CN106558048A - Screw array dystopy fault detection method and system - Google Patents
Screw array dystopy fault detection method and system Download PDFInfo
- Publication number
- CN106558048A CN106558048A CN201611066397.XA CN201611066397A CN106558048A CN 106558048 A CN106558048 A CN 106558048A CN 201611066397 A CN201611066397 A CN 201611066397A CN 106558048 A CN106558048 A CN 106558048A
- Authority
- CN
- China
- Prior art keywords
- screw
- image
- labelling
- area
- subimage
- 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
- G06T7/0006—Industrial image inspection using a design-rule based approach
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of screw array dystopy fault detection method and system, linear first labelling arranged in determining nut, screw area both sides are provided with the second labelling and the 3rd labelling, when normal, first labelling, second labelling and the 3rd labelling are in same straight line, when screw array failure, first labelling is vertical with second labelling and the 3rd labelling, obtain eyelid covering image, and extract screw area image in eyelid covering image, segmentation obtains the subimage in each screw area, calculate the length of nose section in each subimage, when the length of nose section in subimage is less than predetermined threshold value, judge the screw failure in the subimage.In whole process, using various image processing meanses, the accurately automatic diagnosis identification to screw array dystopy failure.
Description
Technical field
The present invention relates to fault detection technique field, more particularly to screw array dystopy fault detection method and system.
Background technology
Screw is usually used in fixed object, when being fixed than two objects for contacting in a big way to certain, can often adopt
It is fixed with screw array, for example eyelid covering screw.
Eyelid covering screw can bring impact directly to the safety of whole housing construction with the presence or absence of abnormal.With aircraft skin it is
Example, the high-quality engineering maintenance of high reliability is aircraft safety flight important guarantee.At present, the Maintenance errors in aviation based on people
It is difficult to avoid the aircraft safety hidden danger caused by human error.
For the detachable eyelid covering of aircraft, the installation of an eyelid covering is fixed and generally requires dozens of even up to a hundred soon
Formula screw is unloaded, flight crew needs the screw that removes and installs of continuous repetition in whole maintenance process, this simple, repeatedly and withered
Dry work, easily goes wrong, and causes neglected loading or does not tighten certain screw, causes potential safety hazard, or even cause flight safety thing
Therefore.This aids in solving this problem in the urgent need to a kind of means of automatization.
The content of the invention
Based on this, it is necessary to a kind of problem for there is no screw array dystopy fault detect at present, there is provided accurate spiral shell
Nail array dystopy fault detection method and system.
A kind of screw array dystopy fault detection method, including step:
The first linear labelling of nut is arranged in determining screw and the second of screw area both sides are respectively arranged at
Labelling and the 3rd labelling, when screw array is normal, first labelling, second labelling and the 3rd labelling are in
Same straight line, when screw array failure, first labelling is vertical with second labelling and the 3rd labelling;
Eyelid covering image is obtained, and extracts screw area image in eyelid covering image;
Segmentation screw area image, obtains the subimage in each screw area;
Calculate the length of nose section in each subimage;
The length of nose section in each subimage is compared with predetermined threshold value, when the length of nose section in subimage is less than
During predetermined threshold value, the screw failure in the subimage is judged.
A kind of screw array dystopy fault detection system, including:
Determining module, for determining the first linear labelling that nut is arranged in screw and being respectively arranged at screw
Second labelling and the 3rd labelling of area both sides, when screw array is normal, first labelling, second labelling and described
3rd labelling is in same straight line, when screw array failure, first labelling and second labelling and the described 3rd
Labelling is vertical;
Image zooming-out module, for obtaining eyelid covering image, and extracts screw area image in eyelid covering image;
Segmentation module, for splitting screw area image, obtains the subimage in each screw area;
Computing module, for calculating the length of nose section in each subimage;
Detection module, for the length of nose section in each subimage is compared with predetermined threshold value, when most long in subimage
When the length of line segment is less than predetermined threshold value, the screw failure in the subimage is judged.
