CN108460763A - A kind of automatic detection recognition method of magnetic powder inspection image - Google Patents
A kind of automatic detection recognition method of magnetic powder inspection image Download PDFInfo
- Publication number
- CN108460763A CN108460763A CN201810254254.4A CN201810254254A CN108460763A CN 108460763 A CN108460763 A CN 108460763A CN 201810254254 A CN201810254254 A CN 201810254254A CN 108460763 A CN108460763 A CN 108460763A
- Authority
- CN
- China
- Prior art keywords
- characteristic parameter
- particle
- image
- measurand
- camera
- 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
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/82—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws
- G01N27/83—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws by investigating stray magnetic fields
- G01N27/84—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables for investigating the presence of flaws by investigating stray magnetic fields by applying magnetic powder or magnetic ink
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10141—Special mode during image acquisition
- G06T2207/10152—Varying illumination
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Electrochemistry (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of automatic detection recognition methods of magnetic powder inspection image, include the following steps:Measurand is irradiated using the ultraviolet source of incidence angles degree, the multiple image under different illumination conditions is shot with camera and camera lens, Pixel-level rate of gray level calculating is carried out to image, obtains binary image.White particle in binary image is filtered by size, and calculates the score of remainder particulate characteristic parameter;The score of different characteristic parameter is summed, the characteristic parameter comprehensive score as the particle;The characteristic parameter comprehensive score of particle is compared with characteristic parameter threshold, characteristic parameter comprehensive score is judged as defect, i.e. crackle in measurand less than the particle of threshold value.The present invention proposes that calculating speed is fast based on three morphological feature nonlinearity in parameters blending algorithms, and stability is high, helps to improve automation and the intelligent level of fluorescent magnetic particle flaw detection, can be widely used in magnetic metal part manufacturing field.
Description
Technical field
The present invention relates to surface of workpiece defect estimation fields more particularly to a kind of magnetic powder inspection image to detect knowledge automatically
Other method.
Background technology
The supersonic detector more expensive compared to price, fluorescentmagnetic particle(powder) crack detection because of at low cost, high sensitivity and
The advantages that detection speed is fast and be widely used.It is real that traditional fluorescentmagnetic particle(powder) surface crack testing relies primarily on artificial cognition
It is existing, it is less efficient and judge that accuracy is not high, or occur flase drop because of visual fatigue, while human body is for a long time in the ultraviolet ring of light
It works under border and easily endangers health.Therefore, fluorescentmagnetic particle(powder) crackle automated detection method has important theory significance and reality
With value.
Retrieval finds that the representative achievement in the field includes at present:
1) paper《Cylindrical workpiece crack automatic detection method based on magnetic powder inspection》It proposes a kind of with machine vision
The cylindrical work external crack automatic testing method of detection technique, the method are to utilize industrial camera and plane black light lamp
Workpiece cylinder magnetic trace image is shot, the feature of doubtful slit region is extracted with digital image processing techniques, then utilizes mould
Classifier technique in formula identification completes the automatic identification of bearing parts cylinder crackle.Since magnetic flaw detection ink recycles the miscellaneous of accumulation
Matter and foam, column surface roughness differ and the reasons such as cylindrical reflection is stronger, cause magnetic trace image segmentation bad.
2) paper《The development of automobile brake disc full-automatic vision magnaflux》Automobile brake disc is devised automatically to regard
Feel magnaflux, improvement is optimized to magnetizing assembly and conveying robot, it is proposed that combined type Threshold Analysis algorithm comes
It distinguish true from false defect.The method has certain reference value in terms of magnetic powder automatic inspection equipment structure optimization, but still does not have
Solve the problems, such as that magnetic powder inspection False Rate is high.
3) utility model patent《Intelligent magnetic particle flaw detector》(publication number:CN105866242A it) devises and adds magnetic automatically and move back
The Mechatronic Systems of magnetic, there is no differentiate that angle is proposed with innovative thinking from workpiece imaging and image intelligent.
