CN110223296A - A kind of screw-thread steel detection method of surface flaw and system based on machine vision - Google Patents
A kind of screw-thread steel detection method of surface flaw and system based on machine vision Download PDFInfo
- Publication number
- CN110223296A CN110223296A CN201910610402.6A CN201910610402A CN110223296A CN 110223296 A CN110223296 A CN 110223296A CN 201910610402 A CN201910610402 A CN 201910610402A CN 110223296 A CN110223296 A CN 110223296A
- Authority
- CN
- China
- Prior art keywords
- image
- screw
- thread steel
- detection
- carried out
- 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
Classifications
-
- 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
- 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
- G06T7/0008—Industrial image inspection checking presence/absence
-
- 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
- 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
- G01N2021/8887—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 based on image processing techniques
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20061—Hough transform
-
- 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
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
-
- 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
- G06T2207/30164—Workpiece; Machine component
Abstract
The invention discloses a kind of screw-thread steel Surface Defect Recognition detection method and systems, which comprises obtains screw-thread steel original image and is pre-processed；Image segmentation and cutting are carried out to pretreated original image, obtain screw-thread steel area image；Hough transformation straight-line detection is carried out to the screw-thread steel area image, the differentiation of positive side is carried out according to straight-line detection result；Defects detection is executed to direct picture or side image；The defect identified is marked in original image, and is visualized.The present invention can fast and accurately judge defect present in direct picture, effectively reduce omission factor and cross inspection rate, improve the precision of detection.
Description
Technical field
The invention belongs to technical field of machine vision, and in particular to a kind of screw-thread steel defects detection side based on machine vision
Method and system.
Background technique
Only there is provided background technical informations relevant to the disclosure for the statement of this part, it is not necessary to so constitute first skill
Art.
In screw-thread steel production process, the screw thread steel product that each blank rolls out will be according to national standard or enterprise
Inner quality standard carries out the detection of product lining bar test experiment room, and detection content includes: size, appearance, surface, minus tolerance of lining bar etc.
Deng, detection in general every 15 to 20 minutes or so is primary, once detect the above problem to overrun needs and alarm, and fast
The whole mill rolling force of velocity modulation or replacement rolling equipment.Due to the factors such as production technology, raw material, press device, screw thread steel surface warp
Often will appear scratch, crackle, pit, the defects of pitted skin, folding, scabbing, if cannot the processing of discovery in time, subsequent production can be given
A large amount of waste products are caused, bring massive losses to enterprise.Due to sample point working environment evil, labor intensity, operational hazards, detection effect
Rate is low, apart from factors such as the remote poor in timeliness of production, makes each screw-thread steel manufacturing enterprise, all there is an urgent need to quick, intelligent point automatic
Analysis system cracks this problem.
Artificial detection mainly relies on human eye to detect and identify defect, but overlong time, eyes can generate vision
Fatigue occurs " motion blur sense ", and people can not just differentiate subtle surface defect, will lead to a large amount of erroneous detections, missing inspection occur in this way.
The eddy current testing device developed by French Lorraine tandem rolling company in 1989, by configuring in slab leptoprosopy and upper and lower surface
Vortex finder carries out the detection of the surface crack defect of hot continuous casting steel billet.But the defect kind that the device detects is fewer,
And the defect characteristic parameter extracted also is extremely limited, and is unable to complete the comprehensive eye exam of the surface quality situation of product, and it is only suitable
For occasion of less demanding.2010 Nian Songzhi are strong, Li Zhuxin, and town etc. utilizes flux-leakage detection method combination continuous wavelet transform
Energy method detects Buried Oil Pipelines crack defect.But structure is complicated for leakage magnetic detection device, it is difficult to safeguard, and be easy by ring
The influence of border factor, cannot detect the roughness of belt steel surface, also cannot achieve the correct classification of surface defect, therefore in industry
It is not used widely in production.Infrared detecting device is succeeded in developing by Elkem company, Norway in nineteen ninety earliest, works as presence
When defect, the faradic stroke that radio-frequency induction coil generates in transfer roller just be will increase, and cause billet surface temperature
It increases, to realize detection function.But requirement of the Infrared Detection Method to environment is relatively high, and cannot achieve to defect type
Carry out Accurate classification.
During Machine Vision Detection algorithm research, many scholars propose the defects detection algorithm of oneself.2006
Posco Jong Pil Yun et al. proposes a kind of real-time defects detection algorithm.The detection algorithm mainly proposes
Morphology operations methods is utilized to carry out defects detection.But there are some incomplete places in the utilization of morphologic algorithm.
2007, Choi et al. proposed when handling wire surface defects detection algorithm by edge filter, laplacian filtering and it is double
The thought that three steps of threshold binarization combine.This method can satisfy the requirement of real-time detection, but want to the precision of image
Ask high, more stringent requirements are proposed to Image Acquisition and transmission process.2010, Choi et al. on the basis of original by
Between row image or row image carries out threshold value selection, using the second-order differential value between image compared with the threshold value of setting, just
Step detection image whether there is surface defect.The algorithm calculates simply, and speed is fast, but higher to the quality requirement of image, into
The detection of one step must take more comprehensive method.2015, Wang Qi et al. was become using median filtering and based on discrete cosine
The image enchancing method changed, pre-processes image, the crest mark of break and tooth bottom scratch defects of thread surface is extracted, to doubtful
Defect area seeks the length and width of its minimum circumscribed rectangle, is compared to judge whether it is defect with standard.But the party
Method universality is general, and anti-interference ability is not strong, is unsatisfactory for the step of industrial development.2015, carrys out illuminate et al. and passing through screw-thread steel just
The Combined Treatment in face and side image devises the sub-pix boundary alignment method based on projection center of gravity, designs based on edge
Screw thread steel dimensions computational algorithm, the detection for screw-thread steel outer dimension defect provides reference frame, but is not directed to specific
The detection of screw-thread steel surface defect is studied.Due to screw-thread steel surface defect wide variety, texture is complicated, form is changeable, to defect
The algorithm of detection is more demanding, so referential algorithm and data are relatively fewer at present.