Screw array dystopy fault detection method of the present invention and system, linear first labelling arranged in determining nut,
Screw area both sides are provided with the second labelling and the 3rd labelling, and when normal, the first labelling, the second labelling and the 3rd labelling are in
Same straight line, when screw array failure, first labelling is vertical with second labelling and the 3rd labelling, obtains
Eyelid covering image, and screw area image in eyelid covering image is extracted, segmentation obtains the subimage in each screw area, in each subimage of calculating most
The length of long line segment, when the length of nose section in subimage is less than predetermined threshold value, judges the screw failure in the subimage.
In whole process, using various image processing meanses, the accurately automatic diagnosis identification to screw array dystopy failure.
Description of the drawings
Fig. 1 is the schematic flow sheet of screw array dystopy fault detection method one embodiment of the present invention;
Fig. 2 is local aircraft skin figure;
Fig. 3 is screw array local aircraft skin figure under normal circumstances;
Fig. 4 is local aircraft skin figure in the case of screw array failure;
Fig. 5 is the schematic flow sheet of second embodiment of screw array dystopy fault detection method of the present invention;
Fig. 6 takes complement for eyelid covering;
Fig. 7 is that eyelid covering removes Background;
Fig. 8 is each screw region threshold segmentation figure;
Fig. 9 is screw skeleton drawing;
Figure 10 is each screw region branch point diagram;
Figure 11 is the locating segmentation figure in each screw region;
Figure 12 is each screw state diagnostic graph;
Figure 13 is the optimization figure in each screw region;
Figure 14 is the structural representation of screw array dystopy fault detection system one embodiment of the present invention;
Figure 15 is the structural representation of second embodiment of screw array dystopy fault detection system of the present invention.
Specific embodiment
As shown in figure 1, a kind of screw array dystopy fault detection method, including step:
S100:The first linear labelling of nut is arranged in determining screw and screw area both sides are respectively arranged at
Second labelling and the 3rd labelling, when screw array is normal, first labelling, second labelling and the 3rd labelling
In same straight line, when screw array failure, first labelling is vertical with second labelling and the 3rd labelling.
The detection object of screw array dystopy fault detection method of the present invention is screw, and specifically, in screw, nut sets
Be equipped with the first linear labelling, screw area both sides are respectively arranged with the second labelling and the 3rd labelling, when screw array it is normal
When, the first labelling, the second labelling and the 3rd labelling are in same straight line, when screw array failure, first labelling
It is vertical with second labelling and the 3rd labelling.By taking the screw in aircraft skin as an example, as shown in Fig. 2 Fig. 2 is covered for aircraft
Pi Tu, in fig. 2, the feature of aircraft screw is have a groove (the first labelling), screw area two in the middle part of screw nut
Side indicates the entopic two articles of short-terms of screw (the second labelling and the 3rd labelling), normal condition, as shown in figure 3, descending these three
Part is in a straight line, failure condition, lower groove as shown in Figure 4 and screw position tag line near normal.Screw event here
Hinder the non-tight condition for screw.
S200:Eyelid covering image is obtained, and extracts screw area image in eyelid covering image.
Eyelid covering image can directly be shot using image acquisition equipment or directly be imported from miscellaneous equipment (such as USB flash disk)
The eyelid covering image deposited, the equipment for shooting eyelid covering image can be digital camera, smart mobile phone or industrial camera.In eyelid covering image
Mainly include screw area image and background image, subsequent operation is primarily directed to screw area image and is operated, carries here
Take screw area image in eyelid covering image.It is non-essential, to avoid interference of the background image to subsequent treatment, need accurately to extract and cover
Screw area image in skin image.
In Fig. 5 wherein one embodiment, step S200 includes:
S220:Eyelid covering image is converted to into gray level image.
For ease of the Treatment Analysis of image, conversion chromatic image is gray-scale maps, screw groove (the first labelling) in gray-scale maps
It is obvious with image background regions contrast with screw both sides tag line (the second labelling and the 3rd labelling).
S240:Carry out taking complementary operation, opening operation, additive operation, medium filtering and adaptive to greyscale image data successively
Thresholding dividing processing is answered, screw area image in eyelid covering image is extracted.