4) patent《Fluorescent magnetic particle flaw detection detection method based on image procossing and detecting system》(publication number:
CN105510429A capture) is carried out to the workpiece for being attached with fluorescentmagnetic particle(powder) liquid using colored CCD, by cromogram
R, G, B figure layer of picture are overlapped and subtract each other processing to improve image preprocessing effect, and calculate those suspected defects on this basis
Length, width, area, the length and width of pattern, to judge whether the those suspected defects are crackle.This process employs color cameras not
With figure layer to the sensitivity difference of lambda1-wavelength, belong to image preprocessing scope, do not relate to the design of imaging system with
Optimization, original image quality is unsatisfactory, has limited to the detection result of this method.
It follows that also lack effective means to fluorescent magnetic particle flaw detection image vision intelligent measurement at present, can not complete pair
The integrated information of welding production obtains and defect intelligent distinguishing, to cause serious safety to the production of extensive automatic welding
Hidden danger.Therefore, those skilled in the art is dedicated to providing a kind of surface defect online test method based on multiple information synthesis
And technology.
Invention content
In view of the above-mentioned deficiency of the prior art, the technical problem to be solved by the present invention is to realize fluorescentmagnetic particle(powder) crackle certainly
Dynamicization detects.
To achieve the above object, the present invention provides a kind of automatic detection recognition method of magnetic powder inspection image, the methods
Include the following steps:
Step 1 is irradiated measurand using the ultraviolet source of incidence angles degree, and quilt is shot with camera and camera lens
Survey multiple image of the object under different illumination conditions;
Step 2 carries out Pixel-level rate of gray level calculating to multiple image of the measurand under different illumination conditions, i.e.,
For each pixel on measurand image, extraction same position pixel shot under different illumination conditions several
Gray value in image calculates the sum of gray-value variation rate;
Step 3, the pixel for the sum of gray-value variation rate higher than given threshold are labeled as white, other element markings
For black, corresponding binary image is obtained;
Step 4 is filtered the white particle in binary image by size, to remaining particle after filtering, is calculated
The score of remainder particulate characteristic parameter;
Step 5 sums the score of different characteristic parameter, the characteristic parameter comprehensive score as the particle;
Step 6 the characteristic parameter comprehensive score of particle is compared with characteristic parameter comprehensive score threshold value, characteristic parameter
Comprehensive score is judged as defect, i.e. crackle in measurand less than the particle of characteristic parameter comprehensive score threshold value.
Further, fluorescentmagnetic particle(powder) liquid is adhered in the measurand;When the measurand is on-plane surface workpiece, use
The annular ultraviolet source of different-diameter and with camera coaxial placement, the camera lens be spherical surface wide-angle lens;The measurand is
When flat work pieces, different angles are formed using bar shaped ultraviolet source and from camera, the camera lens is common high definition camera lens;The phase
Machine is high resolution CCD camera, and the ultraviolet source, the camera lens and the camera are fixed in mounting bracket.
Further, the annular ultraviolet source of the different-diameter includes large scale annular ultraviolet light, middle scale toroidal purple
Outer light and small-size annular ultraviolet light, the annular ultraviolet source of the different-diameter to the measurand carry out high angle, in
Angle and low angle illumination form the ultraviolet source illumination of incidence angles degree;The bar shaped ultraviolet source includes that high angle enters
Penetrate ultraviolet light, middle angle incident uv and low angle incident uv, to the measurand carry out high angle, middle angle and
Low angle illuminates, and forms the ultraviolet source illumination of incidence angles degree.
Further, under the different lighting conditions, the measurand is shot with camera lens, was shot
Ensure that camera immobilizes with measured workpiece relative position in journey;If because actual conditions can not avoid camera and measured workpiece position
Deviation occurs, is matched by image translation, rotation, scale transformation, to reach measured workpiece on the image consistent;In height
Under the conditions of angle illumination, the measured workpiece picture of shooting is that high angle illuminates picture, under the conditions of middle angle illumination, the quilt of shooting
Survey workpiece picture is middle angle illumination picture, and under low angle lighting condition, the measured workpiece picture of shooting illuminates for low angle
Picture.
Further, the average brightness value of high angle illumination picture average brightness and low angle illumination picture it
Than being not less than 3.