The above method respectively has feature, but universal disadvantage is exactly that system structure is complicated, is unfavorable for safeguarding, therefore not in industry
It is widely used in detection.Screw-thread steel detection method of surface flaw based on machine vision, can using machine vision technique
It by property, the accuracy of detection and improves production efficiency and product quality, it is reliable, accurate, fast to be able to achieve screw-thread steel surface defect
The lossless intelligent measurement of speed is of great significance for mitigating labor intensity of workers, improving production efficiency and product quality.
Summary of the invention
To overcome above-mentioned the deficiencies in the prior art, the screw-thread steel defects detection based on machine vision that the present invention provides a kind of
Method and system carry out slant correction to screw-thread steel region using affine transformation, based on Hough transformation straight-line detection to screw-thread steel
Front, side image distinguish judgement, finally execute defects detection respectively for front, side image, can be fast and accurately
Judge existing scratch, crackle, pit, the defects of pitted skin, folding, scabbing, effectively reduces omission factor and cross inspection rate, improve
The precision of detection.
To achieve the above object, one or more embodiments of the invention provides following technical solution:
A kind of screw-thread steel Surface Defect Recognition detection method, comprising the following steps:
It obtains screw-thread steel original image and is pre-processed；
Image segmentation and cutting are carried out to pretreated original image, obtain screw-thread steel area image；
Hough transformation straight-line detection is carried out to the screw-thread steel area image, positive side is carried out according to straight-line detection result
It distinguishes；
Defects detection is executed to direct picture or side image；
The defect identified is marked in original image, and is visualized.
One or more embodiments provide a kind of screw-thread steel Surface Defect Recognition detection system, comprising:
Data acquisition module obtains screw-thread steel original image and is pre-processed；
Area-of-interest obtains module, carries out image segmentation and cutting to pretreated original image, obtains screw-thread steel
Area image；
Positive side discriminating module carries out Hough transformation straight-line detection to the screw-thread steel area image, according to straight-line detection
As a result the differentiation of positive side is carried out；
Defects detection module executes defects detection to direct picture or side image；
Flaw labeling module marks the defective locations identified and defect classification in original image, and is visualized.
One or more embodiments provide a kind of electronic equipment, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, which is characterized in that the processor is realized described when executing described program
A kind of screw-thread steel Surface Defect Recognition detection method.
One or more embodiments provide a kind of computer readable storage medium, are stored thereon with computer program,
It is characterized in that, a kind of screw-thread steel Surface Defect Recognition detection method is realized when which is executed by processor.
The above one or more technical solution there are following the utility model has the advantages that
The present invention carries out slant correction to screw-thread steel region using affine transformation, while orienting screw-thread steel region,
Judgement is distinguished to screw-thread steel front, side image by Hough transformation straight-line detection, is clapped instead of industrial by camera
According to first time positive bat of the mode that number and the number moved back and forth distinguish a front surface and a side surface of image, such as camera
Screw-thread steel direct picture is taken the photograph, after rotating 180 degree, second time reversed shooting screw-thread steel side image, repeatedly, until rotation
360 degree, judgment method is simplified, is improved work efficiency.
The present invention targetedly carries out defects detection to obtained front, side image respectively, can quick and precisely position and
Defect existing for screw thread steel surface of classifying effectively reduces omission factor and crosses inspection rate, improves the precision of detection.
Detailed description of the invention
The Figure of description for constituting a part of the invention is used to provide further understanding of the present invention, and of the invention shows
Examples and descriptions thereof are used to explain the present invention for meaning property, does not constitute improper limitations of the present invention.
Fig. 1 is the overview flow chart of screw-thread steel Surface Defect Recognition detection method in the one or more embodiments of the present invention；
Fig. 2 is the overhaul flow chart of screw-thread steel direct picture surface defect in the one or more embodiments of the present invention；
Fig. 3 is the overhaul flow chart of screw-thread steel side image surface defect in the one or more embodiments of the present invention.
Specific embodiment
It is noted that described further below be all exemplary, it is intended to provide further instruction to the present invention.Unless another
It indicates, all technical and scientific terms used herein has usual with general technical staff of the technical field of the invention
The identical meanings of understanding.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to exemplary embodiments of the present invention.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
In the absence of conflict, the feature in the embodiment and embodiment in the present invention can be combined with each other.
Embodiment one
Present embodiment discloses a kind of screw-thread steel Surface Defect Recognition detection methods, comprising the following steps:
Step 1 is acquired screw-thread steel original image by high-speed industrial camera.
Step 2, image gray processing carry out enhancing to image and mean filter are handled.It specifically, is 8 using contrast, wide
The low pass mask that degree and height are 15 × 15 enhances the contrast of image, emphasizes the high-frequency region (edge and angle) of image, so that
To image seem apparent, using 5 × 5 mean filter to image carry out the disposal of gentle filter.
Step 3 is partitioned into screw-thread steel entirety region by automatic threshold segmentation method and feature selecting.Specifically, right
Image after the disposal of gentle filter carries out Threshold segmentation according to preset threshold, and preset threshold is by automatic threshold segmentation operator according to flat
Image after sliding filtering processing is arranged automatically.Connected domain is sought, Selection Center row coordinate is located at 84-260 and convexity is located at 0.3-
Screw-thread steel overall region between 1.
Step 4, to original image and through step 3, treated that region is corrected and ROI image extracts, specifically:
4-1: the ranks coordinate of the direction that rear region is handled through step 3 and central point is sought；
Using the shape in shape_trans operator transformation region, the minimum circumscribed rectangle of region any direction is obtained, is sought
The features such as ranks coordinate Row, Column of the direction Phi (radian expression) of minimum circumscribed rectangle, central point.
