CN108665452A - A kind of pipeline-weld film scanning storage and identification of Weld Defects and its system based on big data - Google Patents
A kind of pipeline-weld film scanning storage and identification of Weld Defects and its system based on big data Download PDFInfo
- Publication number
- CN108665452A CN108665452A CN201810438166.XA CN201810438166A CN108665452A CN 108665452 A CN108665452 A CN 108665452A CN 201810438166 A CN201810438166 A CN 201810438166A CN 108665452 A CN108665452 A CN 108665452A
- Authority
- CN
- China
- Prior art keywords
- weld
- image
- defect
- weld seam
- big data
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 116
- 238000003860 storage Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 35
- 230000035515 penetration Effects 0.000 claims abstract description 24
- 239000002893 slag Substances 0.000 claims abstract description 23
- 238000001514 detection method Methods 0.000 claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 19
- 230000004927 fusion Effects 0.000 claims abstract description 19
- 238000003909 pattern recognition Methods 0.000 claims abstract description 11
- 238000012216 screening Methods 0.000 claims abstract description 10
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000001228 spectrum Methods 0.000 claims abstract description 7
- 208000037656 Respiratory Sounds Diseases 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 8
- 238000011068 loading method Methods 0.000 claims description 8
- WFKWXMTUELFFGS-UHFFFAOYSA-N tungsten Chemical compound [W] WFKWXMTUELFFGS-UHFFFAOYSA-N 0.000 claims description 7
- 229910052721 tungsten Inorganic materials 0.000 claims description 7
- 239000010937 tungsten Substances 0.000 claims description 7
- 238000003708 edge detection Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000000227 grinding Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 238000005336 cracking Methods 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 230000014759 maintenance of location Effects 0.000 claims 1
- 238000012795 verification Methods 0.000 abstract description 8
- 230000002159 abnormal effect Effects 0.000 abstract description 5
- 230000008439 repair process Effects 0.000 abstract description 5
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 21
- 238000007726 management method Methods 0.000 description 6
- 238000003466 welding Methods 0.000 description 6
- 238000010276 construction Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000009659 non-destructive testing Methods 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 238000007405 data analysis Methods 0.000 description 3
- 238000013499 data model Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000009412 basement excavation Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000004321 preservation Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 235000002918 Fraxinus excelsior Nutrition 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 239000002956 ash Substances 0.000 description 1
- 239000011324 bead Substances 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 239000004205 dimethyl polysiloxane Substances 0.000 description 1
- 235000013870 dimethyl polysiloxane Nutrition 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 210000002445 nipple Anatomy 0.000 description 1
- CXQXSVUQTKDNFP-UHFFFAOYSA-N octamethyltrisiloxane Chemical compound C[Si](C)(C)O[Si](C)(C)O[Si](C)(C)C CXQXSVUQTKDNFP-UHFFFAOYSA-N 0.000 description 1
- 230000005693 optoelectronics Effects 0.000 description 1
- 239000003209 petroleum derivative Substances 0.000 description 1
- 238000004987 plasma desorption mass spectroscopy Methods 0.000 description 1
- 229920000435 poly(dimethylsiloxane) Polymers 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 238000002601 radiography Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000004513 sizing Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
- G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Pathology (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of pipeline-weld film scanning storage and identification of Weld Defects based on big data, including step：(1) Target pipe weld seam egative film is scanned, carries out weld seam film scanning storage；(2) weld seam spectrum data library is established；(3) weld seam big data image deflects pattern recognition model is established, analyzes and determines weld seam slag inclusion, incomplete fusion, lack of penetration risk present in negative information image, find out defect that may be present in radiographic film；(4) weld seam big data image recognition analysis is carried out, and provides report；(5) screening report, and propose hidden danger risk point.Additionally provide pipeline-weld scanning storage and defect recognition algorithm based on big data accordingly.This method completes all related film scannings, identification and storage to system, and monitoring signals are checked in pipeline, and defect, which is excavated, verifies and repair, and interior detection is abnormal to synchronize comparison verification with weld seam egative film defect, has found risk hidden danger that may be present.
Description
Technical field
The present invention relates to weld seam detection technical fields, are put in storage more particularly to the pipeline-weld film scanning based on big data
And Welding Line Flaw Detection technology.
Background technology
Pipeline-weld is important one of the feature of pipeline, and quality directly affects the safety of pipeline, since two thousand five, by
There are many accident caused by welding quality, and according to statistics, construction period and runtime occur more than 20 altogether, the mitered of such as puckery peaceful blue multiple line
Weld seam gas leakage accident, unconcerned big wire bonding seam standard-sized sheet splits oil accident within 2011, is caused by weld seam, although can not rule out outer
The influence of power, but welding condition, temperature and weldquality etc. reflect Novel presentation.Accident caused by weld seam is shown as：Pipe
Road touches dead mouth；Weld seam actinogram is unqualified, hides defective；Weld seam radiographic film it is corresponding with weld bond not on.Weld seam radiographic film
It is the vital document for detecting weldquality, the accuracy differentiated is still important measurement index, but the weld data amount of being related to
Greatly, information is more, and identification difficulty is big.
Have begun to start to walk in terms of the identification judge of radiographic film computer, for weld seam, the auto judge of radioscopic image into
It has gone extensively and in-depth study, has related to the input and output of image, pre-processes, at the contours extract of defect and some pseudo-colours
The basic functions such as reason, and basis for estimation is established for the division of bar defect, it gives a kind of based on the improved of weld image
RHT innovatory algorithms improve the arithmetic speed of system.X-ray digital image is handled with image processing techniques simultaneously
On the basis of, the statistic law used in mode identification method is classified defect, realizes the meter of weld seam digital picture
Calculation machine auxiliary is judged, and using X-ray real-time detecting system on-line checking and analysis, can effectively overcome artificial comment caused by piece accidentally
Sentence, to make weld seam egative film defect on-line checking work objectify, regulation and standardization, be by computer, automatically control, machine
The leap that numerous subjects such as tool transmission, non-destructive testing are effectively combined brings enormous impact to weld seam detection traditional handicraft.
However weld seam big data model, it is related to construction period data, pipeline detection data etc., needs to consider and comment
Sentence, need to consider it is multifactor under the conditions of pipeline data application problem, in terms of pipeline-weld big data model, it is domestic not yet
It is studied.It is well known that weld defect or implicit problem can be found out by way of big data analysis, finds out and touch dead mouth
Whole egative films of position are exactly particularly the big data analysis model by establishing image automatic identification, are transported to in-service pipeline
The quality auto judge of departure date weld seam, this is the important channel for solving construction period weld seam leftover problem.Therefore, pass through big data
Mode identification technology finds out risk hidden danger that may be present, and it is very necessary to establish weld seam big data model.
Invention content
The inventive concept of the present invention is to be directed to can not find weld seam by big data mode identification technology in the prior art
Defect simultaneously carries out specific defect identification, to provide a kind of pipeline-weld film scanning storage and defect recognition based on big data
Method, the method includes the steps：
(1) Target pipe weld seam egative film is scanned, carries out weld seam film scanning storage；
(2) weld seam spectrum data library is established；
(3) weld seam big data image deflects pattern recognition model is established, analyzes and is welded present in determining negative information image
Slag inclusion, incomplete fusion, lack of penetration risk are stitched, defect that may be present in radiographic film is found out；
(4) weld seam big data image recognition analysis is carried out, and provides report；
(5) screening report, and propose hidden danger risk point.