Further, the threshold value of the sum of described gray-value variation rate is 100;The size is area, and area is less than 20
The white particle of pixel is filtered out.
Further, there are three the characteristic parameter of the remainder particulate has altogether, respectively compactedness, perimeter/area, external
Rectangle frame short side/long side;Between the threshold selection of each characteristic parameter is 0.2~0.5, when characteristic parameter is less than threshold value
When, this of particle is scored at 0, and when characteristic parameter is higher than threshold value, particle is scored at the difference of characteristic parameter and threshold value.
Further, the specific value of the threshold value of the characteristic parameter is determined according to the body characteristics of measured workpiece defect.
Further, the characteristic parameter comprehensive score threshold selection is 0.5~0.7.
Further, the specific value of threshold value of the characteristic parameter comprehensive score is according to the body characteristics of measured workpiece defect
It determines.
The method of the present invention has the advantages that compared with prior art:The present invention can overcome workpieces surface condition pair
The influence of magnetic powder image, need to only ensureing the average gray value of high angle illumination image and low angle illumination image, there are 3 times or more
Difference, the image binaryzation method based on gray-value variation rate sum have higher adaptive ability.In the binary image
In, the present invention is proposed based on three morphological feature nonlinearity in parameters blending algorithms, has calculating speed fast and stability
High feature helps to improve automation and the intelligent level of fluorescent magnetic particle flaw detection, can be widely used in magnetic metal
Part manufacturing field.
The technique effect of the design of the present invention, concrete structure and generation is described further below with reference to attached drawing, with
It is fully understood from the purpose of the present invention, feature and effect.
Description of the drawings
Fig. 1 is the imaging method schematic diagram of the preferred embodiment of the present invention;
Fig. 2 is the workpiece photo of the large scale annular ultraviolet light of the preferred embodiment of the present invention;
Fig. 3 is the workpiece photo of the middle scale toroidal ultraviolet light of the preferred embodiment of the present invention;
Fig. 4 is the workpiece photo of the small-size annular ultraviolet light of the preferred embodiment of the present invention;
Fig. 5 is the position of the pixel of the selection of the preferred embodiment of the present invention in the picture;
Fig. 6 is the gray value of the selected pixel of the preferred embodiment of the present invention under different lighting conditions;
Fig. 7 is the sum of gray value and change rate of the selected pixel of the preferred embodiment of the present invention;
Fig. 8 is that the gray-value variation rate of the preferred embodiment of the present invention is more than the pixel label result of threshold value;
Fig. 9 is the particle number of the label picture processing of the preferred embodiment of the present invention;
Figure 10 is the calculation of characteristic parameters result (particles filled degree) of the preferred embodiment of the present invention;
Figure 11 is calculation of characteristic parameters result (particle boundary rectangle short side and the long side of the preferred embodiment of the present invention
Ratio);
Figure 12 is the calculation of characteristic parameters result (ratio of particle circumference and area) of the preferred embodiment of the present invention;
Figure 13 be the present invention a preferred embodiment tag image in particle comprehensive score;
Figure 14 is the image procossing final result of the preferred embodiment of the present invention;
Figure 15 is the imaging method schematic diagram of another preferred embodiment of the present invention.
Wherein, 1-CCD cameras, 2- camera lenses, 3- measured workpieces, 4- crackles, 5- large scale annular ultraviolet lights, size ring in 6-
Shape ultraviolet light, 7- small-size annular ultraviolet lights, 8- fluorescentmagnetic particle(powder)s, 9- high angle incident uvs, angle incidence is ultraviolet in 10-
Light, 11- low angle incident uvs.
Specific implementation mode
Multiple preferred embodiments that the present invention is introduced below with reference to Figure of description, keep its technology contents more clear and just
In understanding.The present invention can be emerged from by many various forms of embodiments, and protection scope of the present invention not only limits
The embodiment that Yu Wenzhong is mentioned.