4-2: slant correction is carried out to original image and through step 3 treated region using affine transformation；
Rigid affine transformation matrix is calculated according to the central point (Row, Column) and direction Phi of minimum circumscribed rectangle, so
Rotation and translation is carried out to image afterwards.The row, column coordinate of the row, column coordinate of origin and transformed point is all minimum circumscribed rectangle
Ranks coordinate Row, Column of central point.If radian Phi < -1.57 (angle=- 90 degree), then the radian of transformed point
For -3.14 (angle=- 180)；If -1.57 < radian Phi < 1.57, then the radian of transformed point is 0 (angle=0)；If
It is radian Phi > 1.57, then the radian of transformed point is 3.14 (angle=180).
Rotation correction operation is carried out to original image first, then the minimum circumscribed rectangle in above-mentioned 4-1 is carried out same
Rotation correction operation.
4-3: using the region after correcting as template, the original image after correction is cut, obtains that there are the ROI of screw-thread steel figures
Picture reduces the interference of substrate processing time and background；
Region similar with minimum circumscribed rectangle is cut out from original image by reduce_doamin operator, as correction
Pretreated image.
4-4: will through step 4-3, treated that ROI image stores, and as pretreated image.
Step 5, for carrying out the judgement of positive side through step 4 treated image, specifically:
5-1: for carrying out side first with the canny operator based on non-maxima suppression through step 4 treated image
Edge detection, improves edge detection accuracy, inhibits false edge；
For the image after step 4 correction process, carried out using 3 × 3 discrete Gaussian function gauss_filter operator
Smooth operation.The side of image is sought using the non-maxima suppression canny filter that filtering parameter is 1.3, threshold value is 15-30
Edge obtains edge amplitude (gradient magnitude) image ImaAmp and edge direction image ImaDir.
5-2: and then using in the Hough transformation hough_lines_dir operator detection edge image of Local gradient direction
Straight line, and with the angle and direction of normal form return straight line.Conllinear point correspondence intersects in parameter space in image space
Line, all straight lines that the same point is intersected in parameter space have conllinear point to be corresponding to it in image space；
Using in the Hough transformation hough_lines_dir operator detection edge image ImaDir of Local gradient direction
Line, the threshold value that the number of collinear points is wherein formed in Hough image is 67.Using gen_region_hline operator by Hesse
Input line described in regular shape is stored as region.
5-3: conllinear straight according to what be can be detected in the collinear lines and side image detected in screw-thread steel direct picture
Line position range difference distinguishes front, side image.Such as the straight line detected in direct picture, slightly towards in image
Lower edges two sides, and the straight line detected in side image is slightly towards picture centre part.Specifically, area_ is utilized
Center operator obtains the ranks coordinate of the central point of straight line, if linear rows coordinate between 0-30 or 160-190, original image
As being cross rib direct picture facing forward, if the row coordinate of straight line, between 50-145, original image is side rib side facing forward
Image.
Step 6, the direct picture distinguished for above-mentioned steps 5 carry out area dividing defects detection, specifically:
6-1: for direct picture, progress sub-pixel edge detection first selects edge based on profile length, and connection is close
Like conllinear contour line, lower edges region is found out, generates its minimum circumscribed rectangle respectively, it will be upper following as template
Edge, which is individually cut out from original image, to be come；
The accurate lower edges of sub-pixel are extracted using the Canny filter in edges_sub_pix operator, are utilized
It is total that union_collinear_contours_xld operator connects the approximation that the maximal clearance length between two contours is 30
Line profile selects edge contour of the contour line total length between 123-1000 using select_shape_xld operator.Point
Not Sheng Cheng where lower edges at minimum circumscribed rectangle, region identical with minimum circumscribed rectangle is cut out from original image, is obtained
To area image ImageA, the ImageB for only including lower edges.
6-2: global threshold processing is carried out respectively for the upper and lower longitudinal rib image of said extracted, with a Rectangle structure cell
Element carries out morphology to it and opens operation；
Using area-of-interest in threshold operator extraction image ImagA, ImageB, using the Rectangle structure cell of 3x3
Element carries out out operation.
6-3: the convexity feature based on region carries out feature selecting, finds out defect region；
Extract connected domain, using select_shape operator select convexity be 0.51369-0.60447, area 0-357
Between region, such region if it exists, then existing defects in lower edges image, if it does not exist, then lower edges image
In be not present defect.
6-4: differences operator is utilized, the cross rib region in addition to lower edges region is extracted from original image；Benefit
It is cut out with crop_domain operator to obtain ImageC.
6-5: for above-mentioned intermediate cross rib part, the position where each cross rib is searched out first with template matching method,
Obtain the features such as center point coordinate, angle, the radius of each cross rib；
Single flawless cross rib image is found first as template, utilizes create_shape_model operator creation one
A shape, get_shape_model_contours operator return to the profile expression of shape, find_shape_
Model operator finds out the best match of shape in image ImageC, obtains its relevant position parameter.
6-6: one is generated at the cross rib that each is matched to and is subject to central point (row, column), direction and long axis
Radius1, minor axis radius Radius2 are slightly larger than the elliptic region of cross rib radius, can partly overlap between adjacent area, step master
If preventing from omitting existing defects region.Cross rib region in original image is cut as template, extracts all cross
Flank carries out defects detection positioning respectively；
For the single screw-thread steel region being matched to, it is first based on its central point, direction value, generates a long axis, short axle half
Diameter is slightly larger than the elliptic region of original template, and it is cut out to come from cross rib image ImageC, does and handles in next step.
6-7: for each the cross rib region extracted, edge detection is carried out first with canny operator, to its side
Edge amplitude (gradient magnitude) image carries out Threshold segmentation, extracts connected component；
For the image that above-mentioned 6-6 is extracted, the non-maxima suppression that filtering parameter is 1.5, threshold value is 10-30 is used
Canny filter seeks the edge of single cross rib image, obtains edge amplitude (gradient magnitude) image ImaAmp1, shakes to edge
Width image carries out the operation of threshold Threshold segmentation, extracts connected domain.