Image is carried out taking complementary operation, screw area and eyelid covering background area is exchanged, as shown in Figure 6.In Fig. 6, image surrounding is inclined
Bright, middle part is partially dark, brightness irregularities, and opening operation and additive operation are carried out to image, removes image background regions brightness disproportionation
Affect.As shown in fig. 7, there is certain interference in detection identification of the screw annular regions to screw area, it is to ensure that eliminating circle ring area makes an uproar
While sound, retain the detailed information of screw groove and tag line, image is processed from medium filtering, after medium filtering
Image, whole screw regional luminance is uneven, is to eliminate impact of the luminance difference to image segmentation, from adaptive threshold pair
Image enters row threshold division, obtains image as shown in Figure 8.
S400:Segmentation screw area image, obtains the subimage in each screw area.
The corresponding screw area image of multiple screws is there are in screw area image, needs enter to each screw area image respectively
Row fault detect.Here, screw area image is obtained to step S200 splitting, the subimage in each screw area is obtained.It is concrete next
Say, dividing method can be:Image obtained by micronization processes step S200 simultaneously subtracts each other therewith, obtains screw skeleton spine image, carries
The branch point of screw skeleton spine image is taken, expansion process is carried out to screw skeleton spine branch point diagram picture, connect each screw area
The branch point in domain, counts the center-of-mass coordinate and region area, image obtained by segmentation step S200 in each screw region, obtains each screw
The subimage in area.
As shown in figure 5, wherein in one embodiment, step S400 includes step:
S410:Micronization processes to greatest extent are carried out to screw area image, screw skeleton image is obtained, wherein, to greatest extent
Micronization processes criterion is, under the premise of the connected pixel area in image is ensured does not rupture in thinning process, to approach picture as far as possible
Plain zone centerline.
Carry out micronization processes first to greatest extent, while the connected pixel area in ensureing image is continuous in thinning process
Split, in thinning process, approach pixel zone centerline as far as possible.Specifically as shown in figure 9, Fig. 9 is to carry out maximum to screw area image
The image obtained after limit micronization processes, in Fig. 9, line-like structures are referred to as the skeleton of image.
S420:Obtain screw skeleton spine image.
By the image after micronization processes and screw area image subtraction, screw skeleton spine image is obtained.In simple terms, it is false
It is A images to determine screw area image, obtains B images, A images are deducted B images, is obtained after micronization processes to greatest extent are carried out to which
Obtain screw skeleton spine image C.That is screw skeleton spine image C=A image-B images.
S430:Extract the branch point in screw skeleton spine image.
For the screw skeleton spine image that step S420 is obtained, branch point therein is extracted.Specifically, remove first
Step S420 obtain screw skeleton spine image in spine, then, by obtain image again with screw skeleton spine image phase
Subtract, obtain screw image framework spine, finally, determine screw skeleton spine image branch point, and as each screw area key point.
In simple terms, it is assumed that step S420 obtain for screw skeleton spine image C, remove the spine in screw skeleton spine image C,
Image D is obtained, then screw skeleton spine image C and image D is subtracted each other, obtain screw image framework spine, by the screw for obtaining
Image framework spine is as screw skeleton spine image branch point, concrete as shown in Figure 10.
S440:Expansion process branch point, to connect the branch point in each screw area.
Expansion process is carried out to the branch point that step S430 is obtained, to connect the branch point in each screw area.
S450:The center-of-mass coordinate and region area in each screw area are counted, segmentation screw area image obtains the son in each screw area
Image.
After the branch point that step S440 connects in each screw area, it is possible to achieve the approximate location to each target area, enter
One step combines each screw area area, then the screw region in divisible image, concrete as shown in figure 11.
S600:Calculate the length of nose section in each subimage.
The subimage in each screw area obtained for step S400, calculates the length of nose section in subimage respectively.Tool
For body, calculate and Hough transformation can be adopted to calculate the nose section in each subimage, i.e., carry out in parameter space simple
Cumulative statistics, by the straight line in accumulator peakvalue's checking image space is found in hough space.
S800:The length of nose section in each subimage is compared with predetermined threshold value, when the length of nose section in subimage
When degree is less than predetermined threshold value, the screw failure in the subimage is judged.
Nose section and threshold value in each subimage of comparison, if the nose section is less than threshold value, in judging the subimage
Screw failure, otherwise, screw in the subimage is normal, as shown in figure 12.
It is non-essential, as shown in figure 5, wherein in one embodiment, also including step before step S800:
S720:Random selection obtains the subimage at least 2 width screw areas.