Preferably, the step (1) detects the special digital system MII-900plus of industrial film using industrial x-ray
System or high intensity viewbox carry out radiographic film digitlization, are clear storable electronization text by the scanning of pipeline radiographic film
Part, and assist in data loading to generalized information system, the MII-900plus provides the scanning software work(that NDT/RT images need
Can and processing measuring function, and execute image processing function, including display, inquiry, measurement, mark, make report, storage and burn
Record.
Preferably, the step (2) includes：
(2-1) is different according to the corresponding blackness of different defect characteristics, establishes egative film defect library, the defect characteristic includes：
Lack of penetration, incomplete fusion, grinding wheel fray and easily cause the characteristic feature of fatigue cracking, including tungsten inclusion, circular flaw and sting
Side；
(2-2) establishes crack defect library according to the typical photographic on egative film, and the crackle includes longitudinal crack, root crack
And transversal crack；The typical photographic includes：There is small sawtooth on the black line of sharp outline or black silk, black line or black silk, has
Bifurcated, thickness and blackness change sometimes；Or mutually wind shape in thicker black line and thinner black silk；The end of line is tapering,
There is Filamentous shade to extend in front of end sometimes；
(2-3) designs concordance list, and radiographic film is finally scanned and tied by typing concordance list data and by concordance list data loading
Fruit is put in storage, and to improve pipeline foundation database, realizes quick search, improves contingency management.
Preferably, the weld seam big data image deflects pattern recognition model of establishing of the step (3) is based on X-ray
Weld image, establish feature extraction and the automatic identification model of defect, including step：
(3-1) pre-processes weld image using the method that mean filter and medium filtering are combined；
(3-2) compares two class algorithm for image enhancement, and image enhancement is carried out using histogram equalizing method；
(3-3) is split welded seam area using iteration threshold image segmentation algorithm, and carries out feature to weld defect
Extraction and feature selecting；
(3-4) carries out Classification and Identification, classification using the SVM model classifiers method based on binary tree to weld defect type
Identify the characteristic parameter collection constituted based on multiple parameters.
Preferably, the step (3-1) further includes applying the local edge for identifying the weld image by Canny operators
Optimal edge detection algorithm, the algorithm be the calculus of variations, the actual edge for identifying the weld image, enabling to the greatest extent
The actual edge in image may mostly be identified, the edge identified will to the greatest extent may be used with the actual edge in real image as far as possible
Edge that can be in close and image can only identify once, it is understood that there may be picture noise should not be identified as edge.
Preferably, the step (3-1) further includes applying edge detection algorithm by Log operators, first to the weld seam
Image does gaussian filtering, then asks its Laplce's second dervative, i.e., the Laplce of described weld image and Gaussian function again
Transformation is filtered operation, finally, the edge of the weld image is obtained by the zero crossing of detection filter result.
Preferably, the step (3-1) further includes finding the weld seam using local difference operator by Roberts operators
Image border, using the difference approximate gradient amplitude detection edge of adjacent two pixel of diagonal.
Preferably, the multiple parameters of the step (3-4) include defect and background gray scale difference, contrast C ON, entropy ENT, circle
Shape degree e and equivalent area S/C；The weld defect type includes：Crackle, lack of penetration, incomplete fusion, stomata, spherical slag inclusion with
And strip slag inclusion.
Preferably, the step (4) includes：
(4-1) handles digitized image, including pretreatment, by sharpen and/or Laplace operator to described
Weld image is handled；
(4-2) defect recognition, and it is sharpened processing and blackness identification；
(4-3) uses the discrete first difference operators of Sobel, and the First-order Gradient for calculating the weld image luminance function is close
Like value, this operator is used in any point of the weld image, generate the corresponding gradient vector of point or its law vector to
Carry out brightness identification；
(4-4) finds out crackle by Roberts Edge contrasts；
(4-5) carries out weld seam other defect identification, and weld defect type includes：Lack of penetration, incomplete fusion, stomata, spherical folder
Slag and strip slag inclusion.
The present invention also aims to provide a kind of pipeline-weld scanning storage and defect recognition system based on big data,
Including：
(1) weld seam film scanning enters library module, for Target pipe weld seam egative film to be scanned, carries out weld seam egative film and sweeps
Retouch storage；
(2) weld seam spectrum data library, for storing the relevant collection of illustrative plates of weld defect；
(3) big data image deflects pattern recognition model and defect recognition module, for establishing weld seam big data image
Defect mode identification model is analyzed and determines weld seam slag inclusion, incomplete fusion, lack of penetration risk present in negative information image, looks for
Defect that may be present in emergent ray egative film；
(4) picture recognition module for carrying out weld seam big data image recognition analysis, and provides report；
(5) it reports screening module, for screening report, and proposes hidden danger risk point.
Using the pipeline-weld scanning storage and defect recognition system and method based on big data, all correlations are completed
Film scanning, identification and storage, complete monitoring signals check in pipeline, the excavation verification of defect and repair, interior detection it is abnormal with
Weld seam egative film defect has carried out synchronous contrast verification, will excavate verification, non-destructive testing, flaw evaluation and repair as routine work
Plan, has found risk hidden danger that may be present.
According to the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will be brighter
The above and other objects, advantages and features of the present invention.
Description of the drawings
Some specific embodiments that the invention will be described in detail by way of example and not limitation with reference to the accompanying drawings hereinafter.
Identical reference numeral denotes same or similar component or part in attached drawing.It should be appreciated by those skilled in the art that these
What attached drawing was not necessarily drawn to scale.The target and feature of the present invention will be apparent from view of following description taken together with the accompanying drawings,
In attached drawing：
Attached drawing 1 is the engineering background work flow diagram implemented according to the embodiment of the present invention；
Attached drawing 2 is the pipeline-weld scanning storage and defect identification method stream based on big data according to the embodiment of the present invention
Cheng Tu；
Attached drawing 3 is several typical defect feature egative film views according to the embodiment of the present invention, and wherein Fig. 3-1 indicates there is circle
Shape defect, the weld seam of tungsten inclusion；Fig. 3-2 indicates the weld seam with circular flaw and incomplete penetration defect；Fig. 3-3 indicates to have lack of penetration
The weld seam of defect；Fig. 3-4 indicates there is the weld seam for not merging defect；Fig. 3-5 indicate with circular flaw, do not merge and undercut lack
Sunken weld seam；Fig. 3-6 indicates that grinding wheel frays and trace rather than does not merge the weld seam of defect；Fig. 3-7 indicate have do not merge, tungsten inclusion
With the weld seam of circular flaw；Fig. 3-8 indicates there is the weld seam for not merging defect；Fig. 3-9 indicates there is the weld seam for not merging defect；
Attached drawing 4 is the typical longitudinal crack according to the embodiment of the present invention, transversal crack and root crack schematic diagram and weld seam
Egative film schematic diagram, wherein Fig. 4-1 indicate the schematic diagram and weld seam egative film schematic diagram of transversal crack and longitudinal crack；Fig. 4-2 is indicated
Root crack schematic diagram and weld seam egative film schematic diagram；Fig. 4-3 indicates transversal crack schematic diagram and weld seam egative film schematic diagram；
Attached drawing 5 is the weld seam egative film schematic diagram for including crackle and other defect according to the embodiment of the present invention, wherein Fig. 5-1
It indicates with crackle, merge and the weld seam of circular flaw；Fig. 5-2 is indicated with crackle, lack of penetration and circular flaw
Weld seam；
Attached drawing 6 is to carry out pretreated negative map to weld image according to the embodiment of the present invention, and wherein Fig. 6-1 is indicated
Weld seam original image and the image comparison after Canny operators and the processing of Log operators；Fig. 6-2 indicates gas hole defect Roberts
Operator treated image；
Attached drawing 7 indicates the binary tree algorithm structure chart according to the ... of the embodiment of the present invention for weld defect classification；
It is attached that Fig. 8 shows the weld seam egative films according to embodiments of the present invention for carrying out weld seam big data image recognition analysis and obtaining successively
Diagram；
Attached drawing 9 indicates pipeline-weld scanning storage and defect recognition system according to the ... of the embodiment of the present invention based on big data
Principle schematic.