6-8: feature selecting is carried out with circularity according to area, finds out defect region.
To connected domain carry out feature selecting, select the region that area is 12-30, circularity is 0.1-0.5, if it exists this
The region of sample, then existing defects in cross rib image, are otherwise not present defect.
Step 7: the side image distinguished for above-mentioned steps 5 carries out defects detection.Because the texture of background is not
May be identical with the texture of present image, the present embodiment is handled by the way that image is transformed to frequency domain, extracts defect point
Spatial domain is changed in contravariant after amount, and the specific location of defect is obtained by operations such as Threshold segmentations.Specifically:
7-1: side image gray processing；
7-2: frequency range locating for defect and background and noise have apparent difference in surface to be detected, pass through band
The frequency content obtained after resistance filtering has apparent inhibition, and prominent defect ingredient to the texture in background.Image is carried out
Fourier transformation is transformed into domain space, removes the interference of cyclical signal；The Gauss of two ' FFT ' modes is generated in frequency domain
Low-pass filter constructs a bandstop filter to extract defect component after being subtracted each other；With bandstop filter to side image
Carry out convolution；Fourier inversion is carried out to spatial domain to image；
Low frequency is mostly that background in image etc. is smoothly located in surface to be detected, and high frequency treatment is mostly the variation such as edge or noise
There are inhibiting effect, and prominent defect in violent place to the texture in background by the frequency content obtained after bandreject filtering
Ingredient.Fourier transformation is carried out to image, domain space is transformed into, removes the interference of cyclical signal；' FFT ' is generated in frequency domain
Mode, the Gaussian filter 1 of Sigma1=10, Sigm2=10 and ' FFT ' mode, the Gauss of Sigma1=3, Sigma2=3 filter
Wave device 2, then filter 1 subtracts filter 2, obtains bandstop filter；Convolution is carried out to side image with bandstop filter, it is right
Image carries out Fourier inversion to spatial domain, enhances the contrast of defect and other regions；
7-3: calculating above-mentioned spatial domain gray value of image range, i.e., using 10 × 10 rectangle as template, traverses input picture,
The difference (max-min) of the maximum gradation value and minimum gradation value of image in each rectangular mask is sought, each difference is constituted
One picture point, all picture points are constituted piece image, as a result return in the form of images, enhanced pair between texture with this
Than degree, facilitate Threshold segmentation.If parameter mask height or mask width are even number, they are changed to next lesser
Singular value；In the boundary of image, gray value is mirrored.
7-4: determine that result described in 7-3 generates the minimum gray value Min and maximum gradation value Max of image；
7-5: using selected threshold value T compared with maximum gradation value Max size come threshold value, to 7-3 described image into
Row segmentation；Select region of the gray value between (max ([T, Max*0.3]), 255) in generation image described in 7-3；
7-6: connected region processing is extracted defect area based on area and circularity feature.
Step 8, the defect area found for each step obtain the area in region and the ranks coordinate of central point first,
Again from Area generation XLD sub-pix profile, with circle approximation XLD profile, and mark in original image sunken region of falling vacant area,
The information such as position.
The present embodiment is the defect inspection method based on Halcon vision algorithm, and defects detection is that view is write by Halcon
Feel algorithm process image, statistical shortcomings information realization being automatically brought into operation of machine improves detection efficiency, reduces manpower waste, can
Realization persistently detects work.
Embodiment two
The purpose of the present embodiment is to provide a kind of screw-thread steel Surface Defect Recognition detection system.
To achieve the goals above, a kind of screw-thread steel Surface Defect Recognition detection system is present embodiments provided, comprising:
Data acquisition module obtains screw-thread steel original image and is pre-processed；
Area-of-interest obtains module, carries out image segmentation and cutting to pretreated original image, obtains screw-thread steel
Area image；
Positive side discriminating module carries out Hough transformation straight-line detection to the screw-thread steel area image, according to straight-line detection
As a result the differentiation of positive side is carried out；
Defects detection module executes defects detection to direct picture or side image；
Flaw labeling module marks the defect identified in original image, and is visualized.
Embodiment three
The purpose of the present embodiment is to provide a kind of electronic equipment.
A kind of electronic equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, the processor realize following steps when executing described program, comprising:
It obtains screw-thread steel original image and is pre-processed；
Image segmentation and cutting are carried out to pretreated original image, obtain screw-thread steel area image；
Hough transformation straight-line detection is carried out to the screw-thread steel area image, positive side is carried out according to straight-line detection result
It distinguishes；
Defects detection is executed to direct picture or side image；
The defect identified is marked in original image, and is visualized.
Example IV
The purpose of the present embodiment is to provide a kind of computer readable storage medium.
A kind of computer readable storage medium, is stored thereon with computer program, realization when which is executed by processor
Following steps:
It obtains screw-thread steel original image and is pre-processed；
Image segmentation and cutting are carried out to pretreated original image, obtain screw-thread steel area image；
Hough transformation straight-line detection is carried out to the screw-thread steel area image, positive side is carried out according to straight-line detection result
It distinguishes；
Defects detection is executed to direct picture or side image；
The defect identified is marked in original image, and is visualized.
Each step involved in above embodiments two, three and four is corresponding with embodiment of the method one, and specific embodiment can
Referring to the related description part of embodiment one.Term " computer readable storage medium " is construed as including that one or more refers to
Enable the single medium or multiple media of collection；It should also be understood as including any medium, any medium can be stored, be encoded
Or it carries instruction set for being executed by processor and processor is made either to execute in the present invention method.
The above one or more embodiment has following technical effect that
The present invention carries out slant correction to screw-thread steel region using affine transformation, while orienting screw-thread steel region.
Subregion judgement is carried out to screw-thread steel direct picture by Hough transformation straight-line detection, is taken pictures instead of industrially by camera secondary
The mode that number and the number moved back and forth distinguish a front surface and a side surface of image, such as first time positive shooting spiral shell of camera
Line steel direct picture, after rotating 180 degree, second time reversed shooting screw-thread steel side image, repeatedly, until rotation 360
Degree, simplifies judgment method, improves work efficiency.