The subimage in the screw area obtained to step S400, randomly selects at least 2 width.To improve the standard of final fault detect
Exactness, can obtain the subimage in more screw areas here, it might even be possible to obtain the subimage in whole screw areas.
S740:Between the second labelling and the 3rd labelling in the subimage in the screw area for calculating selection, longest distance is equal
Value.
Calculation procedure S720 obtain screw area subimage between the second labelling and the 3rd labelling longest distance it is equal
Value, so that the accuracy of data is improved by the way of averaging.
S760:Average is multiplied by into proportionality coefficient, predetermined threshold value is obtained, proportionality coefficient is more than or equal to 0.8 and is less than or equal to
0.9。
The average that step S740 is obtained is multiplied by into 0.8~0.9, predetermined threshold value is obtained.
Specifically, randomly choose 5~10 width of S400 neutron images, when screw it is normal, the most long straight line in the subimage
Through the first labelling, the second labelling and the 3rd labelling, when the nose section in screw failure, the subimage cannot run through three simultaneously
Individual labelling, therefore calculate the average of longest distance between the second labelling and the 3rd labelling in each subimage, this value is multiplied by into 0.8~
0.9 used as threshold value.
Screw array dystopy fault detection method of the present invention, linear first labelling arranged in determining nut, screw area
Both sides are provided with the second labelling and the 3rd labelling, and when normal, the first labelling, the second labelling and the 3rd labelling are in same
Straight line, when screw array failure, first labelling is vertical with second labelling and the 3rd labelling, obtains eyelid covering figure
Picture, and screw area image in eyelid covering image is extracted, segmentation obtains the subimage in each screw area, calculates nose section in each subimage
Length, when nose section in subimage length be less than predetermined threshold value when, judge the screw failure in the subimage.Whole mistake
Cheng Zhong, using various image processing meanses, the accurately automatic diagnosis identification to screw array dystopy failure.
As shown in figure 5, also including before step S400 in one embodiment wherein:
S300:Screw area image is expanded and corrosion treatmentCorrosion Science successively.
(second marks tag line after normal screw area Threshold segmentation in screw groove line segment (the first labelling) Yu screw both sides
Note and the 3rd labelling) do not plan a successor between section, to connect screw groove line segment and the tag line of screw both sides, successively to image
Expanded and corrosion treatmentCorrosion Science, to optimize screw area image, as shown in figure 13.
As shown in figure 14, a kind of screw array dystopy fault detection system, including:
Determining module 100, for determining the first linear labelling that nut is arranged in screw and being respectively arranged at
Second labelling and the 3rd labelling of screw area both sides, when screw array is normal, first labelling, second labelling and
3rd labelling is in same straight line, when screw array failure, first labelling and second labelling and described
3rd labelling is vertical.
Image zooming-out module 200, for obtaining eyelid covering image, and extracts screw area image in eyelid covering image.
Segmentation module 400, for splitting screw area image, obtains the subimage in each screw area.
Computing module 600, for calculating the length of nose section in each subimage.
Detection module 800, for the length of nose section in each subimage is compared with predetermined threshold value, when in subimage most
When the length of long line segment is less than predetermined threshold value, the screw failure in the subimage is judged.
Screw array dystopy fault detection system of the present invention, linear first arranged in the determination nut of determining module 100
Labelling, screw area both sides are provided with the second labelling and the 3rd labelling, when normal, the first labelling, the second labelling and the 3rd mark
In same straight line, when screw array failure, first labelling is hung down note with second labelling and the 3rd labelling
Directly, image zooming-out module 200 obtains eyelid covering image, and extracts screw area image in eyelid covering image, and the segmentation segmentation of module 400 is obtained
The subimage in each screw area, computing module 600 calculate the length of nose section in each subimage, when nose section in subimage
When length is less than predetermined threshold value, detection module 800 judges the screw failure in the subimage.In whole process, using various figures
As processing means, the accurately automatic diagnosis identification to screw array dystopy failure, it is ensured that by the safety of eyelid covering housing construction.
As shown in figure 15, wherein in one embodiment, image zooming-out module 200 includes:
Gradation conversion unit 220, for eyelid covering image is converted to gray level image.
Extraction unit 240, for carrying out successively taking complementary operation, opening operation, additive operation, intermediate value filter to greyscale image data
Ripple and self-adaptive filtering method are processed, and extract screw area image in eyelid covering image.