In the accompanying drawings, the identical component of structure is indicated with same numbers label, everywhere the similar component of structure or function with
Like numeral label indicates.The size and thickness of each component shown in the drawings are to be arbitrarily shown, and there is no limit by the present invention
The size and thickness of each component.In order to keep diagram apparent, some places suitably exaggerate the thickness of component in attached drawing.
The automatic detection recognition method of magnetic powder inspection image provided by the present invention, including illumination imaging method and defect image
Recognizer two parts.
Embodiment one
As shown in Figure 1, the intermediate frequency that the detected workpiece 3 in the present embodiment is medium carbon steel workpiece is heat-treated surface, it is detected
The surface of workpiece 3 is curved surface, is a kind of on-plane surface workpiece.The crackle of workpiece surface such as 4 example of crackle.For on-plane surface workpiece,
Using the annular ultraviolet source and camera 1,2 coaxial placement of camera lens of 3 different-diameters, camera lens 2 is spherical surface wide-angle lens, ultraviolet
Light source carries out high angle, middle angle and low angle illumination to being detected workpiece 3.Under different lighting conditions, with 1 He of CCD camera
Camera lens 2 is shot to being detected workpiece 3, ensures that camera immobilizes with measured workpiece relative position in shooting process.Annular
Ultraviolet source, camera 1 and camera lens 2 are fixed in mounting bracket.
As shown in Fig. 2, being the workpiece image obtained when being illuminated using large scale annular ultraviolet light 5;Fig. 3 is size in using
The workpiece image that annular ultraviolet light 6 obtains when illuminating;Fig. 4 is the workpiece figure obtained when being illuminated using small-size annular ultraviolet light 7
Picture.L3 is the diameter of large scale annular ultraviolet light 5, and L2 is the diameter of middle scale toroidal ultraviolet light 6, and L1 is small size
The diameter of annular ultraviolet light 7.
The light source of different incidence angles irradiates successively, and camera 1 and the relative position of workpiece 3 remain unchanged, if because of actual conditions
When can not avoid position deviation, it should be matched by transformation such as image translation, rotation, scalings, be existed with reaching subject
It is consistent on image.After shooting finishes, accurate corresponding three width image is obtained, then image node-by-node algorithm carries to treated
The gray value of each pixel is taken, rate of gray level is calculated and is summed, that is, extracts same position pixel under different illumination conditions
Gray value in three width images of shooting calculates the sum of gray-value variation rate.The position of the pixel that Fig. 5 displays are chosen in the picture
It sets, Fig. 6 shows that the selected gray value of pixel under different lighting conditions, Fig. 7 show that the gray value of selected pixel becomes
The sum of rate.
Due to luminescence mechanism difference, the magnetic powder particle of workpiece surface cracks aggregation is more quick to ultraviolet angle of light
Sense, the sum of gray-value variation rate of cracks pixel are higher.Pixel for the sum of rate of gray level higher than given threshold clicks through
Line flag.It completes that pixel is marked after traversal calculates, is labeled as by gray-value variation rate and higher than the pixel of threshold value white
Color, other pixels are labeled as black, obtain corresponding binary image.Binarization method avoids workpieces surface condition to single width figure
The otherness of the influence of image brightness, different lighting condition hypographs is the premise of this method, usually requires that high angle illuminates picture
The ratio between the average brightness value of average brightness and low angle illumination picture is not less than 3.In the present embodiment, using 100 as gray value
Change rate and threshold value carry out element marking, i.e., the pixel that gray-value variation rate is more than 100 are labeled as white, by rest of pixels mark
It is denoted as black, image is as shown in Figure 8 after obtained label.
In binary image, first white particle is filtered by size, i.e., area is less than to the particle of 20 pixels
It filters out, then compactedness, perimeter/area, boundary rectangle frame short side/long side totally 3 characteristic parameters is calculated to remainder particulate.Such as Fig. 9
It is shown, the particle number of label picture processing.
Between the threshold selection of each characteristic parameter is 0.2~0.5, specific value can be according to the body of measured workpiece defect
Feature determines.When characteristic parameter is less than threshold value, this of particle is scored at 0;When characteristic parameter is higher than threshold value, particle
It is divided into the difference of characteristic parameter and threshold value.Score calculating is carried out to three morphological feature parameters of each particle, and three are obtained
Dividing final characteristic parameter comprehensive score of the summation as the particle, the threshold selection of characteristic parameter comprehensive score is 0.5~0.7,
Specific value can be determined according to the body characteristics of measured workpiece defect.When the final characteristic parameter comprehensive score of particle is less than threshold value
It is judged to defect, otherwise it is assumed that being normal surface.The result of calculation of characteristic parameter is as shown in Figure 10, Figure 11 and Figure 12, wherein filling out
Degree of filling threshold value is set as 0.2, and the short long side of boundary rectangle is set as 0.4 than threshold value, particle circumference and area than threshold value be set as 0.5.
As shown in figure 13, the comprehensive score of each particle in image is marked.Particle 0#, 1#, 7#, 8#, 10# and 11#'s is comprehensive
It closes score and is substantially less than other particles, can accurately be branched away crackle particle for threshold value with 0.5.Figure 14 is image procossing
Final result.Testing result shows that this method accurately detects the crackle of workpiece surface.
Embodiment two
It is that the detected illumination imaging method of workpiece is different from the difference of embodiment one.
As shown in figure 15, the surface of the detected workpiece 3 in the present embodiment is plane, and workpiece surface adheres to fluorescentmagnetic particle(powder) 8.
For the workpiece that surface is plane, different angles are formed using bar shaped ultraviolet source and from camera, bar shaped ultraviolet source includes height
Angle incident uv 9, middle angle incident uv 10 and low angle incident uv 11, to measurand carry out high angle,
Middle angle and low angle illumination form the ultraviolet source illumination of incidence angles degree.Camera is high resolution CCD camera 1, camera lens
2 be common high definition camera lens, and ultraviolet source, camera lens 2 and camera 1 are fixed in mounting bracket.Using high angle incident uv
9, when middle angle incident uv 10 and low angle incident uv 11 are illuminated to being detected workpiece 3, three width of shooting are different
Workpiece image.The light source of different incidence angles irradiates successively, and camera 1 and the relative position of workpiece 3 remain unchanged, if because of practical feelings
When condition can not avoid position deviation, it should be matched by transformation such as image translation, rotation, scalings, to reach subject
On the image consistent.
After shooting finishes, the process of Identification of Cracks is carried out to image, it is identical with embodiment one.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that the ordinary skill of this field is without wound
The property made labour, which according to the present invention can conceive, makes many modifications and variations.Therefore, all technician in the art
Pass through the available technology of logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Scheme, all should be in the protection domain being defined in the patent claims.
Claims (10)
1. a kind of automatic detection recognition method of magnetic powder inspection image, which is characterized in that the described method comprises the following steps:
Step 1 is irradiated measurand using the ultraviolet source of incidence angles degree, with camera and tested pair of camera lens shooting
As the multiple image under different illumination conditions;
Step 2 carries out Pixel-level rate of gray level calculating to multiple image of the measurand under different illumination conditions, i.e., for
Each pixel on measurand image, the multiple image that extraction same position pixel is shot under different illumination conditions
In gray value, calculate the sum of gray-value variation rate;
Step 3, the pixel for the sum of gray-value variation rate higher than given threshold are labeled as white, and other pixels are labeled as black
Color obtains corresponding binary image;
Step 4 is filtered the white particle in binary image by size, to remaining particle after filtering, is calculated remaining
The score of particle characteristic parameter;
Step 5 sums the score of different characteristic parameter, the characteristic parameter comprehensive score as the particle;
The characteristic parameter comprehensive score of particle is compared step 6 with characteristic parameter comprehensive score threshold value, characteristic parameter synthesis
Score is judged as defect, i.e. crackle in measurand less than the particle of characteristic parameter comprehensive score threshold value.
2. the automatic detection recognition method of magnetic powder inspection image as described in claim 1, which is characterized in that in the measurand
Adhere to fluorescentmagnetic particle(powder) liquid;When the measurand is on-plane surface workpiece, using the annular ultraviolet source and and camera of different-diameter
Coaxial placement, the camera lens are spherical surface wide-angle lens;The measurand be flat work pieces when, using bar shaped ultraviolet source and with
Camera forms different angles, and the camera lens is common high definition camera lens;The camera is high resolution CCD camera, the ultraviolet light
Source, the camera lens and the camera are fixed in mounting bracket.