The present invention is to obtained top edge defects detection image, lower edge defects detection image, central defect detection image
Judged, can fast and accurately judge defect present in direct picture, effectively reduce omission factor and cross inspection rate, is improved
The precision of detection.
The invention proposes a kind of detection methods for screw-thread steel surface defect, and structure is very simple, has expanded screw thread
The use scope of steel identification, improves the surface quality of screw-thread steel finished product, promotes the hair of screw-thread steel defects detection in the industrial production
Exhibition and application, meet the needs of user in practice.
The present invention finds the position of single cross rib, then successively carries out by carrying out template matching to whole cross rib part
The detection of defect judges, solve the problems, such as screw-thread steel because overall gray value close to due to divide hardly possible；To dividing for cross rib part
Block detection, also improves the accuracy and speed of defects detection.
The present invention uses two low-pass filters, and bandstop filter is constructed after being subtracted each other and extracts defect component,
Have apparent inhibition to the texture in background by the frequency content obtained after bandreject filtering, enhance in surface to be detected defect and
The difference of background and noise；Reconstruct obtains defect image after carrying out Fourier inversion, obtains defect using threshold operation
Specific location.
It will be understood by those skilled in the art that each module or each step of aforementioned present invention can be filled with general computer
It sets to realize, optionally, they can be realized with the program code that computing device can perform, it is thus possible to which they are stored
Be performed by computing device in the storage device, perhaps they are fabricated to each integrated circuit modules or by they
In multiple modules or step be fabricated to single integrated circuit module to realize.The present invention is not limited to any specific hardware and
The combination of software.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Above-mentioned, although the foregoing specific embodiments of the present invention is described with reference to the accompanying drawings, not protects model to the present invention
The limitation enclosed, those skilled in the art should understand that, based on the technical solutions of the present invention, those skilled in the art are not
Need to make the creative labor the various modifications or changes that can be made still within protection scope of the present invention.
Claims (10)
1. a kind of screw-thread steel Surface Defect Recognition detection method, which comprises the following steps:
It obtains screw-thread steel original image and is pre-processed；
Image segmentation and cutting are carried out to pretreated original image, obtain screw-thread steel area image；
Hough transformation straight-line detection is carried out to the screw-thread steel area image, the area of positive side is carried out according to straight-line detection result
Point；
Defects detection is executed to direct picture or side image；
The defect identified is marked in original image, and is visualized.
2. a kind of screw-thread steel Surface Defect Recognition detection method as described in claim 1, which is characterized in that the pretreatment packet
It includes: picture contrast is enhanced using low pass mask；The disposal of gentle filter is carried out to image using mean filter.
3. a kind of screw-thread steel Surface Defect Recognition detection method as described in claim 1, which is characterized in that pretreated original
Beginning image carries out image segmentation and cutting includes:
It selects to obtain screw-thread steel entirety region based on automatic threshold segmentation and connected domain；
Seek the direction of the minimum circumscribed rectangle in the region and the ranks coordinate of central point；
Rigid affine transformation matrix is calculated according to the direction of minimum circumscribed rectangle and center position；
Rotation correction is carried out to original image and the minimum circumscribed rectangle based on the radiation transformation matrix；
Based on the minimum circumscribed rectangle after correction, original image after clipping correction obtains screw-thread steel area image.
4. a kind of screw-thread steel Surface Defect Recognition detection method as described in claim 1, which is characterized in that carry out sentencing for positive side
It is disconnected to include:
For screw-thread steel area image, edge detection is carried out；
Based on the straight line in Hough transformation operator detection edge image；
It is different with collinear lines position range detectable in side image according to screw-thread steel direct picture, to front, side
Image distinguishes.
5. a kind of screw-thread steel Surface Defect Recognition detection method as described in claim 1, which is characterized in that executed to direct picture
Defects detection includes:
For direct picture, sub-pixel edge detection is carried out, upper and lower edge region and intermediate region are obtained；
For upper and lower edge region, its minimum circumscribed rectangle is generated respectively, and original image is cut based on minimum circumscribed rectangle
Obtain upper and lower longitudinal rib area image；
For upper and lower longitudinal rib area image, global threshold processing is carried out respectively, and morphology is carried out using rectangular configuration element and opens behaviour
Make, defect region is determined according to the convexity feature in region；
For intermediate cross rib area image, each cross rib position is searched using template matching method, and obtain each cross rib
Center point coordinate, angle and radius；Generation one is consistent with the cross rib central point and direction at the cross rib that each is matched to,
And long axis, short axle are greater than the elliptic region of cross rib radius, adjacent ellipses are interregional to partly overlap, and is cut based on these elliptic regions
Original image obtains the corresponding cross rib area image of each cross rib；
For each cross rib area image, edge detection is carried out first with canny operator, threshold is carried out to its edge amplitude image
Value segmentation, extracts connected component, determines defect region with circularity according to area.
6. a kind of screw-thread steel Surface Defect Recognition detection method as described in claim 1, which is characterized in that executed to side image
Defects detection includes:
Fourier transformation is carried out to side image, is transformed into domain space；
Bandstop filter is constructed, convolution is carried out to side image, Fourier inversion is carried out after convolution, is transformed into spatial domain；
Space area image degree of comparing is enhanced based on mask, then executes Threshold segmentation；
Area and circularity feature based on connected region extract defect area.
7. a kind of screw-thread steel Surface Defect Recognition detection method as described in claim 1, which is characterized in that get the bid in original image
Remember that the defect identified includes:
For each defect area, the area in region and the ranks coordinate of central point are obtained；
From Area generation sub-pix profile, and using the approximate sub-pix profile of circle；
The area of marking of defects region and position in original image.