Specific implementation mode
Attached drawing 1 is the engineering background work flow diagram implemented according to the embodiment of the present invention, can be obtained according to weld seam radiographic film
Property is obtained, this project is divided into two parts and carries out respectively：(1) part of pipeline-weld radiographic film can be found；(2) no-welding-seam
The part of radiographic film.1 main working process includes with reference to the accompanying drawings：1) weld seam negative plate digitization：Work is detected using industrial x-ray
The special digital system of industry film：MII-900plus or high intensity viewbox, carry out radiographic film digitlization, and by its
Individually it is converted into the single file of 330mm.It is required that color depth：8bit/16bit grayscale (the grayscale of 256 grey stratum/65536
Layer)；Optical resolution：Default per inch 300 points (300dpi), maximum should be able to reach 2400dpi；Weld seam scanning blackness reaches
2.5 or more, maximum blackness can reach 4.7.2) it establishes weld seam egative film database and is put in storage：Weld seam egative film database is established, including
Concordance list designs, and radiographic film scanning result is finally put in storage by concordance list data inputting, three big step of concordance list data loading,
To improve pipeline foundation database, quick search is realized, it is horizontal to improve contingency management.3) piece is repaiied before the identification of scanning egative film：In order to
The identification for ensuring defect of pipeline carries out graphic operation to weld tabs using specialized software, handles the textural characteristics of weld tabs, special needle
To the egative film of poor quality, fully defect is exposed, it is ensured that the defect of pipeline weld tabs is easy to identify, not dead angle, non-blind area.Side
The non-welding bead image of edge label should be cleared up；4) Defect Scanning identification and evaluation：It is identified and is evaluated using weld seam radiographic film
System carries out defect characteristic extraction and the automatic identification of weld image, selects image enhancement technique, carried out to weld defect feature
Extraction, finds out weld defect feature, such as crackle, lack of penetration, incomplete fusion, stomata, spherical slag inclusion and strip slag inclusion, most automatically
Small identification flaw size feature should be not more than 1mm, and evaluate the acceptability of defect, and provide weld seam egative film automatically and lack
Fall into identification report；5) manual review's software recognition result when necessary：RT-III grades of personnel are selected, irregularly to weld seam egative film number
Change, identify, result of reexamining is checked, it is ensured that project implementation quality；6) detection signal check in：To there are the welderings of weld seam egative film
Detection signal is checked in seam, is paid close attention to：Construction period is rated II grades of egative film, reprocesses mouth, Jin Kou, pipe nipple (length
1.5m or less) both ends weld seam, elbow both ends weld seam, wear crossing pipeline both ends weld seam, geological disaster high risk pipeline section weld seam and interior
The weld seam of examining report is abnormal；7) weld seam negative plate digitization file and identification, result of reexamining storage：Plan weld seam negative plate digitization
File enters DPMS systems, plans to enter Technical Data Management in the presence of abnormal weld seam negative plate digitization file and identification, result of reexamining
In system PDMS systems, and project end result is subjected to archive of company level by DCC；8) defect is excavated verification and is repaired：According to
Flaw evaluation is as a result, check post is excavated in selection, and is repaired in time.
Attached drawing 2 is the pipeline-weld scanning storage and defect identification method stream based on big data according to the embodiment of the present invention
Cheng Tu, including step：
(1) Target pipe weld seam egative film is scanned, carries out weld seam film scanning storage, including be suspicious and resentful pipeline foundation letter
Breath, pipeline welding data simultaneously carry out pipeline welding film scanning；
The step detects the special digital system MII-900plus systems or high intensity of industrial film using industrial x-ray
Viewbox carries out radiographic film digitlization, is clear storable electronic file by the scanning of pipeline radiographic film, and assist by
In data loading to generalized information system, the Integrity Management for substantially increasing pipeline owner is horizontal, and scans egative film with good
Quality maintains the effect consistent with the blackness of original negative, clarity, and being capable of amplifying observation.
The preservations such as record and report are detected according to JBT4730.1-2005 bearing device non-destructive testing part 1 General Requirements
After phase must not be less than 7 years, 7 years, if user needs to preserve with handover user.In addition, SY4056-93《Petroleum gas steel
Sizing In Butt Welded Pipes radiography and quality grading》The photographic density allowable range is required to be in standard：1.2~3.5, it provides effectively
Examining report.Egative film is provided to owner, if desired oneself preserves egative film, then egative film archive should at least 5 years.That is, bottom
The holding time of piece was generally for 7 years, while to meet the requirements such as temperature, humidity, and to prevent it from making moist, water logging etc. no
Sharp influence factor, long-term preservation is difficult and condition is harsh.Egative film needs are moisture-proof, seal, is sun-proof.Drier is not permanently to have
Effect is replaced 1 time for general 1 year, and unsuitable over-drying, otherwise film becomes fragile.Most of all, pipeline radiographic film substantial amounts,
Enterprise does not have the ability for preserving egative film and space, and special messenger is needed to take care of.Inconvenience remotely is had access to, onsite application generally requires
Take scene to.And the generalized information system that enterprise uses at present is badly in need of ray data, film scanning image storage enriches pipeline foundation number
According to library, it can be achieved that quick search.And egative film electronization can be solved the problems, such as into above-mentioned four class.Scanner is can not to scan bottom
Piece, it needs to use film scanning digital imaging system.Dedicated X-ray film digital scanner, can be to existing traditional x-ray
Film scanning is allowed to be converted into digitized image.It is rotated using He-Ne Lasers by polyhedron using the principle of opto-electronic conversion
Formula reflective mirror is scanned existing X-ray film, is turned the optical signal received by the automatic tracking receiver of quick multichannel
Become electric signal, digital signal data is converted by analog/digital (A/D) converter, realizes that film image is converted to number
Change image, so as to store and recycle in a computer.High performance X-ray film digital scanner can make conversion loss
The caused distortion factor reaches very little.