As shown in figure 15, wherein in one embodiment, screw array dystopy fault detection system also includes:
Optimization module 300, for being expanded successively and corrosion treatmentCorrosion Science to screw area image.
As shown in figure 15, wherein in one embodiment, segmentation module 400 includes:
Skeleton image acquiring unit 410, for micronization processes to greatest extent are carried out to screw area image, obtains screw skeleton
Image, wherein, micronization processes criterion is before the connected pixel area in guarantee image does not rupture in thinning process to greatest extent
Put, approach pixel zone centerline as far as possible.
Spine image acquisition unit 420, for obtaining screw skeleton spine image.
Bifurcation extracting unit 430, for extracting the branch point in screw skeleton spine image.
Expansion cell 440, for expansion process branch point, to connect the branch point in each screw area.
Cutting unit 450, for counting the center-of-mass coordinate and region area in each screw area, segmentation screw area image is obtained
The subimage in each screw area.
Embodiment described above only expresses the several embodiments of the present invention, and its description is more concrete and detailed, but and
Therefore can not be construed as limiting the scope of the patent.It should be pointed out that for one of ordinary skill in the art comes
Say, without departing from the inventive concept of the premise, some deformations and improvement can also be made, these belong to the protection of the present invention
Scope.Therefore, the protection domain of patent of the present invention should be defined by claims.
Claims (10)
1. a kind of screw array dystopy fault detection method, it is characterised in that including step:
The first linear labelling that nut is arranged in determining screw and the second labelling for being respectively arranged at screw area both sides
With the 3rd labelling, when screw array is normal, first labelling, second labelling and the 3rd labelling are in same
Bar straight line, when screw array failure, first labelling is vertical with second labelling and the 3rd labelling;
Eyelid covering image is obtained, and extracts screw area image in the eyelid covering image;
Split screw area image, obtain the subimage in each screw area;
Calculate the length of nose section in each subimage;
The length of nose section in each subimage is compared with predetermined threshold value, when the length of nose section in the subimage
During less than the predetermined threshold value, the screw failure in the subimage is judged.
2. screw array dystopy fault detection method according to claim 1, it is characterised in that the acquisition eyelid covering
Include the step of screw area image in image:
The eyelid covering image is converted to into gray level image;
The greyscale image data is carried out taking complementary operation, opening operation, additive operation, medium filtering and adaptive threshold successively
Dividing processing, extracts screw area image in the eyelid covering image.
3. screw array dystopy fault detection method according to claim 1, it is characterised in that the segmentation screw
Also include before the step of area's image, subimage in acquisition each screw area:
Screw area image is expanded and corrosion treatmentCorrosion Science successively.
4. screw array dystopy fault detection method according to claim 1, it is characterised in that the segmentation screw
The step of area's image, subimage for obtaining each screw area, includes:
Micronization processes to greatest extent are carried out to screw area image, screw skeleton image is obtained, wherein, it is described thin to greatest extent
It is, under the premise of the connected pixel area in image is ensured does not rupture in thinning process, to approach pixel as far as possible to change process criterion
Zone centerline;
Obtain screw skeleton spine image;
Extract the branch point in the screw skeleton spine image;
Branch point described in expansion process, to connect the branch point in each screw area;
The center-of-mass coordinate and region area in each screw area are counted, splits screw area image, obtain the subimage in each screw area.
5. screw array dystopy fault detection method according to claim 1, it is characterised in that each son of the calculating
Include the step of the length of nose section in image:
Nose section in each subimage is calculated using Hough transformation.
6. screw array dystopy fault detection method according to claim 1, it is characterised in that described by each subgraph
Also include before the step of length of nose section is compared with predetermined threshold value as in:
Random selection obtains the subimage at least 2 width screw areas;
Calculate the average of longest distance between second labelling and the 3rd labelling in the subimage in the screw area for choosing;
The average is multiplied by into proportionality coefficient, predetermined threshold value is obtained, the proportionality coefficient is more than or equal to 0.8 and is less than or equal to 0.9.