8. a kind of screw-thread steel Surface Defect Recognition detection system characterized by comprising
Data acquisition module obtains screw-thread steel original image and is pre-processed；
Area-of-interest obtains module, carries out image segmentation and cutting to pretreated original image, obtains screw-thread steel region
Image；
Positive side discriminating module carries out Hough transformation straight-line detection to the screw-thread steel area image, according to straight-line detection result
Carry out the differentiation of positive side；
Defects detection module executes defects detection to direct picture or side image；
Flaw labeling module marks the defect identified in original image, and is visualized.
9. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes a kind of such as the described in any item spiral shells of claim 1-7 when executing described program
Line steel surface defect recognition detection method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
A kind of such as claim 1-7 described in any item screw-thread steel Surface Defect Recognition detection methods are realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910610402.6A CN110223296B (en) | 2019-07-08 | 2019-07-08 | Deformed steel bar surface defect detection method and system based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910610402.6A CN110223296B (en) | 2019-07-08 | 2019-07-08 | Deformed steel bar surface defect detection method and system based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110223296A true CN110223296A (en) | 2019-09-10 |
| CN110223296B CN110223296B (en) | 2021-06-11 |
Family
ID=67812250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910610402.6A Active CN110223296B (en) | 2019-07-08 | 2019-07-08 | Deformed steel bar surface defect detection method and system based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110223296B (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717902A (en) * | 2019-09-29 | 2020-01-21 | 中山市瑞福达触控显示技术有限公司 | Processing method for displaying image edge |
| CN110991667A (en) * | 2019-11-28 | 2020-04-10 | 中国铁道科学研究院集团有限公司 | Railway track facility abnormity identification method and system |
| CN111028215A (en) * | 2019-12-06 | 2020-04-17 | 上海大学 | Method for detecting end surface defects of steel coil based on machine vision |
| CN111047556A (en) * | 2019-11-13 | 2020-04-21 | 广智微芯(扬州)有限公司 | Strip steel surface defect detection method and device |
| CN111598831A (en) * | 2020-03-26 | 2020-08-28 | 浙江科技学院 | Detection method and system for cylindrical roller bearing |
| CN111855672A (en) * | 2020-07-29 | 2020-10-30 | 佛山市南海区广工大数控装备协同创新研究院 | Method for detecting COF flexible board defects |
| CN112164052A (en) * | 2020-09-30 | 2021-01-01 | 西南交通大学 | Railway sleeper defect detection method based on terahertz imaging |
| CN112378338A (en) * | 2020-11-25 | 2021-02-19 | 上海里莱技术服务中心 | External thread detection method based on Hough transformation |
| CN112629407A (en) * | 2020-11-24 | 2021-04-09 | 西安理工大学 | Deformed steel bar size measuring method based on image analysis |
| CN112834457A (en) * | 2021-01-23 | 2021-05-25 | 中北大学 | Metal microcrack three-dimensional characterization system and method based on reflective laser thermal imaging |
| CN112881404A (en) * | 2021-01-12 | 2021-06-01 | 王玮佳 | Deformed steel bar appearance defect detection device |
| CN112884694A (en) * | 2019-11-13 | 2021-06-01 | 合肥欣奕华智能机器有限公司 | Defect detection method, device, equipment and medium for flat display panel |
| CN112991251A (en) * | 2019-11-29 | 2021-06-18 | 合肥欣奕华智能机器有限公司 | Method, device and equipment for detecting surface defects |
| CN113327257A (en) * | 2021-06-02 | 2021-08-31 | 蚌埠凯盛工程技术有限公司 | Method and device for judging automobile glass in different areas |
| CN113505811A (en) * | 2021-06-10 | 2021-10-15 | 常州理工科技股份有限公司 | Machine vision imaging method for hub production |
| CN113744194A (en) * | 2021-08-04 | 2021-12-03 | 武汉钢铁有限公司 | Steel plate defect detection method and computer readable storage medium |
| CN114354491A (en) * | 2021-12-30 | 2022-04-15 | 苏州精创光学仪器有限公司 | DCB ceramic substrate defect detection method based on machine vision |
| CN115375693A (en) * | 2022-10-27 | 2022-11-22 | 浙江托普云农科技股份有限公司 | Method, system and device for detecting defects of probe of agricultural information acquisition sensor |
| TWI792351B (en) * | 2021-06-15 | 2023-02-11 | 鴻海精密工業股份有限公司 | Method for detecting defects, electronic device and storage medium |
| CN116452589A (en) * | 2023-06-16 | 2023-07-18 | 山东伟国板业科技有限公司 | Intelligent detection method for surface defects of artificial board based on image processing |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010164318A (en) * | 2009-01-13 | 2010-07-29 | Nippon Steel Corp | Method and device for inspecting flaw |
| US8615125B2 (en) * | 2010-10-08 | 2013-12-24 | Omron Corporation | Apparatus and method for inspecting surface state |
| CN104121853A (en) * | 2014-07-15 | 2014-10-29 | 深圳市大族激光科技股份有限公司 | Casing detection device and objective table thereof |
| CN105631414A (en) * | 2015-12-23 | 2016-06-01 | 上海理工大学 | Vehicle-borne multi-obstacle classification device and method based on Bayes classifier |
| US20160293502A1 (en) * | 2015-03-31 | 2016-10-06 | Lam Research Corporation | Method and apparatus for detecting defects on wafers |
| CN106011672A (en) * | 2016-07-29 | 2016-10-12 | 常州东大中天钢铁研究院有限公司 | Philippine-standard straight screw-thread steel and low-cost production technology thereof |