MII-900plus belongs to special pipeline ray digitization system, and detecting industrial film specifically for industrial x-ray carries
The scanning software function and processing measuring function needed for NDT/RT images, can increase image dark space level of detail with it is clear
Degree, meets the needs of NDT/RT is to image.Operate it is simple, feature-rich, facilitate storage, quickly share.MII-900plus's sweeps
Software (MiiNDT) is retouched, not only simplifies procedures, save the time, image specially treated measuring function etc. is also provided, can perform
A series of image processing functions, including display, inquiry, measurement, mark, make report, storage, burning, allow NDT/RT images more
It easily observes, store and shares.MII-900plus arranges in pairs or groups dedicated scanning software (MiiNDT) can for industrial file word
It supports one to one equal proportion printing in kind, can directly compare original material object, immediately know that damage position, facilitate construction, promote effect
Energy.NDT/RT non-destructive testing negative plate digitization systems are using efficiently industrial film image collection technology, long-range interpretation optimal images
Acquisition scheme simplifies existing work flow and provides whole resource-sharing.
Technical parameter is：Scanning range：It is maximum：14 inches * 52 inches；Color depth：8bit/16bit grayscale (256 ashes
The grey stratum in stratum/65536)；Optical resolution：Default per inch 300 points (300dpi) is up to 2400dpi；Maximum dynamic is close
Degree：4.7Dmax；Appearance and size：260*474*235mm (long * wide * high).
Working environment and realization function：1) tool interface system：Hi-speed USB interface (2.0 interfaces of USB)；(2) operating system：
Windows XP/2000；(3) egative film maximal density can be scanned：4.7D；(4) equipment regulative mode：It is adjusted through software i.e. adjustable
Save speed and brightness；(5) imaging sensor：CCD is linear；(6) color depth：Grayscale 8/16bit (256 ranks/65536 rank)；(7)
Sweep speed：≤ 18sec@300dpi, grayscale (14 " × 17 ")；(8) bundled software：MiiNDT；(9) auxiliary sentences piece function：Make
With the image software accurately measuring of mating profession, defect, partial enlargement are marked, brightness regulation etc. assists sentencing piece；(10) report system
Make function：Picture can be directly printed in report, keeps report description more concise；(11) digital document management work(
Energy：Negative plate digitization stores, transmission, the management such as inquiry；(12) remotely sentence piece：Digitized image can be converted international standard
DICONDE formats (can not be changed), and realization remotely sentences piece.
The embodiment is by taking the two wires of Shan capital as an example, caliber 1016MM, and egative film overall length is 3.19 meters, bottom panel width 80mm,
Ray film making is divided into 2 egative films, every 1.7 meters or so of length.It is calculated according to 1000km, egative film sum is 200,000 left sides in total
It is right.Using MII-900plus systems, egative film density 0-4.7Dmax can be scanned, general photographic density is 3.5.
(2) data check is carried out after establishing weld seam spectrum data library, including original welding record acquisition, obtains the quality of data
Egative film data are received after report, and weld data is aligned after data loading, then weld data is put in storage, is specifically included：
(2-1) is different according to the corresponding blackness of different defect characteristics, establishes egative film defect library, the defect characteristic includes：
Lack of penetration, incomplete fusion, grinding wheel fray and easily cause the characteristic feature of fatigue cracking, including tungsten inclusion, circular flaw and sting
Side, several typical defect feature egative film views as shown in Figure 3, wherein Fig. 3-1 indicate there is circular flaw, the weld seam of tungsten inclusion；
Fig. 3-2 indicates the weld seam with circular flaw and incomplete penetration defect；Fig. 3-3 indicates the weld seam with incomplete penetration defect；Fig. 3-4 tables
Show with the weld seam for not merging defect；Fig. 3-5 is indicated with circular flaw, is not merged and the weld seam of undercut defect；Fig. 3-6 is indicated
Grinding wheel, which frays, trace rather than does not merge the weld seam of defect；Fig. 3-7 indicate have do not merge, the weld seam of tungsten inclusion and circular flaw；
Fig. 3-8 indicates there is the weld seam for not merging defect；Fig. 3-9 indicates there is the weld seam for not merging defect.
(2-2) establishes crack defect library according to the typical photographic on egative film, and the crackle includes longitudinal crack, root crack
And transversal crack；The typical photographic includes：There is small sawtooth on the black line of sharp outline or black silk, black line or black silk, has
Bifurcated, thickness and blackness change sometimes；Or mutually wind shape in thicker black line and thinner black silk；The end of line is tapering,
There is in front of end Filamentous shade to extend sometimes, typical longitudinal crack as shown in Figure 4, transversal crack and root crack schematic diagram and
Weld seam egative film schematic diagram, wherein Fig. 4-1 indicate the schematic diagram and weld seam egative film schematic diagram of transversal crack and longitudinal crack；Fig. 4-2
Indicate root crack schematic diagram and weld seam egative film schematic diagram；Fig. 4-3 indicates transversal crack schematic diagram and weld seam egative film schematic diagram.Figure
5 is include the weld seam egative film schematic diagram of crackle and other defect in embodiment, wherein Fig. 5-1 indicate to have crackle, do not merge with
And the weld seam of circular flaw；Fig. 5-2 is indicated with crackle, lack of penetration and circular flaw weld seam.
(2-3) designs concordance list, and radiographic film is finally scanned and tied by typing concordance list data and by concordance list data loading
Fruit is put in storage, and to improve pipeline foundation database, realizes quick search, improves contingency management.Wherein concordance list and data dictionary point
Shown in not following Tables 1 and 2：
1 concordance list of table
2 data dictionary of table
(3) weld seam big data image deflects pattern recognition model is established, analyzes and is welded present in determining negative information image
Slag inclusion, incomplete fusion, lack of penetration risk are stitched, finds out defect that may be present in radiographic film, which includes establishing defect
Database, the acquisition of weld defect criterion and the clustering for carrying out weld image are finally completed the identification point of weld seam big data
Analysis, it is the weld image based on X-ray to establish weld seam big data image deflects pattern recognition model, establishes the feature extraction of defect
With automatic identification model, including step：
(3-1) pre-processes weld image using the method that mean filter and medium filtering are combined, pretreated
Mode includes：
(3-1-1) applies the optimal edge detection algorithm for the local edge for identifying the weld image by Canny operators,
The algorithm is the calculus of variations, the actual edge for identifying the weld image, enabling identify image as much as possible
In actual edge, the edge identified will as far as possible with the actual edge in real image as close possible to and image in
Edge can only identify once, it is understood that there may be picture noise should not be identified as edge；
(3-1-2) applies edge detection algorithm by Log operators, does gaussian filtering to the weld image first, then
Its Laplce's second dervative, i.e., the Laplace transform of the described weld image and Gaussian function is asked to be filtered operation again, most
Afterwards, obtain the edge of the weld image by the zero crossing of detection filter result, weld seam original image and by Canny operators and
Image comparison after the processing of Log operators is as in Figure 6-1；Or
(3-1-3) finds the weld image edge by Roberts operators using local difference operator, using diagonal line
The difference approximate gradient amplitude detection edge of adjacent two pixel in direction, as in fig. 6-2；
(3-2) compares two class algorithm for image enhancement, and image enhancement is carried out using histogram equalizing method；
(3-3) is split welded seam area using iteration threshold image segmentation algorithm, and carries out feature to weld defect
Extraction and feature selecting；
(3-4) carries out Classification and Identification, classification using the SVM model classifiers method based on binary tree to weld defect type
Identify the characteristic parameter collection that is constituted based on multiple parameters, plurality of parameter include defect and background gray scale difference, contrast C ON,
Entropy ENT, circularity e and equivalent area S/C；The weld defect type includes：Crackle, lack of penetration, incomplete fusion, stomata, ball
Shape slag inclusion and strip slag inclusion, shown in the binary tree SVM algorithm structure chart classified for weld defect such as Fig. 7, wherein A is indicated not
Fusion, B indicate lack of penetration, and C indicates that stomata, D expressions are mingled with, and by the description of the figure, obtains the SVM structure charts of full blast；
Table 3 and table 4 indicate to separate the knowledge of defect type and sample in the embodiment using each parameter of 5 characteristic parameter collection respectively
Not rate, as shown in the table.
Table 3 separates defect type using each parameter of 5 characteristic parameter collection
The discrimination statistical chart that table 4 is determined according to sample type
| Sample names | Sample size | Identify quantity | Discrimination (%) |
| Crackle | 32 | 30 | 93.8 |
| Stomata | 137 | 129 | 94.2 |
| Incomplete fusion | 19 | 17 | 89.5 |
| It is lack of penetration | 28 | 25 | 89.3 |
| Spherical slag inclusion | 41 | 38 | 92.7 |
| Strip slag inclusion | 21 | 17 | 81.0 |
| Total amount | 278 | 256 | 92.1 |
(4) it uses software to carry out weld seam big data image recognition analysis, and provides report, specifically comprise the following steps：
(4-1) handles digitized image, including pretreatment, by sharpen and/or Laplace operator to described
Weld image is handled, and the artwork as shown in Fig. 8-1 and such as Fig. 8-2 are shown to pass through pretreated digitized image, such as schemes
By sharpening and/or Laplace operator is to the weld image image that carries out that treated shown in 8-3；
(4-2) defect recognition as shown in fig. 8-4 and is sharpened processing and blackness identification, as shown in Fig. 8-5；
(4-3) uses the discrete first difference operators of Sobel, and the First-order Gradient for calculating the weld image luminance function is close
Like value, this operator is used in any point of the weld image, generate the corresponding gradient vector of point or its law vector to
Brightness identification is carried out, as shown in Fig. 8-6；
(4-4) finds out crackle by Roberts Edge contrasts, as shown in Fig. 8-7；
(4-5) carries out weld seam other defect identification, and weld defect type includes：Lack of penetration, incomplete fusion, stomata, spherical folder
Slag and strip slag inclusion；
(5) screening report, and propose hidden danger risk point.
Attached drawing 9 also provides a kind of pipeline-weld scanning storage and defect recognition system based on big data, including：
(1) weld seam film scanning enters library module, for Target pipe weld seam egative film to be scanned, carries out weld seam egative film and sweeps
Retouch storage；
(2) weld seam spectrum data library, for storing the relevant collection of illustrative plates of weld defect；
(3) big data image deflects pattern recognition model and defect recognition module, for establishing weld seam big data image
Defect mode identification model is analyzed and determines weld seam slag inclusion, incomplete fusion, lack of penetration risk present in negative information image, looks for
Defect that may be present in emergent ray egative film；
(4) picture recognition module for carrying out weld seam big data image recognition analysis, and provides report；
(5) it reports screening module, for screening report, and proposes hidden danger risk point.
Using the pipeline-weld scanning storage and defect recognition system and method based on big data, all correlations are completed
Film scanning, identification and storage, complete monitoring signals check in pipeline, the excavation verification of defect and repair, interior detection it is abnormal with
Weld seam egative film defect has carried out synchronous contrast verification, will excavate verification, non-destructive testing, flaw evaluation and repair as routine work
Plan, has found risk hidden danger that may be present.
It, will not be by these embodiments although the present invention is described by reference to specific illustrative embodiment
Restriction and only limited by accessory claim.It should be understood by those skilled in the art that can be without departing from the present invention's
The example of the present invention can be modified and be changed in the case of protection domain and spirit.
Claims (10)
1. a kind of pipeline-weld film scanning storage and identification of Weld Defects based on big data, it is characterised in that the side
Method includes step：
(1) Target pipe weld seam egative film is scanned, carries out weld seam film scanning storage；
(2) weld seam spectrum data library is established；
(3) weld seam big data image deflects pattern recognition model is established, analyzes and determines the folder of weld seam present in negative information image
Slag, incomplete fusion, lack of penetration risk find out defect that may be present in radiographic film；
(4) weld seam big data image recognition analysis is carried out, and provides report；
(5) screening report, and propose hidden danger risk point.
2. a kind of pipeline-weld scanning storage and defect identification method based on big data according to claim 1, special
Sign is that the step (1) detects the special digital system MII-900plus systems or height of industrial film using industrial x-ray
Intensity viewbox carries out radiographic film digitlization, is clear storable electronic file by the scanning of pipeline radiographic film, and auxiliary
It helps in data loading to generalized information system, the MII-900plus provides the scanning software function and processing that NDT/RT images need
Measuring function, and image processing function is executed, including display, inquiry, measurement, mark, making report, storage and burning.
3. a kind of pipeline-weld scanning storage and defect identification method based on big data according to claim 1, special
Sign is that the step (2) includes：
(2-1) is different according to the corresponding blackness of different defect characteristics, establishes egative film defect library, the defect characteristic includes：It does not weld
Thoroughly, incomplete fusion, grinding wheel fray and easily cause the characteristic feature of fatigue cracking, including tungsten inclusion, circular flaw and undercut；
(2-2) establishes crack defect library according to the typical photographic on egative film, and the crackle includes longitudinal crack, root crack and cross
To crackle；The typical photographic includes：There is small sawtooth on the black line of sharp outline or black silk, black line or black silk, there is bifurcated,
Thickness and blackness change sometimes；Or mutually wind shape in thicker black line and thinner black silk；The end of line is tapering, before end
Side has Filamentous shade to extend sometimes；
(2-3) designs concordance list, and typing concordance list data and by concordance list data loading finally enter radiographic film scanning result
Quick search is realized in library to improve pipeline foundation database, improves contingency management.
4. a kind of pipeline-weld film scanning storage and Welding Line Flaw Detection side based on big data according to claim 1
Method, it is characterised in that the weld seam big data image deflects pattern recognition model of establishing of the step (3) is based on X-ray
Weld image establishes feature extraction and the automatic identification model of defect, including step：
(3-1) pre-processes weld image using the method that mean filter and medium filtering are combined；
(3-2) compares two class algorithm for image enhancement, and image enhancement is carried out using histogram equalizing method；
(3-3) is split welded seam area using iteration threshold image segmentation algorithm, and carries out feature extraction to weld defect
And feature selecting；
(3-4) carries out Classification and Identification, Classification and Identification using the SVM model classifiers method based on binary tree to weld defect type
The characteristic parameter collection constituted based on multiple parameters.