3. the automatic detection recognition method of magnetic powder inspection image as claimed in claim 2, which is characterized in that the different-diameter
Annular ultraviolet source includes large scale annular ultraviolet light, middle scale toroidal ultraviolet light and small-size annular ultraviolet light, the difference
The annular ultraviolet source of diameter carries out high angle, middle angle and low angle to the measurand and illuminates, and forms different incidence angles
The ultraviolet source of degree illuminates;The bar shaped ultraviolet source includes high angle incident uv, middle angle incident uv and low angle
Incident uv is spent, carrying out high angle, middle angle and low angle to the measurand illuminates, and forms the purple of incidence angles degree
Outer light source illuminates.
4. the automatic detection recognition method of magnetic powder inspection image as claimed in claim 3, which is characterized in that in the different illuminations
Under the conditions of, the measurand is shot with camera lens, ensures camera and measured workpiece relative position in shooting process
It immobilizes;If because actual conditions can not avoid camera and measured workpiece position that deviation occurs, pass through image translation, rotation, contracting
It puts transformation to be matched, to reach measured workpiece on the image consistent;Under high angle lighting condition, the measured workpiece of shooting
Picture is that high angle illuminates picture, and under the conditions of middle angle illumination, the measured workpiece picture of shooting is middle angle illumination picture,
Under low angle lighting condition, the measured workpiece picture of shooting is that low angle illuminates picture.
5. the automatic detection recognition method of magnetic powder inspection image as claimed in claim 4, which is characterized in that the high angle illumination
Picture average brightness and the ratio between the average brightness value of low angle illumination picture are not less than 3.
6. the automatic detection recognition method of magnetic powder inspection image as described in claim 1, which is characterized in that the gray-value variation
The threshold value of the sum of rate is 100;The size is area, and the white particle that area is less than 20 pixels is filtered out.
7. the automatic detection recognition method of magnetic powder inspection image as described in claim 1, which is characterized in that the remainder particulate
There are three characteristic parameter has altogether, respectively compactedness, perimeter/area, boundary rectangle frame short side/long side;Each feature ginseng
Between several threshold selections is 0.2~0.5, when characteristic parameter is less than threshold value, this of particle is scored at 0, works as characteristic parameter
When higher than threshold value, particle is scored at the difference of characteristic parameter and threshold value.
8. the automatic detection recognition method of magnetic powder inspection image as described in claim 1, which is characterized in that the characteristic parameter
The specific value of threshold value is determined according to the body characteristics of measured workpiece defect.
9. the automatic detection recognition method of magnetic powder inspection image as described in claim 1, which is characterized in that the characteristic parameter is comprehensive
It closes score threshold and is chosen to be 0.5~0.7.
10. the automatic detection recognition method of magnetic powder inspection image as described in claim 1, which is characterized in that the characteristic parameter
The specific value of threshold value of comprehensive score is determined according to the body characteristics of measured workpiece defect.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810254254.4A CN108460763B (en) | 2018-03-26 | 2018-03-26 | Automatic detection and identification method for magnetic powder inspection image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810254254.4A CN108460763B (en) | 2018-03-26 | 2018-03-26 | Automatic detection and identification method for magnetic powder inspection image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108460763A true CN108460763A (en) | 2018-08-28 |
CN108460763B CN108460763B (en) | 2021-03-30 |
Family
ID=63236616
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810254254.4A Active CN108460763B (en) | 2018-03-26 | 2018-03-26 | Automatic detection and identification method for magnetic powder inspection image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108460763B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111398411A (en) * | 2020-04-28 | 2020-07-10 | 秦皇岛市盛通无损检测有限责任公司 | Magnetic automatic check out system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101852768A (en) * | 2010-05-05 | 2010-10-06 | 电子科技大学 | Workpiece flaw identification method based on compound characteristics in magnaflux powder inspection environment |
CN103353480A (en) * | 2013-07-09 | 2013-10-16 | 中国科学院声学研究所 | Automatic ultrasonic flaw detection method and device for locomotive wheel shaft |
CN104062354A (en) * | 2013-03-18 | 2014-09-24 | 宝山钢铁股份有限公司 | Steel pipe magnetic powder inspection fluorescent image detection apparatus and detection method |
WO2014183385A1 (en) * | 2013-05-17 | 2014-11-20 | 中兴通讯股份有限公司 | Terminal and image processing method therefor |
CN107180428A (en) * | 2017-06-09 | 2017-09-19 | 电子科技大学 | A kind of the magneto optic images Enhancement Method based on optical flow algorithm |
-
2018
- 2018-03-26 CN CN201810254254.4A patent/CN108460763B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101852768A (en) * | 2010-05-05 | 2010-10-06 | 电子科技大学 | Workpiece flaw identification method based on compound characteristics in magnaflux powder inspection environment |
CN104062354A (en) * | 2013-03-18 | 2014-09-24 | 宝山钢铁股份有限公司 | Steel pipe magnetic powder inspection fluorescent image detection apparatus and detection method |
WO2014183385A1 (en) * | 2013-05-17 | 2014-11-20 | 中兴通讯股份有限公司 | Terminal and image processing method therefor |
CN103353480A (en) * | 2013-07-09 | 2013-10-16 | 中国科学院声学研究所 | Automatic ultrasonic flaw detection method and device for locomotive wheel shaft |
CN107180428A (en) * | 2017-06-09 | 2017-09-19 | 电子科技大学 | A kind of the magneto optic images Enhancement Method based on optical flow algorithm |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111398411A (en) * | 2020-04-28 | 2020-07-10 | 秦皇岛市盛通无损检测有限责任公司 | Magnetic automatic check out system |
Also Published As
Publication number | Publication date |
---|---|
CN108460763B (en) | 2021-03-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhou et al. | Automated visual inspection of glass bottle bottom with saliency detection and template matching | |
CA2986537C (en) | Method of inspecting a steel strip | |
US6667800B1 (en) | Method and device for measuring and quantifying surface defects on a test surface | |
US20060147092A1 (en) | Intelligent digital graphics inspection system and method | |
KR20160090359A (en) | Surface defect detection method and surface defect detection device | |
CN104483320B (en) | Digitized defect detection device and detection method of industrial denitration catalyst | |
CN111999305B (en) | Machine vision hardware system and hub spline combination gap vision detection method | |
CN109142509B (en) | Round steel magnetic powder flaw detection method and device | |
JP4045742B2 (en) | Nondestructive inspection method and apparatus | |
CN108257171A (en) | Car radar assembling aperture detection method based on light vision | |
CN107664644A (en) | A kind of apparent automatic detection device of object based on machine vision and method | |
Liu et al. | An automatic system for bearing surface tiny defect detection based on multi-angle illuminations | |
CN106872488A (en) | A kind of double surface defect visible detection methods of rapid large-area transparent substrate and device | |
CN111487192A (en) | Machine vision surface defect detection device and method based on artificial intelligence | |
CN112858351A (en) | X-ray nondestructive inspection method based on machine vision | |
CN109461156A (en) | The threaded closure plug assembly and detection method of view-based access control model | |
Fu et al. | Medicine glass bottle defect detection based on machine vision | |
CN107145896A (en) | Dysnusia identifying system based on fluorescentmagnetic particle(powder) | |
CN111426693A (en) | Quality defect detection system and detection method thereof | |
JP2011013007A (en) | Magnetic particle flaw inspection apparatus | |
Li et al. | The method for glass bottle defects detecting based on machine vision | |
CN111879789A (en) | Metal surface defect detection method and system | |
Chiou et al. | Flaw detection of cylindrical surfaces in PU-packing by using machine vision technique | |
Ma et al. | A machine vision assisted system for fluorescent magnetic particle inspection of railway wheelsets | |
CN108460763A (en) | A kind of automatic detection recognition method of magnetic powder inspection image |
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 |