| CN106053479A (en) * | 2016-07-21 | 2016-10-26 | 湘潭大学 | System for visually detecting workpiece appearance defects based on image processing |
| CN106204614A (en) * | 2016-07-21 | 2016-12-07 | 湘潭大学 | A kind of workpiece appearance defects detection method based on machine vision |
| CN107478657A (en) * | 2017-06-20 | 2017-12-15 | 广东工业大学 | Stainless steel surfaces defect inspection method based on machine vision |
| CN107507206A (en) * | 2017-06-09 | 2017-12-22 | 合肥工业大学 | A kind of depth map extracting method based on conspicuousness detection |
| CN107610104A (en) * | 2017-08-31 | 2018-01-19 | 南通兴华达高实业有限公司 | Crack detecting method at a kind of elevator compensation chain R based on machine vision |
| CN108548820A (en) * | 2018-03-28 | 2018-09-18 | 浙江理工大学 | Cosmetics paper labels defect inspection method |
| CN108560345A (en) * | 2018-05-04 | 2018-09-21 | 刘刚 | Road construction pretreatment unit and its construction method |
| CN109001212A (en) * | 2018-07-17 | 2018-12-14 | 五邑大学 | A kind of stainless steel soup ladle defect inspection method based on machine vision |
| CN109118471A (en) * | 2018-06-26 | 2019-01-01 | 广东工业大学 | A kind of polishing workpiece, defect detection method suitable under complex environment |
| CN109827515A (en) * | 2018-12-28 | 2019-05-31 | 甘肃第一建设集团有限责任公司 | A kind of the screw steel wire area of bed detection system and method for separate type |
-
2019
- 2019-07-08 CN CN201910610402.6A patent/CN110223296B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010164318A (en) * | 2009-01-13 | 2010-07-29 | Nippon Steel Corp | Method and device for inspecting flaw |
| US8615125B2 (en) * | 2010-10-08 | 2013-12-24 | Omron Corporation | Apparatus and method for inspecting surface state |
| CN104121853A (en) * | 2014-07-15 | 2014-10-29 | 深圳市大族激光科技股份有限公司 | Casing detection device and objective table thereof |
| US20160293502A1 (en) * | 2015-03-31 | 2016-10-06 | Lam Research Corporation | Method and apparatus for detecting defects on wafers |
| CN105631414A (en) * | 2015-12-23 | 2016-06-01 | 上海理工大学 | Vehicle-borne multi-obstacle classification device and method based on Bayes classifier |
| CN106053479A (en) * | 2016-07-21 | 2016-10-26 | 湘潭大学 | System for visually detecting workpiece appearance defects based on image processing |
| CN106204614A (en) * | 2016-07-21 | 2016-12-07 | 湘潭大学 | A kind of workpiece appearance defects detection method based on machine vision |
| CN106011672A (en) * | 2016-07-29 | 2016-10-12 | 常州东大中天钢铁研究院有限公司 | Philippine-standard straight screw-thread steel and low-cost production technology thereof |
| CN107507206A (en) * | 2017-06-09 | 2017-12-22 | 合肥工业大学 | A kind of depth map extracting method based on conspicuousness detection |
| CN107478657A (en) * | 2017-06-20 | 2017-12-15 | 广东工业大学 | Stainless steel surfaces defect inspection method based on machine vision |
| CN107610104A (en) * | 2017-08-31 | 2018-01-19 | 南通兴华达高实业有限公司 | Crack detecting method at a kind of elevator compensation chain R based on machine vision |
| CN108548820A (en) * | 2018-03-28 | 2018-09-18 | 浙江理工大学 | Cosmetics paper labels defect inspection method |
| CN108560345A (en) * | 2018-05-04 | 2018-09-21 | 刘刚 | Road construction pretreatment unit and its construction method |
| CN109118471A (en) * | 2018-06-26 | 2019-01-01 | 广东工业大学 | A kind of polishing workpiece, defect detection method suitable under complex environment |
| CN109001212A (en) * | 2018-07-17 | 2018-12-14 | 五邑大学 | A kind of stainless steel soup ladle defect inspection method based on machine vision |
| CN109827515A (en) * | 2018-12-28 | 2019-05-31 | 甘肃第一建设集团有限责任公司 | A kind of the screw steel wire area of bed detection system and method for separate type |
Non-Patent Citations (4)
| Title |
|---|
| JING YANG等: "Research On Non-destructive Testing For Defects Of Steel Bar In Concrete", 《APPLIED MECHANICS AND MATERIALS》 * |
| NIRBHAR NEOGI等: "Review of vision-based steel surface inspection systems", 《EURASIP JOURNAL ON IMAGE AND VIDEO PROCESSING》 * |
| 来煜: "螺纹钢表面缺陷检测算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 田利等: "机器视觉在半导体器件塑封缺陷检测中的应用", 《电脑知识与技术》 * |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717902B (en) * | 2019-09-29 | 2023-06-09 | 中山市瑞福达触控显示技术有限公司 | Processing method for display image edge |
| CN110717902A (en) * | 2019-09-29 | 2020-01-21 | 中山市瑞福达触控显示技术有限公司 | Processing method for displaying image edge |
| CN111047556A (en) * | 2019-11-13 | 2020-04-21 | 广智微芯(扬州)有限公司 | Strip steel surface defect detection method and device |
| CN112884694B (en) * | 2019-11-13 | 2023-12-22 | 合肥欣奕华智能机器股份有限公司 | Defect detection method, device, equipment and medium for flat display panel |
| CN111047556B (en) * | 2019-11-13 | 2024-04-05 | 广智微芯(扬州)有限公司 | Strip steel surface defect detection method and device |
| CN112884694A (en) * | 2019-11-13 | 2021-06-01 | 合肥欣奕华智能机器有限公司 | Defect detection method, device, equipment and medium for flat display panel |
| CN110991667A (en) * | 2019-11-28 | 2020-04-10 | 中国铁道科学研究院集团有限公司 | Railway track facility abnormity identification method and system |
| CN112991251A (en) * | 2019-11-29 | 2021-06-18 | 合肥欣奕华智能机器有限公司 | Method, device and equipment for detecting surface defects |
| CN112991251B (en) * | 2019-11-29 | 2023-01-17 | 合肥欣奕华智能机器股份有限公司 | Method, device and equipment for detecting surface defects |
| CN111028215A (en) * | 2019-12-06 | 2020-04-17 | 上海大学 | Method for detecting end surface defects of steel coil based on machine vision |
| CN111598831A (en) * | 2020-03-26 | 2020-08-28 | 浙江科技学院 | Detection method and system for cylindrical roller bearing |
| CN111855672A (en) * | 2020-07-29 | 2020-10-30 | 佛山市南海区广工大数控装备协同创新研究院 | Method for detecting COF flexible board defects |
| CN112164052B (en) * | 2020-09-30 | 2021-10-15 | 西南交通大学 | Railway sleeper defect detection method based on terahertz imaging |
| CN112164052A (en) * | 2020-09-30 | 2021-01-01 | 西南交通大学 | Railway sleeper defect detection method based on terahertz imaging |
| CN112629407A (en) * | 2020-11-24 | 2021-04-09 | 西安理工大学 | Deformed steel bar size measuring method based on image analysis |
| CN112629407B (en) * | 2020-11-24 | 2024-03-22 | 西安理工大学 | Deformed steel bar dimension measuring method based on image analysis |
| CN112378338A (en) * | 2020-11-25 | 2021-02-19 | 上海里莱技术服务中心 | External thread detection method based on Hough transformation |
| CN112881404A (en) * | 2021-01-12 | 2021-06-01 | 王玮佳 | Deformed steel bar appearance defect detection device |
| CN112834457A (en) * | 2021-01-23 | 2021-05-25 | 中北大学 | Metal microcrack three-dimensional characterization system and method based on reflective laser thermal imaging |
| CN113327257A (en) * | 2021-06-02 | 2021-08-31 | 蚌埠凯盛工程技术有限公司 | Method and device for judging automobile glass in different areas |
| CN113327257B (en) * | 2021-06-02 | 2022-07-26 | 蚌埠凯盛工程技术有限公司 | Method and device for judging automobile glass in different areas |
| CN113505811A (en) * | 2021-06-10 | 2021-10-15 | 常州理工科技股份有限公司 | Machine vision imaging method for hub production |
| TWI792351B (en) * | 2021-06-15 | 2023-02-11 | 鴻海精密工業股份有限公司 | Method for detecting defects, electronic device and storage medium |
| CN113744194B (en) * | 2021-08-04 | 2024-03-19 | 武汉钢铁有限公司 | Steel plate defect detection method and computer readable storage medium |
| CN113744194A (en) * | 2021-08-04 | 2021-12-03 | 武汉钢铁有限公司 | Steel plate defect detection method and computer readable storage medium |
| CN114354491A (en) * | 2021-12-30 | 2022-04-15 | 苏州精创光学仪器有限公司 | DCB ceramic substrate defect detection method based on machine vision |
| CN115375693A (en) * | 2022-10-27 | 2022-11-22 | 浙江托普云农科技股份有限公司 | Method, system and device for detecting defects of probe of agricultural information acquisition sensor |
| CN116452589A (en) * | 2023-06-16 | 2023-07-18 | 山东伟国板业科技有限公司 | Intelligent detection method for surface defects of artificial board based on image processing |
| CN116452589B (en) * | 2023-06-16 | 2023-10-20 | 山东伟国板业科技有限公司 | Intelligent detection method for surface defects of artificial board based on image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110223296B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110223296A (en) | A kind of screw-thread steel detection method of surface flaw and system based on machine vision | |
| Luo et al. | A cost-effective and automatic surface defect inspection system for hot-rolled flat steel | |
| Yang et al. | Defect detection in magnetic tile images based on stationary wavelet transform | |
| CN107389701A (en) | A kind of PCB visual defects automatic checkout system and method based on image | |
| CN105335973B (en) | Apply to the visual processing method of strip machining production line | |
| CN104732536A (en) | Sub-pixel edge detection method based on improved morphology | |
| CN116228747B (en) | Metal cabinet processing quality monitoring method | |
| CN108171102A (en) | A kind of part method for quickly identifying of view-based access control model | |
| CN110021024A (en) | A kind of image partition method based on LBP and chain code technology | |
| CN109584215A (en) | A kind of online vision detection system of circuit board | |
| CN107240086B (en) | A kind of fabric defects detection method based on integral nomography | |
| Shi et al. | A faster-rcnn based chemical fiber paper tube defect detection method | |
| Yang et al. | Vial bottle mouth defect detection based on machine vision | |
| Ji et al. | Features extraction and classification of wood defect based on HU invariant moment and wavelet moment and BP neural network | |
| Li et al. | Weld image recognition algorithm based on deep learning | |
| Yao | Image segmentation based on Sobel edge detection | |
| Dong et al. | A rapid detection method for the surface defects of mosaic ceramic tiles | |
| CN112381140B (en) | Abrasive particle image machine learning identification method based on new characteristic parameters | |
| Zhang et al. | An improved Harris corner detection algorithm for noised images | |
| CN113436102A (en) | Method for detecting surface defects of machined part | |
| CN106504251B (en) | A kind of electric aluminum foil cubic texture detection method of content based on picture processing | |
| Yu et al. | Surface defect detection of hight-speed railway hub based on improved YOLOv3 algorithm | |
| CN114494225A (en) | Paper tube breakage defect detection method based on shape characteristics | |
| Zhong et al. | Iris location algorithm based on the CANNY operator and gradient Hough transform | |
| Kong et al. | Scratch detection of round buttons based on machine 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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230818 Address after: Block A, Building 2, Intelligent Transportation Industrial Park, No. 11777, Tourist Road, Jinan Area, China (Shandong) Pilot Free Trade Zone, Jinan City, Shandong Province, 250000 Patentee after: SHANDONG ZHENGCHEN POLYTRON TECHNOLOGIES Co.,Ltd. Address before: 250101 1000, Feng Ming Road, Lingang Development Zone, Licheng District, Ji'nan, Shandong Patentee before: SHANDONG JIANZHU University |