5. a kind of pipeline-weld film scanning storage and Welding Line Flaw Detection side based on big data according to claim 4
Method, it is characterised in that the step (3-1) further includes applying the local edge for identifying the weld image by Canny operators
Optimal edge detection algorithm, the algorithm are the calculus of variations, the actual edge for identifying the weld image, enabling to the greatest extent may be used
Can mostly identify actual edge in image, the edge identified will as far as possible with the actual edge in real image as far as possible
Edge in close and image can only identify once, it is understood that there may be picture noise should not be identified as edge.
6. a kind of pipeline-weld scanning storage and defect identification method based on big data according to claim 4, special
Sign is that the step (3-1) further includes applying edge detection algorithm by Log operators, is Gauss to the weld image first
Filtering, then asks its Laplce's second dervative, i.e., the Laplace transform of the described weld image and Gaussian function to be filtered again
Wave operation finally obtains the edge of the weld image by the zero crossing of detection filter result.
7. a kind of pipeline-weld film scanning storage and Welding Line Flaw Detection side based on big data according to claim 4
Method, it is characterised in that described further includes finding the weld image edge using local difference operator by Roberts operators, is adopted
With the difference approximate gradient amplitude detection edge of adjacent two pixel of diagonal.
8. a kind of pipeline-weld film scanning storage and Welding Line Flaw Detection side based on big data according to claim 4
Method, it is characterised in that the multiple parameters of the step (3-4) include defect and background gray scale difference, contrast C ON, entropy ENT, circle
Spend e and equivalent area S/C；The weld defect type includes：Crackle, lack of penetration, incomplete fusion, stomata, spherical slag inclusion and
Strip slag inclusion.
9. a kind of pipeline-weld scanning storage and defect identification method based on big data according to claim 1, special
Sign is that the step (4) includes：
(4-1) handles digitized image, including pretreatment, by sharpen and/or Laplace operator to the weld seam
Image is handled；
(4-2) defect recognition, and it is sharpened processing and blackness identification；
(4-3) uses the discrete first difference operators of Sobel, calculates the approximation of the First-order Gradient of the weld image luminance function
Value uses this operator in any point of the weld image, generate the corresponding gradient vector of point or its law vector into
Row brightness identifies；
(4-4) finds out crackle by Roberts Edge contrasts；
(4-5) carries out weld seam other defect identification, and weld defect type includes：Lack of penetration, incomplete fusion, stomata, spherical slag inclusion with
And strip slag inclusion.
10. a kind of pipeline-weld film scanning storage and Welding Line Flaw Detection system based on big data, for implementing according to power
Profit requires any pipeline-weld film scanning storages and identification of Weld Defects based on big data of 1-9, feature
Be include：
(1) weld seam film scanning enters library module, for Target pipe weld seam egative film to be scanned, carries out weld seam film scanning and enters
Library；
(2) weld seam spectrum data library, for storing the relevant collection of illustrative plates of weld defect；
(3) big data image deflects pattern recognition model and defect recognition module, for establishing weld seam big data image deflects
Pattern recognition model is analyzed and determines weld seam slag inclusion, incomplete fusion, lack of penetration risk present in negative information image, finds out and penetrate
Defect that may be present in line egative film；
(4) picture recognition module for carrying out weld seam big data image recognition analysis, and provides report；
(5) it reports screening module, for screening report, and proposes hidden danger risk point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810438166.XA CN108665452B (en) | 2018-05-09 | 2018-05-09 | Pipeline-weld film scanning storage and identification of Weld Defects and its system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810438166.XA CN108665452B (en) | 2018-05-09 | 2018-05-09 | Pipeline-weld film scanning storage and identification of Weld Defects and its system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108665452A true CN108665452A (en) | 2018-10-16 |
| CN108665452B CN108665452B (en) | 2019-06-07 |
Family
ID=63778936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810438166.XA Active CN108665452B (en) | 2018-05-09 | 2018-05-09 | Pipeline-weld film scanning storage and identification of Weld Defects and its system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108665452B (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109741297A (en) * | 2018-11-28 | 2019-05-10 | 歌尔股份有限公司 | Product quality detection method and device |
| CN109741295A (en) * | 2018-11-28 | 2019-05-10 | 歌尔股份有限公司 | Product quality detection method and device |
| CN109800824A (en) * | 2019-02-25 | 2019-05-24 | 中国矿业大学(北京) | A kind of defect of pipeline recognition methods based on computer vision and machine learning |
| CN109856893A (en) * | 2019-03-12 | 2019-06-07 | 苏州大学 | A kind of full-automatic radiographic film projection reading set mechanism |
| CN110084807A (en) * | 2019-05-06 | 2019-08-02 | 河南大学 | A kind of detection method that seam inspection egative film image is faked |
| CN110287968A (en) * | 2019-07-01 | 2019-09-27 | 河南大学 | A kind of detection method that the seam inspection egative film image based on LBP texture is faked |
| CN110969611A (en) * | 2019-12-03 | 2020-04-07 | 广州特种承压设备检测研究院 | Pipeline weld defect detection method, device and system and storage medium |
| CN111141765A (en) * | 2020-01-17 | 2020-05-12 | 中海石油气电集团有限责任公司 | Automatic detection and identification device for weld defects of storage tank |
| CN111159836A (en) * | 2018-11-07 | 2020-05-15 | 中国石油化工股份有限公司 | Method and device for judging interactive influence of pipeline defects |
| CN111340258A (en) * | 2020-03-13 | 2020-06-26 | 新兴铸管股份有限公司 | Cast tube product quality management method based on Internet |
| CN111461569A (en) * | 2020-04-17 | 2020-07-28 | 广东大鹏液化天然气有限公司 | Safety management method for pipeline body |
| CN111597597A (en) * | 2020-05-11 | 2020-08-28 | 内蒙古京泰发电有限责任公司 | Method for creating PDMS (polydimethylsiloxane) round-the-sky and square-earth structural body equipment model |
| CN111808367A (en) * | 2020-07-09 | 2020-10-23 | 浙江七色鹿色母粒有限公司 | Improvement method for plastic PPR silver grain whitening defect |
| CN112734693A (en) * | 2020-12-18 | 2021-04-30 | 平安科技（深圳）有限公司 | Pipeline weld defect detection method and related device |
| CN113034442A (en) * | 2021-03-04 | 2021-06-25 | 北京科技大学设计研究院有限公司 | Hot-rolled strip steel surface quality grading method based on defect distribution map |
| CN113055542A (en) * | 2020-12-15 | 2021-06-29 | 陕西泰诺特检测技术有限公司 | Radiographic negative scanning device and analysis and evaluation method |
| CN113192070A (en) * | 2021-06-30 | 2021-07-30 | 深圳市超准视觉科技有限公司 | Weld image processing method and system based on computer vision |
| CN113447501A (en) * | 2021-06-28 | 2021-09-28 | 国网黑龙江省电力有限公司电力科学研究院 | Radiographic inspection method for power plant metal pipeline under operating condition |
| CN113516736A (en) * | 2021-04-23 | 2021-10-19 | 扬州能煜检测科技有限公司 | Method and device for establishing and processing discode file mark and electronic equipment |
| CN113591055A (en) * | 2021-07-27 | 2021-11-02 | 中国核工业第二二建设有限公司 | Credible weld joint digital inspection method based on block chain technology |
| CN114549836A (en) * | 2022-02-17 | 2022-05-27 | 沈阳派得林科技有限责任公司 | Method for accurately aligning long-distance pipeline radiographic image with internal detection information |
| CN114648415A (en) * | 2020-12-18 | 2022-06-21 | 苏州国双软件有限公司 | Method and device for generating pipeline integrity management scheme |
| CN114820499A (en) * | 2022-04-20 | 2022-07-29 | 上海电气集团股份有限公司 | Pipeline welding seam detection method |
| CN116228703A (en) * | 2023-02-21 | 2023-06-06 | 北京远舢智能科技有限公司 | Defect sample image generation method and device, electronic equipment and medium |
| CN118067726A (en) * | 2024-04-19 | 2024-05-24 | 天津博爱管道科技集团有限公司 | Spiral steel pipe defect detection system based on laser image recognition |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101101277A (en) * | 2007-08-10 | 2008-01-09 | 华南理工大学 | High-resolution welding seam supersonic image-forming damage-free detection method and detection system |
| CN101118225A (en) * | 2007-08-09 | 2008-02-06 | 中国航天科技集团公司长征机械厂 | Method for analyzing welding quality of aluminium alloy with X radiograph |
| CN101122585A (en) * | 2007-09-12 | 2008-02-13 | 天津大学 | Automatic identification method for supersonic phased array for detecting oil gas pipeline girth weld defect type |
| CN101559512A (en) * | 2009-05-21 | 2009-10-21 | 山东大学 | Welding track detection and control method of plate butt weld based on laser ranging |
| CN101782529A (en) * | 2009-01-20 | 2010-07-21 | 中国特种设备检测研究院 | Weld defect computer imaging and defect recognition system |
| CN103134809A (en) * | 2013-03-14 | 2013-06-05 | 苏州华源包装股份有限公司 | Welded line defect detection method |
| CN103914838A (en) * | 2014-03-25 | 2014-07-09 | 西安石油大学 | Method for identifying defects of industrial x-ray weld joint image |
| CN104574418A (en) * | 2015-01-27 | 2015-04-29 | 西安工业大学 | Pressure vessel weld defect identification method and device based on neural network |
| CN104865277A (en) * | 2015-05-11 | 2015-08-26 | 四川大学 | Method for automatically identifying pipe weld defect based on double-wall double-projection transillumination image |
| CN104914111A (en) * | 2015-05-18 | 2015-09-16 | 北京华检智研软件技术有限责任公司 | Strip steel surface defect on-line intelligent identification and detection system and detection method |
| CN104977313A (en) * | 2014-04-09 | 2015-10-14 | 四川省特种设备检验研究院 | Method and device for detecting and identifying X-ray image defects of welding seam |
| CN105588845A (en) * | 2016-01-04 | 2016-05-18 | 江苏科技大学 | Weld defect characteristic parameter extraction method |
-
2018
- 2018-05-09 CN CN201810438166.XA patent/CN108665452B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101118225A (en) * | 2007-08-09 | 2008-02-06 | 中国航天科技集团公司长征机械厂 | Method for analyzing welding quality of aluminium alloy with X radiograph |
| CN101101277A (en) * | 2007-08-10 | 2008-01-09 | 华南理工大学 | High-resolution welding seam supersonic image-forming damage-free detection method and detection system |
| CN101122585A (en) * | 2007-09-12 | 2008-02-13 | 天津大学 | Automatic identification method for supersonic phased array for detecting oil gas pipeline girth weld defect type |
| CN101782529A (en) * | 2009-01-20 | 2010-07-21 | 中国特种设备检测研究院 | Weld defect computer imaging and defect recognition system |
| CN101559512A (en) * | 2009-05-21 | 2009-10-21 | 山东大学 | Welding track detection and control method of plate butt weld based on laser ranging |
| CN103134809A (en) * | 2013-03-14 | 2013-06-05 | 苏州华源包装股份有限公司 | Welded line defect detection method |
| CN103914838A (en) * | 2014-03-25 | 2014-07-09 | 西安石油大学 | Method for identifying defects of industrial x-ray weld joint image |
| CN104977313A (en) * | 2014-04-09 | 2015-10-14 | 四川省特种设备检验研究院 | Method and device for detecting and identifying X-ray image defects of welding seam |
| CN104574418A (en) * | 2015-01-27 | 2015-04-29 | 西安工业大学 | Pressure vessel weld defect identification method and device based on neural network |
| CN104865277A (en) * | 2015-05-11 | 2015-08-26 | 四川大学 | Method for automatically identifying pipe weld defect based on double-wall double-projection transillumination image |
| CN104914111A (en) * | 2015-05-18 | 2015-09-16 | 北京华检智研软件技术有限责任公司 | Strip steel surface defect on-line intelligent identification and detection system and detection method |
| CN105588845A (en) * | 2016-01-04 | 2016-05-18 | 江苏科技大学 | Weld defect characteristic parameter extraction method |
Non-Patent Citations (4)
| Title |
|---|
| 李坤等: "基于分离程度的二叉树的多类SVM在焊接缺陷分类和识别中的应用", 《四川大学学报（自然科学版）》 * |
| 李坤等: "基于自适应SVM决策树的焊缝缺陷类型识别", 《无损检测》 * |
| 李大力: "底片扫描仪小型化及焊缝故障识别技术研究", 《中国优秀硕士学位论文全文数据库》 * |
| 罗爱民等: "基于改进二叉树多分类SVM的焊缝缺陷分类方法", 《焊接学报》 * |
Cited By (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111159836A (en) * | 2018-11-07 | 2020-05-15 | 中国石油化工股份有限公司 | Method and device for judging interactive influence of pipeline defects |
| CN109741295A (en) * | 2018-11-28 | 2019-05-10 | 歌尔股份有限公司 | Product quality detection method and device |
| CN109741297A (en) * | 2018-11-28 | 2019-05-10 | 歌尔股份有限公司 | Product quality detection method and device |
| CN109741295B (en) * | 2018-11-28 | 2021-04-20 | 歌尔股份有限公司 | Product quality detection method and device |
| CN109800824A (en) * | 2019-02-25 | 2019-05-24 | 中国矿业大学(北京) | A kind of defect of pipeline recognition methods based on computer vision and machine learning |
| CN109800824B (en) * | 2019-02-25 | 2019-12-20 | 中国矿业大学(北京) | Pipeline defect identification method based on computer vision and machine learning |
| CN109856893A (en) * | 2019-03-12 | 2019-06-07 | 苏州大学 | A kind of full-automatic radiographic film projection reading set mechanism |
| CN110084807B (en) * | 2019-05-06 | 2022-09-27 | 河南大学 | Detection method for false detection of welding seam flaw detection negative image |
| CN110084807A (en) * | 2019-05-06 | 2019-08-02 | 河南大学 | A kind of detection method that seam inspection egative film image is faked |
| CN110287968B (en) * | 2019-07-01 | 2022-09-06 | 河南大学 | Detection method for welding line flaw detection negative image counterfeiting based on LBP texture |
| CN110287968A (en) * | 2019-07-01 | 2019-09-27 | 河南大学 | A kind of detection method that the seam inspection egative film image based on LBP texture is faked |
| CN110969611A (en) * | 2019-12-03 | 2020-04-07 | 广州特种承压设备检测研究院 | Pipeline weld defect detection method, device and system and storage medium |
| CN111141765B (en) * | 2020-01-17 | 2024-06-07 | 上海万物智允工业科技有限责任公司 | Automatic detection and recognition device for weld defects of storage tank |
| CN111141765A (en) * | 2020-01-17 | 2020-05-12 | 中海石油气电集团有限责任公司 | Automatic detection and identification device for weld defects of storage tank |
| CN111340258A (en) * | 2020-03-13 | 2020-06-26 | 新兴铸管股份有限公司 | Cast tube product quality management method based on Internet |
| CN111461569A (en) * | 2020-04-17 | 2020-07-28 | 广东大鹏液化天然气有限公司 | Safety management method for pipeline body |
| CN111461569B (en) * | 2020-04-17 | 2021-02-05 | 广东大鹏液化天然气有限公司 | Safety management method for pipeline body |
| CN111597597A (en) * | 2020-05-11 | 2020-08-28 | 内蒙古京泰发电有限责任公司 | Method for creating PDMS (polydimethylsiloxane) round-the-sky and square-earth structural body equipment model |
| CN111808367B (en) * | 2020-07-09 | 2023-05-30 | 浙江七色鹿色母粒有限公司 | Improvement method for plastic PPR silver grain whitening defect |
| CN111808367A (en) * | 2020-07-09 | 2020-10-23 | 浙江七色鹿色母粒有限公司 | Improvement method for plastic PPR silver grain whitening defect |
| CN113055542A (en) * | 2020-12-15 | 2021-06-29 | 陕西泰诺特检测技术有限公司 | Radiographic negative scanning device and analysis and evaluation method |
| CN114648415A (en) * | 2020-12-18 | 2022-06-21 | 苏州国双软件有限公司 | Method and device for generating pipeline integrity management scheme |
| CN112734693B (en) * | 2020-12-18 | 2024-06-07 | 平安科技（深圳）有限公司 | Pipeline weld defect detection method and related device |
| CN112734693A (en) * | 2020-12-18 | 2021-04-30 | 平安科技（深圳）有限公司 | Pipeline weld defect detection method and related device |
| CN113034442B (en) * | 2021-03-04 | 2023-10-13 | 北京科技大学设计研究院有限公司 | Hot rolled strip steel surface quality grading method based on defect distribution map |
| CN113034442A (en) * | 2021-03-04 | 2021-06-25 | 北京科技大学设计研究院有限公司 | Hot-rolled strip steel surface quality grading method based on defect distribution map |
| CN113516736A (en) * | 2021-04-23 | 2021-10-19 | 扬州能煜检测科技有限公司 | Method and device for establishing and processing discode file mark and electronic equipment |
| CN113516736B (en) * | 2021-04-23 | 2022-12-02 | 江苏迪业检测科技有限公司 | Method and device for establishing and processing dicode file mark and electronic equipment |
| CN113447501A (en) * | 2021-06-28 | 2021-09-28 | 国网黑龙江省电力有限公司电力科学研究院 | Radiographic inspection method for power plant metal pipeline under operating condition |
| CN113192070A (en) * | 2021-06-30 | 2021-07-30 | 深圳市超准视觉科技有限公司 | Weld image processing method and system based on computer vision |
| CN113591055A (en) * | 2021-07-27 | 2021-11-02 | 中国核工业第二二建设有限公司 | Credible weld joint digital inspection method based on block chain technology |
| CN114549836B (en) * | 2022-02-17 | 2024-05-24 | 沈阳派得林科技有限责任公司 | Accurate alignment method for radiographic image and internal detection information of long-distance pipeline |
| CN114549836A (en) * | 2022-02-17 | 2022-05-27 | 沈阳派得林科技有限责任公司 | Method for accurately aligning long-distance pipeline radiographic image with internal detection information |
| CN114820499A (en) * | 2022-04-20 | 2022-07-29 | 上海电气集团股份有限公司 | Pipeline welding seam detection method |
| CN116228703A (en) * | 2023-02-21 | 2023-06-06 | 北京远舢智能科技有限公司 | Defect sample image generation method and device, electronic equipment and medium |
| CN116228703B (en) * | 2023-02-21 | 2024-01-12 | 北京远舢智能科技有限公司 | Defect sample image generation method and device, electronic equipment and medium |
| CN118067726A (en) * | 2024-04-19 | 2024-05-24 | 天津博爱管道科技集团有限公司 | Spiral steel pipe defect detection system based on laser image recognition |
| CN118067726B (en) * | 2024-04-19 | 2024-07-09 | 天津博爱管道科技集团有限公司 | Spiral steel pipe defect detection system based on laser image recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108665452B (en) | 2019-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108665452B (en) | Pipeline-weld film scanning storage and identification of Weld Defects and its system | |
| CN105931227B (en) | A kind of face of weld defect characteristic extracting method for sweeping curve based on gradation of image B | |
| Shafeek et al. | Automatic inspection of gas pipeline welding defects using an expert vision system | |
| US20190331301A1 (en) | Method for leakage detection of underground pipeline corridor based on dynamic infrared thermal image processing | |
| Huang et al. | A pavement crack detection method combining 2D with 3D information based on Dempster‐Shafer theory | |
| Tsai et al. | Critical assessment of pavement distress segmentation methods | |
| CN105069790A (en) | Rapid imaging detection method for gear appearance defect | |
| CN110097547B (en) | Automatic detection method for welding seam negative film counterfeiting based on deep learning | |
| CN103499303A (en) | Wool fineness automatic measuring method | |
| CN109872300A (en) | A kind of vision significance detection method of friction plate open defect | |
| CN112150410A (en) | Automatic detection method and system for weld defects | |
| CN115803619A (en) | Information processing device, determination method, and information processing program | |
| Wang et al. | Automated crack detection for digital radiography aircraft wing inspection | |
| CN101118225A (en) | Method for analyzing welding quality of aluminium alloy with X radiograph | |
| CN105510364A (en) | Nondestructive testing system for industrial part flaws based on X rays and detection method thereof | |
| Yan et al. | Multi-scale multi-intensity defect detection in ray image of weld bead | |
| CN117541483B (en) | Structural performance evaluation method and system for diffusion plate | |
| CN113155839A (en) | Steel plate outer surface defect online detection method based on machine vision | |
| CN111353611A (en) | Automatic generation system and method for in-service inspection and overhaul inspection report of nuclear power station | |
| CN103177426B (en) | Morphology-based strong-interference laser edge image restoration method | |
| Zheng et al. | Design of an advanced automatic inspection system for aircraft parts based on fluorescent penetrant inspection analysis | |
| Xu et al. | Fiber-image analysis part I: Fiber-image enhancement | |
| Ashraf et al. | Machine learning-based pavement crack detection, classification, and characterization: a review | |
| Devereux et al. | Automated object detection for visual inspection of nuclear reactor cores | |
| CN111739036B (en) | Hyperspectrum-based file handwriting counterfeiting detection method |
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 |