7. a kind of screw array dystopy fault detection system, it is characterised in that include:
Determining unit, for determining the first linear labelling that nut is arranged in screw and being respectively arranged at screw area two
Second labelling and the 3rd labelling of side, when screw array is normal, first labelling, second labelling and the described 3rd
Labelling is in same straight line, when screw array failure, first labelling and second labelling and the 3rd labelling
Vertically;
Image zooming-out module, for obtaining eyelid covering image, and extracts screw area image in the eyelid covering image;
Segmentation module, for splitting screw area image, obtains the subimage in each screw area;
Computing module, for calculating the length of nose section in each subimage;
Detection module, for the length of nose section in each subimage is compared with predetermined threshold value, when in the subimage
When the length of nose section is less than the predetermined threshold value, the screw failure in the subimage is judged.
8. screw array dystopy fault detection system according to claim 7, it is characterised in that described image extraction module
Including:
Gradation conversion unit, for the eyelid covering image is converted to gray level image;
Extraction unit, for carrying out successively taking complementary operation, opening operation, additive operation, medium filtering to the greyscale image data
And self-adaptive filtering method is processed, screw area image in the eyelid covering image is extracted.
9. screw array dystopy fault detection system according to claim 7, it is characterised in that also include:
Optimization module, for being expanded successively and corrosion treatmentCorrosion Science to screw area image.
10. screw array dystopy fault detection system according to claim 7, it is characterised in that segmentation module includes:
Skeleton image acquiring unit, for micronization processes to greatest extent are carried out to screw area image, obtains screw skeleton drawing
Picture, wherein, micronization processes criterion is that connected pixel area in image is ensured does not rupture in thinning process premise to greatest extent
Under, pixel zone centerline is approached as far as possible;
Spine image acquisition unit, for obtaining screw skeleton spine image;
Bifurcation extracting unit, for extracting the branch point in the screw skeleton spine image;
Expansion cell, for branch point described in expansion process, to connect the branch point in each screw area;
Cutting unit, for counting the center-of-mass coordinate and region area in each screw area, splits screw area image, obtains each spiral shell
The subimage in nail area.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611066397.XA CN106558048B (en) | 2016-11-28 | 2016-11-28 | Screw array dystopy fault detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611066397.XA CN106558048B (en) | 2016-11-28 | 2016-11-28 | Screw array dystopy fault detection method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106558048A true CN106558048A (en) | 2017-04-05 |
CN106558048B CN106558048B (en) | 2019-05-24 |
Family
ID=58445446
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611066397.XA Active CN106558048B (en) | 2016-11-28 | 2016-11-28 | Screw array dystopy fault detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106558048B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112184588A (en) * | 2020-09-29 | 2021-01-05 | 哈尔滨市科佳通用机电股份有限公司 | Image enhancement system and method for fault detection |
CN113486739A (en) * | 2021-06-22 | 2021-10-08 | 深圳无境创新科技有限公司 | Screw detection method and device, electronic equipment and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5221984B2 (en) * | 2008-02-29 | 2013-06-26 | 三菱重工業株式会社 | Fastener looseness detection system and looseness detection method |
CN103500328A (en) * | 2013-10-16 | 2014-01-08 | 北京航空航天大学 | Method for automatically detecting deflection fault of railway wagon locking plate |
CN104318582A (en) * | 2014-11-14 | 2015-01-28 | 西南交通大学 | Detection method for bad state of rotating double-lug component pin of high-speed rail contact network on basis of image invariance target positioning |
CN104680517A (en) * | 2015-01-22 | 2015-06-03 | 清华大学 | Looseness detection method of bolt |
CN105241679A (en) * | 2015-09-21 | 2016-01-13 | 中国铁道科学研究院电子计算技术研究所 | Method for detecting hidden fault of motor train unit |
CN105354827A (en) * | 2015-10-08 | 2016-02-24 | 成都唐源电气有限责任公司 | Method and system for intelligently identifying clamp nut shedding in catenary image |
CN105424711A (en) * | 2015-12-04 | 2016-03-23 | 清华大学苏州汽车研究院(吴江) | Automatic test equipment and method for screw looseness of locomotive |
CN105424330A (en) * | 2014-08-29 | 2016-03-23 | 北京航空航天大学 | Lorry shaft end bolt fault detecting method and lorry shaft end bolt fault detecting device |
-
2016
- 2016-11-28 CN CN201611066397.XA patent/CN106558048B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5221984B2 (en) * | 2008-02-29 | 2013-06-26 | 三菱重工業株式会社 | Fastener looseness detection system and looseness detection method |
CN103500328A (en) * | 2013-10-16 | 2014-01-08 | 北京航空航天大学 | Method for automatically detecting deflection fault of railway wagon locking plate |
CN105424330A (en) * | 2014-08-29 | 2016-03-23 | 北京航空航天大学 | Lorry shaft end bolt fault detecting method and lorry shaft end bolt fault detecting device |
CN104318582A (en) * | 2014-11-14 | 2015-01-28 | 西南交通大学 | Detection method for bad state of rotating double-lug component pin of high-speed rail contact network on basis of image invariance target positioning |
CN104680517A (en) * | 2015-01-22 | 2015-06-03 | 清华大学 | Looseness detection method of bolt |
CN105241679A (en) * | 2015-09-21 | 2016-01-13 | 中国铁道科学研究院电子计算技术研究所 | Method for detecting hidden fault of motor train unit |
CN105354827A (en) * | 2015-10-08 | 2016-02-24 | 成都唐源电气有限责任公司 | Method and system for intelligently identifying clamp nut shedding in catenary image |
CN105424711A (en) * | 2015-12-04 | 2016-03-23 | 清华大学苏州汽车研究院(吴江) | Automatic test equipment and method for screw looseness of locomotive |
Non-Patent Citations (2)
Title |
---|
周志峰: "《光杆六角螺栓的视觉检测》", 《光学技术》 * |
周桐等: "《基于机器视觉的螺钉多参数在线监测》", 《机械工业标准化与质量》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112184588A (en) * | 2020-09-29 | 2021-01-05 | 哈尔滨市科佳通用机电股份有限公司 | Image enhancement system and method for fault detection |
CN113486739A (en) * | 2021-06-22 | 2021-10-08 | 深圳无境创新科技有限公司 | Screw detection method and device, electronic equipment and storage medium |
CN113486739B (en) * | 2021-06-22 | 2024-05-24 | 深圳无境创新科技有限公司 | Screw detection method, device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN106558048B (en) | 2019-05-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
GB2593639A (en) | Risk event identification in maritime data and usage thereof | |
EP2608536B1 (en) | Method for counting objects and apparatus using a plurality of sensors | |
US11328421B2 (en) | Image processing apparatus, image processing method, and storage medium | |
CN110781839A (en) | Sliding window-based small and medium target identification method in large-size image | |
US20190197313A1 (en) | Monitoring device | |
CN109740609B (en) | Track gauge detection method and device | |
US8090151B2 (en) | Face feature point detection apparatus and method of the same | |
CN106558048B (en) | Screw array dystopy fault detection method and system | |
CN111080668A (en) | Brake pad wear fault detection method and system | |
CN116433938A (en) | Abnormal detection method, system and storage medium for bolt looseness of subway gearbox | |
US11386538B2 (en) | Image processing apparatus, image processing method, and storage medium | |
CN111626104B (en) | Cable hidden trouble point detection method and device based on unmanned aerial vehicle infrared thermal image | |
US11922680B2 (en) | Image determination device, image determination method, and non-transitory computer readable medium storing program | |
US10786197B2 (en) | Evaluation method for site of color irregularity and color irregularity site evaluation apparatus | |
CN113435353A (en) | Multi-mode-based in-vivo detection method and device, electronic equipment and storage medium | |
CN109472289B (en) | Key point detection method and device | |
CN116002480A (en) | Automatic detection method and system for accidental falling of passengers in elevator car | |
CN115393791A (en) | Foreign matter detection method, device and computer readable storage medium | |
CN114998194A (en) | Product defect detection method, system and storage medium | |
CN110634124A (en) | Method and equipment for area detection | |
Muzammil et al. | Application of image processing techniques for the extraction of vehicle number plates over ARM target board | |
CN113020428A (en) | Processing monitoring method, device and equipment of progressive die and storage medium | |
CN111898425A (en) | State judgment method and device for switching-on and switching-off indicator of transformer substation | |
CN111709360A (en) | Safety rope wearing identification method and system | |
CN106127100B (en) | A kind of robot vision identifying system and its image processing method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |