the defect image and the defect identification model) and a type, and a schematic diagram of a display result is shown in fig. 3.
Further processing in the program can obtain the amplitude and sound path, such as obtaining the sound path and the amplitude from the detection data according to the position information.
If the input detection image CRH1A hollow car moving axis channel is 45+ forward C display image, the defect lower limit is 5, the defect upper limit is 40, the step is 1.1, the neighborhood is 3 and other parameters; the computer can output the result that the defect size in the image is 5 × 5 to 40 × 40 pixels and the hit frequency is more than or equal to 3 times according to the defect identification model, the output result has a coincidence block diagram or a non-defect block diagram, the output result is stored in structured data after frame regression and cluster analysis, and the detection result is as follows:
[{id:"1",t＝"1",channel:"45+",x:"244",y:"267",h:"28",w:"28",l:"9",v:"45",k:"1.3",d＝"55"},{id:"2",t＝"1",channel:"45+",x:"382",y:"88",h:"27",w:"27",l:"9",v:"65",k:"1.3",d＝"60"},{id:"3",t＝"1",channel:"45+",x:"500",y:"177",h:"29",w:"29",l:"9",v:"70",k:"1.3",d＝"55"},{id:"4",t＝"1",channel:"45+",x:"630",y:"267",h:"28",w:"28",l:"9",v:"80",k:"1.3",d＝"55"},{id:"5",t＝"1",channel:"45+",x:"1057",y:"87",h:"28",w:"28",l:"9",v:"66",k:"1.3",d＝"60"},{id:"6",t＝"1",channel:"45+",x:"1330",y:"88",h:"26",w:"26",l:"9",v:"90",k:"1.3",d＝"55"},{id:"7",t＝"2",channel:"45+",x:"1589",y:"267",h:"29",w:"29",l:"9",v:"77",k:"1.3",d＝"55"},{id:"8",t＝"1",channel:"45+",x:"1876",y:"179",h:"27",w:"27",l:"9",v:"80",k:"1.3",d＝"70"},{id:"9",t＝"1",channel:"45+",x:"1986",y:"87",h:"29",w:"29",l:"9",v:"77",k:"1.3",d＝"55"},{id:"10",t＝"1",channel:"45+",x:"2086",y:"2",h:"27",w:"27",l:"9",v:"89",k:"1.3",d＝"60"},{id:"11",t＝"1",channel:"45+",x:"2125",y:"266",h:"29",w:"29",l:"9",v:"90",k:"1.3",d＝"55"}]。
the above results indicate that 11 defects are found, where id is the serial number, t is the defect type, x is the defect x-axis coordinate, y is the y-axis coordinate, channel is the channel name, h is the defect height, w is the defect width, l is the stage value, v is the amplitude, k is the matching degree, and d is the sound path.
The output results are tabulated as follows:
| id | t | Channel | x | y | h | w | l | | k | d | |
| 1 | 1 | 45+ | 244 | 267 | 28 | 28 | 9 | 45 | 1.3 | 55 | |
| 2 | 1 | 45+ | 382 | 88 | 27 | 27 | 9 | 65 | 1.3 | 60 | |
| 3 | 1 | 45+ | 500 | 177 | 29 | 29 | 9 | 70 | 1.3 | 55 | |
| 4 | 1 | 45+ | 630 | 267 | 28 | 28 | 9 | 80 | 1.3 | 55 | |
| 5 | 1 | 45+ | 1057 | 87 | 28 | 28 | 9 | 66 | 1.3 | 60 | |
| 6 | 1 | 45+ | 1330 | 88 | 26 | 26 | 9 | 90 | 1.3 | 55 | |
| 7 | 2 | 45+ | 1589 | 267 | 29 | 29 | 9 | 77 | 1.3 | 55 | |
| 8 | 1 | 45+ | 1876 | 179 | 27 | 27 | 9 | 80 | 1.3 | 70 | |
| 9 | 1 | 45+ | 1986 | 87 | 29 | 29 | 9 | 77 | 1.3 | 55 | |
| 10 | 1 | 45+ | 2086 | 2 | 27 | 27 | 9 | 89 | 1.3 | 60 | |
| 11 | 1 | 45+ | 2125 | 266 | 29 | 29 | 9 | 90 | 1.3 | 55 |
and step Z300, screening the defects of the defect identification result according to the specific parameters.
After the computer loads the defect identification result, defect screening is carried out according to specific parameters, and the screened parameters comprise defect types, threshold values (wave amplitudes) and positions.
And step Z310, screening the defects according to the defect types.
Assuming that 1 represents a transverse defect, 2 represents a longitudinal defect, 3 represents a flat-bottom hole, 4 represents a surface defect, and so on, if only a transverse defect is selected, the result with t of 1 is matched and displayed in the result file, and other types are not displayed;
and step Z320, screening the defects through an amplitude threshold value.
If the amplitude threshold is adjusted to 50, the defect with the amplitude of the defect plus the floating value (the default value of the floating value is 0) larger than 50 is displayed, and the defect with the amplitude smaller than or equal to 50 is not displayed.
And step Z330, screening the defects through the positions.
Screening according to position information such as radius, axial direction, circumferential direction and the like, if the defect position is characterized by short sound path amplitude, comparing the sound path of a certain position with the contour line (an axial contour and a wheel contour) of the position according to the characteristic, if the floating value (the default value of the floating value is 0) of the sound path is smaller than the contour line, displaying, otherwise, not displaying; if the hollow axle B or C is scanned and displayed, the axial display range is set to be 100-500 mm, the defect in the closed section is displayed, otherwise, the defect is not displayed; if the hollow axle C is scanned and displayed, the circumferential display range is set to be 0-180 degrees, the defect in the closed region is displayed, otherwise, the defect is not displayed;
several parameters can be used alone or in combination, such as adjusting the amplitude threshold to 50 and type 1 transverse defect, then the defect will be shown to be transverse, the floating value of the amplitude is greater than 50, and the defect that is not in condition will not be shown.
As shown in fig. 4, the schematic diagram of the defect recognition result of the hollow axle before screening is shown, and as shown in fig. 5, the schematic diagram of the defect recognition result of the hollow axle after screening is shown, and the defect recognition result after defect screening is more accurate.
And step Z400, displaying the screened defects and alarming.
And loading the screened defect data into an image to be displayed in a defect display frame mode, wherein the information of the defect display frame comprises a serial number, amplitude, type and the like.
The defect warning method is characterized by comprising the following steps of warning defects, and warning and displaying information related to the defects in the forms of images, sounds, photoelectricity, lists and the like in multiple warning modes.
And step Z500, utilizing the images contained in the defect identification result to supplement and construct a defect identification model.
Preferably, the label of the defect display box can display information such as serial number, amplitude, type and the like, and corresponding response events can be generated when the defect display box is clicked, double-clicked and dragged by a mouse.
The mouse response events include but are not limited to events such as moving, clicking, double clicking and the like, for example, the mouse holds a left button to drag a certain defect display box, the position of the defect box can be moved and the defect position information can be corrected, a right button clicks a certain defect box, a popup menu is displayed, and the popup menu comprises 'deleting', 'adding to a positive sample', 'adding to a negative sample';
according to a certain defect display frame mouse response event, adding a defect image of a defect display frame to a defect positive sample library or a defect negative sample library can be manually selected.
More preferably, the computer automatically adds the defect image to a defect positive sample library or a defect negative sample library according to a comparison result of the defect image matching degree and a matching degree threshold, wherein the matching degree threshold comprises a positive sample matching degree threshold and a negative sample matching degree threshold, and the matching degree threshold can be obtained according to the defect identification model. The matching degree can be obtained through a matching algorithm, and the matching algorithm can be one of a matching algorithm based on gray values, a matching algorithm based on shapes and a matching algorithm based on feature points.
And adding the defect image matching degree which is more than or equal to the positive sample matching degree threshold value into a defect positive sample library, adding the defect image matching degree which is less than or equal to the negative sample matching degree threshold value into a defect negative sample library, and not processing the intermediate value. If the threshold value of the matching degree of the positive sample is 1 and the threshold value of the matching degree of the negative sample is-1, the defect image with the matching degree of more than or equal to 1 is added into a defect positive sample library, the defect image with the matching degree of less than or equal to-1 is added into a defect negative sample library, and the defect positive sample is preprocessed, wherein the preprocessing method comprises at least one of the following steps: adjusting at least one of brightness, gray scale and contrast; adjusting at least one of a size and an offset; and (3) performing self-adaptive graph cutting and the like, namely performing normalization processing on the size and the format of the defect image, wherein the processed size of the defect image is 20 pixels by 20 pixels, the format is stored as BMP, and the defect positive sample template file is generated from the defect positive sample library after the normalization processing.
The invention adopts the machine vision technology in the field of artificial intelligence to obtain and construct the image characteristic information of the defect, and the computer analyzes the detection data/image of the ultrasonic detection by utilizing the image characteristic information to identify the defect in the detected workpiece and can prompt the defect information (including but not limited to position, size, amplitude, type and the like) of the detection personnel in the forms of image, table, sound, light (including but not limited to a warning light) and the like.
Meanwhile, the image characteristic information of the defects is continuously supplemented and perfected along with the increase of detection samples and the abundance of defect types, and the accuracy of data analysis is further improved.
Compared with a non-artificial intelligence method, the method has the autonomous learning capability, and can be suitable for more types of defects by increasing the types of positive samples; and non-defect interference samples are put into the defect negative sample library, so that non-defect interference information can be accurately distinguished, and the accuracy of defect identification is further improved.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (10)
1. A wheel axle defect identification model construction method based on machine vision is characterized in that: the method comprises the following steps:
s100, acquiring detection data of the wheel axle by using ultrasonic flaw detection equipment, and storing the detection data in a binary file;
s200, acquiring sample data through the binary file, and constructing a defect positive sample library and a defect negative sample library;
step S300, preprocessing a defect positive sample library to generate a defect positive sample template file;
and S400, constructing a defect identification model through a defect positive sample template file, a sample matching algorithm and a defect negative sample library.
2. The method for constructing the wheel axle defect identification model based on the machine vision is characterized in that: the ultrasonic flaw detection equipment is configured with corresponding channels and scanning modes according to the shaft type or the wheel type to obtain detection data.
3. The method for constructing the wheel axle defect identification model based on the machine vision is characterized in that: the step S200 further includes the steps of:
step S210, performing cluster analysis on the binary files;
step S220, extracting an image from the binary file after the clustering analysis;
step S230, capturing a defect position image according to the image characteristics of the defect to acquire sample data;
step S240, the defective sample data forms a defective positive sample library, and the non-defective sample data forms a defective negative sample library.
4. The method for constructing the wheel axle defect identification model based on the machine vision is characterized in that: the defect positive sample library comprises defect images; the defect negative sample library comprises a non-defect image, a noise image and a transition arc image.
5. The method for constructing the wheel axle defect identification model based on the machine vision is characterized in that: the defect images include flat-bottom hole images, transverse defect images, longitudinal defect images, outer surface defect images, inner surface defect images.
6. The method for constructing the wheel axle defect identification model based on the machine vision is characterized in that: the preprocessing includes a normalization process, i.e., unifying the size and format of the defect image.
7. A wheel axle defect identification method based on machine vision is characterized in that: the method comprises the following steps:
step Z100, acquiring detection data of the wheel axle by using ultrasonic flaw detection equipment, storing the detection data in a binary file, and extracting a detection image from the binary file;
step Z200, carrying out defect identification on the detection image by using the axle defect identification model of any one of claims 1-6;
step Z300, defect screening is carried out on the defect identification result according to the specific parameters;
and step Z400, displaying the screened defects and alarming.
8. The wheel axle defect identification method based on the machine vision as claimed in claim 7, characterized in that: the step Z300 further includes: the specific parameter at least comprises one of defect type, threshold value and position.
9. The wheel axle defect identification method based on the machine vision as claimed in claim 7, characterized in that: the axle defect identification method further comprises the following steps: step Z500, building a wheel axle defect identification model according to any one of claims 1 to 6 by using the image supplement contained in the defect identification result.
10. The wheel axle defect identification method based on the machine vision as claimed in claim 9, characterized in that: step Z500 further comprises: the image contained in the defect identification result can be manually added to a defect positive sample library or a defect negative sample library; the computer can also automatically add the images contained in the defect identification result to a defect positive sample library or a defect negative sample library according to the comparison result of the defect image matching degree and the matching degree threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128206.8A CN111257422B (en) | 2020-02-28 | 2020-02-28 | Wheel axle defect identification model construction method and defect identification method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010128206.8A CN111257422B (en) | 2020-02-28 | 2020-02-28 | Wheel axle defect identification model construction method and defect identification method based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111257422A true CN111257422A (en) | 2020-06-09 |
| CN111257422B CN111257422B (en) | 2023-09-08 |
Family
ID=70946130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010128206.8A Active CN111257422B (en) | 2020-02-28 | 2020-02-28 | Wheel axle defect identification model construction method and defect identification method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111257422B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113470018A (en) * | 2021-09-01 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Hub defect identification method, electronic device, device and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012008142A (en) * | 2011-08-30 | 2012-01-12 | Mitsubishi Heavy Ind Ltd | Program, processing unit and processing method for processing ultrasonic flaw detection data |
| CN107451997A (en) * | 2017-07-31 | 2017-12-08 | 南昌航空大学 | A kind of automatic identifying method of the welding line ultrasonic TOFD D scanning defect types based on deep learning |
| CN107749058A (en) * | 2017-10-23 | 2018-03-02 | 中南大学 | A kind of machine vision detection method and system of boiler tubing surface defect |
| CN109658397A (en) * | 2018-12-12 | 2019-04-19 | 广州地铁集团有限公司 | A kind of rail polling method and system |
| US20190161919A1 (en) * | 2017-11-30 | 2019-05-30 | Sperry Rail Holdings, Inc. | System and method for inspecting a rail using machine learning |
| CN109829907A (en) * | 2019-01-31 | 2019-05-31 | 浙江工业大学 | A kind of metal shaft surface defect recognition method based on deep learning |
| CN110297041A (en) * | 2019-07-01 | 2019-10-01 | 天津工业大学 | A kind of 3D woven composite defect inspection method based on FCN and GRU |
-
2020
- 2020-02-28 CN CN202010128206.8A patent/CN111257422B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012008142A (en) * | 2011-08-30 | 2012-01-12 | Mitsubishi Heavy Ind Ltd | Program, processing unit and processing method for processing ultrasonic flaw detection data |
| CN107451997A (en) * | 2017-07-31 | 2017-12-08 | 南昌航空大学 | A kind of automatic identifying method of the welding line ultrasonic TOFD D scanning defect types based on deep learning |
| CN107749058A (en) * | 2017-10-23 | 2018-03-02 | 中南大学 | A kind of machine vision detection method and system of boiler tubing surface defect |
| US20190161919A1 (en) * | 2017-11-30 | 2019-05-30 | Sperry Rail Holdings, Inc. | System and method for inspecting a rail using machine learning |
| CN109658397A (en) * | 2018-12-12 | 2019-04-19 | 广州地铁集团有限公司 | A kind of rail polling method and system |
| CN109829907A (en) * | 2019-01-31 | 2019-05-31 | 浙江工业大学 | A kind of metal shaft surface defect recognition method based on deep learning |
| CN110297041A (en) * | 2019-07-01 | 2019-10-01 | 天津工业大学 | A kind of 3D woven composite defect inspection method based on FCN and GRU |
Non-Patent Citations (1)
| Title |
|---|
| 赵欣洋等: "基于深度学习的不规则特征识别检测技术", 《轻工机械》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113470018A (en) * | 2021-09-01 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Hub defect identification method, electronic device, device and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111257422B (en) | 2023-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113362326B (en) | Method and device for detecting defects of welding spots of battery | |
| CN114757949B (en) | Wire and cable defect detection method and system based on computer vision | |
| CN111862064B (en) | Silver wire surface flaw identification method based on deep learning | |
| CN114359270B (en) | Computer vision-based automobile engine oil way copper sleeve defect detection method | |
| CN110253019B (en) | Quality monitoring and control method for selective laser melting | |
| CN112083017A (en) | Intelligent detection device, method and medium for internal defects of welding seam | |
| CN115082683A (en) | Injection molding defect detection method based on image processing | |
| CN111815555A (en) | Metal additive manufacturing image detection method and device combining anti-neural network with local binary | |
| CN114706358B (en) | Method and system for processing welding process information of straight welded pipe | |
| CN111667462A (en) | Automatic detection method and system for small and weak targets in welding defects | |
| CN115984284A (en) | X-ray hull weld image detection method for ship maintenance | |
| CN108038838A (en) | A kind of cotton fibriia species automatic testing method and system | |
| CN114897898A (en) | Wood board quality classification method based on image processing | |
| CN114782329A (en) | Bearing defect damage degree evaluation method and system based on image processing | |
| CN115311484B (en) | Optimized clustering segmentation method for weld data | |
| CN114910480A (en) | Wafer surface defect detection method based on machine vision | |
| CN115719332A (en) | Welding quality detection method | |
| CN111161237A (en) | Fruit and vegetable surface quality detection method, storage medium and sorting device thereof | |
| CN114926407A (en) | Steel surface defect detection system based on deep learning | |
| CN113012123A (en) | Classification recognition and quantitative analysis method and system for defect and damage of carbon fiber composite material | |
| CN111257422B (en) | Wheel axle defect identification model construction method and defect identification method based on machine vision | |
| CN115035092A (en) | Image-based bottle detection method, device, equipment and storage medium | |
| CN115880280A (en) | Detection method for quality of steel structure weld joint | |
| CN116630329A (en) | Online visual detection method for multi-axis multi-channel numerical control system | |
| CN115294527A (en) | Subway tunnel damage detection method based on computer vision |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |