CN110111322A - A kind of tunnel defect identifying system based on image - Google Patents
A kind of tunnel defect identifying system based on image Download PDFInfo
- Publication number
- CN110111322A CN110111322A CN201910394410.1A CN201910394410A CN110111322A CN 110111322 A CN110111322 A CN 110111322A CN 201910394410 A CN201910394410 A CN 201910394410A CN 110111322 A CN110111322 A CN 110111322A
- Authority
- CN
- China
- Prior art keywords
- image
- disease
- region
- tunnel
- pixel
- 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.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 18
- 201000010099 disease Diseases 0.000 claims abstract description 43
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 43
- 238000000605 extraction Methods 0.000 claims abstract description 13
- 238000003709 image segmentation Methods 0.000 claims abstract description 13
- 239000000284 extract Substances 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 238000006243 chemical reaction Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 9
- 230000000877 morphologic effect Effects 0.000 claims description 6
- 238000001514 detection method Methods 0.000 abstract description 10
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000007689 inspection Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The tunnel defect identifying system based on image that the present invention relates to a kind of, belong to tunnel defect detection technique field, it include: the image that image preprocessing system receives the vcehicular tunnel that image capturing system passes over, and the image after acquisition is pre-processed, increase the contrast in lining cutting apparent disease region and background area, output image is obtained, the spotted noise of output image, block distortion and linear noise section are extracted, then is filtered out；The image that noise was filtered out by image preprocessing system is carried out binary conversion treatment using image segmentation by image segmentation processing system；Image characteristic extraction system extracts the characteristic of image segmentation processing system treated image, realizes the automatic identification in the apparent disease region of tunnel-liner.The present invention is by denoising image, to enhance image, and carries out feature extraction and classification to target using the calculation for returning loss, achievees the purpose that automatic detection.
Description
Technical field
The invention belongs to tunnel defect detection technique fields, are related to a kind of tunnel defect identifying system based on image.
Background technique
During vcehicular tunnel Defect inspection, the mode for generalling use acquisition disease geo-radar image is detected, however mesh
Preceding detection image is not clear enough, cause it is subsequent there is error during identifying image, and at present when identifying image
It needs to carry out tune ginseng manually, leads to low efficiency, the problem of period is long and the poor robustness of detection method, in last image parameter meter
In the engineering of calculation, there is computational accuracy inaccuracy, strong influence vcehicular tunnel disease inspection in current calculation
The accuracy of survey.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of tunnel defect identifying system based on image.
In order to achieve the above objectives, the invention provides the following technical scheme:
The tunnel defect identifying system based on image that the present invention provides a kind of, comprising:
Image capturing system is acquired the image of vcehicular tunnel；
Image preprocessing system receives the image of vcehicular tunnel that described image acquisition system passes over, and to acquisition after
Image pre-processed, increase lining cutting apparent disease region and background area contrast, obtain output image, to output scheme
Spotted noise, block distortion and the linear noise section of picture extract, then are filtered out；
Image segmentation processing system will filter out the image of noise using image segmentation by described image pretreatment system
Carry out binary conversion treatment；Image characteristic extraction system to the characteristic of described image segmentation processing system treated image into
Row extracts, and realizes the automatic identification in the apparent disease region of tunnel-liner；Disease Characters categorizing system is to described image feature extraction
Disease geo-radar image after system identification is classified；
Parameters Computing System carries out parameter calculating to the sorted image of the Disease Characters categorizing system, obtains disease area
Disease morphological parameters in domain.
As a preferred technical solution of the present invention, described image Feature Extraction System uses region motion network and side
Boundary's frame returns loss and is identified, formula are as follows:
In formula, NregIndicating the number of target frame, i indicates the index of anchors,Indicate corresponding GT (GroundTurth)
Probability, tiIndicate anchors for GT prediction drift amount,Indicate that anchors is l for the true excursions amount of GTreg
() indicates to return loss to the bounding box of single anchors.
The Disease Characters categorizing system is classified using classification CNN network with total Classification Loss, formula are as follows:
N in formulaclsClassification number, piIndicate that anchors is the probability of target frame, lcls() is divided single target frame
Class loss.
As a preferred technical solution of the present invention, described image pretreatment system is in the process extracted to spotted noise
In, first count the quantity of pixel in each connected region, given threshold To, and to pixel quantity nkLess than the company of the threshold value
Logical domain removal, spotted noise extract formula are as follows: Is(x, y)={ Ck(x,y)|nk< To, k=1,2 ..., n }
In formula, Is(x,y),Ck(x, y) respectively indicates spotted noise and connected region.
Block distortion is removed using the method for calculating rectangular degree.Rectangular degree is defined as the total of some connected region kind pixel
Quantity nkWith the smallest boundary rectangle area Sk=(yb-ya)*(xb-xaThe ratio between), (xa,ya),(xb,yb) respectively indicate the connection
The area pixel point set commercial affairs lower left corner and upper right angular coordinate.Utilize D=nk/SkRectangular degree is calculated, and is made an uproar using formulas Extraction bulk
Sound:
Is(x, y)={ Ck(x, y) | D < TD, k=1,2 ..., n }.
As a preferred technical solution of the present invention, the Parameters Computing System is carrying out disease form in disease region
When parameter calculates, comprising the following steps:
D1: scanning image sequence, finds the 1st pixel belonged to not yet, if the pixel is (x0, y0)；
D2: centered on (x0, y0), consider 4 neighborhood territory pixels of (x0, y0)；It, will if (x, y) meets growth criterion
(x0, y0) and (x, y) merges (in the same area), while will be pressed into storehouse；
D3: taking out a pixel from storehouse, it is returned to step D2 as (x0, y0)；
D4: when storehouse is empty, step D1 is returned to；
D5: step D1-D4 is repeated when meeting termination condition, growth terminates.It, both can basis after generating 3D region
Three-dimensional Line Integral obtains the area of disease.
The beneficial effects of the present invention are:
The present invention is by denoising image, to enhance image, and using return loss calculation come
Feature extraction and classification are carried out to target, achievees the purpose that automatic detection, finally can accurately obtain disease morphological parameters, greatly
The precision for improving Defect inspection.
Other advantages, target and feature of the invention will be illustrated in the following description to a certain extent, and
And to a certain extent, based on will be apparent to those skilled in the art to investigating hereafter, Huo Zheke
To be instructed from the practice of the present invention.Target of the invention and other advantages can be realized by following specification and
It obtains.
Detailed description of the invention
To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention is made below in conjunction with attached drawing excellent
The detailed description of choosing, in which:
Fig. 1 is overall structure diagram of the invention.
Specific embodiment
Illustrate embodiments of the present invention below by way of specific specific example, those skilled in the art can be by this specification
Other advantages and efficacy of the present invention can be easily understood for disclosed content.The present invention can also pass through in addition different specific realities
The mode of applying is embodied or practiced, the various details in this specification can also based on different viewpoints and application, without departing from
Various modifications or alterations are carried out under spirit of the invention.It should be noted that diagram provided in following embodiment is only to show
Meaning mode illustrates basic conception of the invention, and in the absence of conflict, the feature in following embodiment and embodiment can phase
Mutually combination.
Wherein, the drawings are for illustrative purposes only and are merely schematic diagrams, rather than pictorial diagram, should not be understood as to this
The limitation of invention；Embodiment in order to better illustrate the present invention, the certain components of attached drawing have omission, zoom in or out, not
Represent the size of actual product；It will be understood by those skilled in the art that certain known features and its explanation may be omitted and be in attached drawing
It is understood that.
Referring to Fig. 1, being a kind of tunnel defect identifying system based on image, comprising:
Image capturing system is acquired the image of vcehicular tunnel；
Image preprocessing system receives the image for the vcehicular tunnel that image capturing system passes over, and to the figure after acquisition
As being pre-processed, increases the contrast in lining cutting apparent disease region and background area, obtain output image, to output image
Spotted noise, block distortion and linear noise section extract, then are filtered out；
Image segmentation processing system is carried out the image that noise was filtered out by image preprocessing system using image segmentation
Binary conversion treatment；Image characteristic extraction system extracts the characteristic of image segmentation processing system treated image,
Realize the automatic identification in the apparent disease region of tunnel-liner；After Disease Characters categorizing system identifies image characteristic extraction system
Disease geo-radar image is classified；
Parameters Computing System carries out parameter calculating to the sorted image of Disease Characters categorizing system, obtains in disease region
Disease morphological parameters.
Image preprocessing:
Image after original image that detection system is acquired or splicing, can have part uneven illumination phenomenon,
And picture contrast is lower, and the apparent disease region of lining cutting and background area brightness change are unobvious.It is that can produce in image segmentation
Raw many black patch noises also influence crack extract work.So being balanced illumination to the apparent disease geo-radar image of lining cutting first, improve
Contrast.
This system carries out image enhancement using improved template light fast filtering algorithm.First to original image Gauss low pass
Filtering obtains background image, by original image and background image difference processing, balances illumination, differentiated image is carried out linear
It stretches, increases the contrast in lining cutting apparent disease region and background area, obtain output image.After image enhancement, in image
Still there are a large amount of noise, algorithm for design is needed to filter out, only the apparent defect information of remaining lining cutting.So by not having largely
The data of value are got rid of.
According to the shape and distribution of noise, noise is classified as three categories, spotted noise including discrete distribution is got together
Block distortion and line-like linear noise, for the different minimizing technology of three kinds of Noise Designs.It is removed in spotted noise
In, count the quantity of pixel in each connected region, given threshold To, pixel quantity nkConnected domain less than the threshold value is gone
It removes, spotted noise extracts formula as shown in formula:
Is(x, y)={ Ck(x,y)|nk< To, k=1,2 ..., n }
In formula, Is(x,y),Ck(x, y) respectively indicates spotted noise and connected region.
Block distortion is removed using the method for calculating rectangular degree.Rectangular degree is defined as the total of some connected region kind pixel
Quantity nkWith the smallest boundary rectangle area Sk=(yb-ya)*(xb-xaThe ratio between), (xa,ya),(xb,yb) respectively indicate the connection
The area pixel point set commercial affairs lower left corner and upper right angular coordinate.Utilize D=nk/SkRectangular degree is calculated, and is made an uproar using formulas Extraction bulk
Sound:
Is(x, y)={ Ck(x, y) | D < TD, k=1,2 ..., n }
In linear noise remove, using the straight line in Hough transformation detection bianry image, straight line is filtered out, with straight line phase
There are spotted noises and block distortion for part even, are filtered out using the algorithm of design, then detect straight line again, threadiness is made an uproar
Sound filters out, and spotted noise, the block distortion being connected with linear noise are filtering out one time one by one, can remove in image so not
With the noise of distribution.
The apparent disease recognition of tunnel-liner:
The apparent Defect inspection identification model of tunnel-liner for designing fasterr-cnn, establishes sample using Disease Characters image
This library, builds deep learning frame, and training network model is returned using region motion network and bounding box and lost, realizes tunnel
The automatic identification in the apparent disease region of lining cutting, wherein it is as follows to return loss formula for bounding box:
In formula, NregIndicating the number of target frame, i indicates the index of anchors,Indicate corresponding GT (Ground
Turth probability), tiIndicate anchors for GT prediction drift amount,Indicate anchors for the true excursions amount of GT
It is lreg() indicates to return loss to the bounding box of single anchors.
Later using classification CNN network and total Classification Loss, the automatic classification to the apparent disease region of lining cutting is completed, wherein
The formula of Classification Loss is as follows:
N in formulaclsClassification number, piIndicate that anchors is the probability of target frame, lcls() is divided single target frame
Class loss.
Combining classification loses and bounding box returns loss and generates total losses function, so that optimizing network obtains maturation
Fasterr-cnn model, formula are as follows:
L({pi},{ti)=Lcls+λLreg
λ is constant for balancing two sub- loss functions in formula.
The fasterr-cnn model that training is completed solves traditional classification algorithm practical application dependent on manual and adjusts ginseng, because
And low efficiency, the problem of period is long and the poor robustness of detection method, realize full automatic trained detection process with quick and precisely
The apparent Defect inspection identification of high tunnel-liner with robustness.
The measurement of disease morphological parameters:
And for areal calculation, region-growing method can be used.The basic thought of region growing is that will have similar quality
Pixel set get up to constitute region.It refers specifically to first need the region divided that a sub-pixel conduct is looked for grow to each
Then point will have the pixel of same or similar property (pre-determined according to certain with sub-pixel in sub-pixel and neighborhood
Growth or similarity criterion determine) it is merged into the region where sub-pixel.These new pixels are continued as new seed
Process above, the pixel until not meeting condition can be included.Such a region is just grown into.Region growing
Generally comprise three important problems: selection, growth criterion and the termination condition of seed point.The growth chosen in this programme
Seed is single pixel point, is growth criterion with neighborhood territory pixel and the small Mr. Yu's threshold value of sub-pixel gray scale difference value, and termination condition is
No pixel or region meet the condition that growth district is added.Those error matching points be exactly when forming workpiece three-dimensional appearance extremely
The extreme point of spatial dimension where protruding from workpiece, so using algorithm of region growing by the similar region merging technique of property, thus
Reject those extreme points.
The key step of algorithm of region growing is as follows:
1. scanning to image sequence, the 1st pixel belonged to not yet is found, if the pixel is (x0, y0)；2. with (x0,
Y0 centered on), consider 4 neighborhood territory pixels of (x0, y0).If (x, y) meets growth criterion, (x0, y0) and (x, y) is merged
(in the same area), while will be pressed into storehouse；3. a pixel is taken out from storehouse, it when (x0, y0) makees back to step
Suddenly 2.；4. when storehouse is empty, 1. back to step；5. repeat step 1. -4. when meeting termination condition, growth terminates.
After generating 3D region, both the area of disease can be obtained according to three-dimensional Line Integral.
Finally, it is stated that the above examples are only used to illustrate the technical scheme of the present invention and are not limiting, although referring to compared with
Good embodiment describes the invention in detail, those skilled in the art should understand that, it can be to skill of the invention
Art scheme is modified or replaced equivalently, and without departing from the objective and range of the technical program, should all be covered in the present invention
Scope of the claims in.
Claims (4)
1. a kind of tunnel defect identifying system based on image characterized by comprising
Image capturing system is acquired the image of vcehicular tunnel；
Image preprocessing system receives the image for the vcehicular tunnel that described image acquisition system passes over, and to the figure after acquisition
As being pre-processed, increases the contrast in lining cutting apparent disease region and background area, obtain output image, to output image
Spotted noise, block distortion and linear noise section extract, then are filtered out；
Image segmentation processing system is carried out the image that noise was filtered out by described image pretreatment system using image segmentation
Binary conversion treatment；Image characteristic extraction system mentions the characteristic of described image segmentation processing system treated image
It takes, realizes the automatic identification in the apparent disease region of tunnel-liner；Disease Characters categorizing system is to described image Feature Extraction System
Disease geo-radar image after identification is classified；
Parameters Computing System carries out parameter calculating to the sorted image of the Disease Characters categorizing system, obtains in disease region
Disease morphological parameters.
2. a kind of tunnel defect identifying system based on image according to claim 1, it is characterised in that: described image feature mentions
It takes system to return loss with bounding box using region motion network to be identified, formula are as follows:
In formula, NregIndicating the number of target frame, i indicates the index of anchors,Indicate corresponding GT (Ground Turth)
Probability, tiIndicate anchors for GT prediction drift amount,Indicate that anchors is l for the true excursions amount of GTreg(·)
It indicates to return loss to the bounding box of single anchors.
The Disease Characters categorizing system is classified using classification CNN network with total Classification Loss, formula are as follows:
N in formulaclsClassification number, piIndicate that anchors is the probability of target frame, lcls() is the damage of the classification to single target frame
It loses.
3. a kind of tunnel defect identifying system based on image according to claim 1, it is characterised in that: described image is pre-
Processing system first counts the quantity of pixel in each connected region, given threshold T during extracting to spotted noiseo,
And to pixel quantity nkConnected domain less than the threshold value removes, and spotted noise extracts formula are as follows: Is(x, y)={ Ck(x,y)|
nk< To, k=1,2 ..., n }
In formula, Is(x,y),Ck(x, y) respectively indicates spotted noise and connected region.
Block distortion is removed using the method for calculating rectangular degree.Rectangular degree is defined as the total quantity of some connected region kind pixel
nkWith the smallest boundary rectangle area Sk=(yb-ya)*(xb-xaThe ratio between), (xa,ya),(xb,yb) respectively indicate the connected region
The pixel point set commercial affairs lower left corner and upper right angular coordinate.Utilize D=nk/SkRectangular degree is calculated, and utilizes formulas Extraction block distortion:
Is(x, y)={ Ck(x, y) | D < TD, k=1,2 ..., n }.
4. a kind of tunnel defect identifying system based on image according to claim 1, it is characterised in that: the parameter meter
Calculation system is when carrying out disease morphological parameters calculating in disease region, comprising the following steps:
D1: scanning image sequence, finds the 1st pixel belonged to not yet, if the pixel is (x0, y0)；
D2: centered on (x0, y0), consider 4 neighborhood territory pixels of (x0, y0)；If (x, y) meets growth criterion, by (x0, y0)
Merge (in the same area) with (x, y), while will be pressed into storehouse；
D3: taking out a pixel from storehouse, it is returned to step D2 as (x0, y0)；
D4: when storehouse is empty, step D1 is returned to；
D5: step D1-D4 is repeated when meeting termination condition, growth terminates.It, both can be according to three-dimensional after generating 3D region
Line Integral obtains the area of disease.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910394410.1A CN110111322A (en) | 2019-05-13 | 2019-05-13 | A kind of tunnel defect identifying system based on image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910394410.1A CN110111322A (en) | 2019-05-13 | 2019-05-13 | A kind of tunnel defect identifying system based on image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110111322A true CN110111322A (en) | 2019-08-09 |
Family
ID=67489732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910394410.1A Pending CN110111322A (en) | 2019-05-13 | 2019-05-13 | A kind of tunnel defect identifying system based on image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110111322A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126415A (en) * | 2019-12-12 | 2020-05-08 | 创新奇智(合肥)科技有限公司 | Tunnel steel bar detection counting system and method based on radar detection image |
| CN111369515A (en) * | 2020-02-29 | 2020-07-03 | 上海交通大学 | Tunnel water stain detection system and method based on computer vision |
| CN111562260A (en) * | 2020-04-14 | 2020-08-21 | 江苏大学 | Lotus root mud hole detection method and device based on machine vision |
| CN111882664A (en) * | 2020-07-08 | 2020-11-03 | 武汉武大卓越科技有限责任公司 | Multi-window accumulated difference crack extraction method |
| CN117514352A (en) * | 2023-11-17 | 2024-02-06 | 湖南科技大学 | Colliery water damage early warning system based on video identification |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
| CN104021574A (en) * | 2014-07-04 | 2014-09-03 | 武汉武大卓越科技有限责任公司 | Method for automatically identifying pavement diseases |
| CN104504388A (en) * | 2014-12-17 | 2015-04-08 | 长安大学 | Pavement crack identification and feature extraction algorithm and system |
| CN105719259A (en) * | 2016-02-19 | 2016-06-29 | 上海理工大学 | Pavement crack image detection method |
| CN106841216A (en) * | 2017-02-28 | 2017-06-13 | 浙江工业大学 | Tunnel defect automatic identification equipment based on panoramic picture CNN |
| CN108918539A (en) * | 2018-07-27 | 2018-11-30 | 同济大学 | A kind of apparent Defect inspection device and method of tunnel structure |
-
2019
- 2019-05-13 CN CN201910394410.1A patent/CN110111322A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
| CN104021574A (en) * | 2014-07-04 | 2014-09-03 | 武汉武大卓越科技有限责任公司 | Method for automatically identifying pavement diseases |
| CN104504388A (en) * | 2014-12-17 | 2015-04-08 | 长安大学 | Pavement crack identification and feature extraction algorithm and system |
| CN105719259A (en) * | 2016-02-19 | 2016-06-29 | 上海理工大学 | Pavement crack image detection method |
| CN106841216A (en) * | 2017-02-28 | 2017-06-13 | 浙江工业大学 | Tunnel defect automatic identification equipment based on panoramic picture CNN |
| CN108918539A (en) * | 2018-07-27 | 2018-11-30 | 同济大学 | A kind of apparent Defect inspection device and method of tunnel structure |
Non-Patent Citations (4)
| Title |
|---|
| CE LI等: "Tunnel crack detection using coarse-to-fine region localization and edge detection", 《WIRES》, 3 April 2019 (2019-04-03), pages 3 * |
| 赵小川编著: "《MATLAB图像处理 能力提高与应用案例》", 31 January 2014, 北京航空航天大学出版社, pages: 87 * |
| 高泰: "隧道裂缝图像采集与检测技术研究", 《中国优秀硕士学位论文全文数据库》, 15 June 2018 (2018-06-15), pages 49 - 59 * |
| 黄宏伟等著: "《城市地铁盾构隧道病害快速检测与工程实践》", 31 January 2019, 上海科学技术出版社, pages: 205 - 206 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126415A (en) * | 2019-12-12 | 2020-05-08 | 创新奇智(合肥)科技有限公司 | Tunnel steel bar detection counting system and method based on radar detection image |
| CN111126415B (en) * | 2019-12-12 | 2022-07-08 | 创新奇智(合肥)科技有限公司 | Tunnel steel bar detection counting system and method based on radar detection image |
| CN111369515A (en) * | 2020-02-29 | 2020-07-03 | 上海交通大学 | Tunnel water stain detection system and method based on computer vision |
| CN111562260A (en) * | 2020-04-14 | 2020-08-21 | 江苏大学 | Lotus root mud hole detection method and device based on machine vision |
| CN111882664A (en) * | 2020-07-08 | 2020-11-03 | 武汉武大卓越科技有限责任公司 | Multi-window accumulated difference crack extraction method |
| CN117514352A (en) * | 2023-11-17 | 2024-02-06 | 湖南科技大学 | Colliery water damage early warning system based on video identification |
| CN117514352B (en) * | 2023-11-17 | 2024-05-07 | 湖南科技大学 | Colliery water damage early warning system based on video identification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110111322A (en) | A kind of tunnel defect identifying system based on image | |
| Pape et al. | 3-D histogram-based segmentation and leaf detection for rosette plants | |
| CN109636784B (en) | Image saliency target detection method based on maximum neighborhood and super-pixel segmentation | |
| WO2021109697A1 (en) | Character segmentation method and apparatus, and computer-readable storage medium | |
| CN109816644A (en) | A kind of bearing defect automatic checkout system based on multi-angle light source image | |
| CN107220649A (en) | A kind of plain color cloth defects detection and sorting technique | |
| CN105550660B (en) | A kind of recognition methods of woven fabric structure structure species | |
| CN110827235B (en) | Steel plate surface defect detection method | |
| CN110245593A (en) | A kind of images of gestures extraction method of key frame based on image similarity | |
| CN108829711B (en) | Image retrieval method based on multi-feature fusion | |
| CN114118144A (en) | Anti-interference accurate aerial remote sensing image shadow detection method | |
| Ji et al. | Target recognition method of green pepper harvesting robot based on manifold ranking | |
| CN110687121B (en) | Intelligent online detection and automatic grading method and system for ceramic tiles | |
| CN105719275A (en) | Parallel combination image defect segmentation method | |
| CN109871900A (en) | The recognition positioning method of apple under a kind of complex background based on image procossing | |
| CN109781737A (en) | A kind of detection method and its detection system of hose surface defect | |
| CN111080696A (en) | Underwater sea cucumber identification and positioning method based on computer vision | |
| CN104123554A (en) | SIFT image characteristic extraction method based on MMTD | |
| CN109886932A (en) | Gear ring of wheel speed sensor detection method of surface flaw based on SVM | |
| CN116704516B (en) | Visual inspection method for water-soluble fertilizer package | |
| CN110473174A (en) | A method of pencil exact number is calculated based on image | |
| CN110176020A (en) | A kind of bird's nest impurity method for sorting merging 2D and 3D rendering | |
| CN109961432A (en) | A kind of detection method and system of filter cloth breakage | |
| CN111738931A (en) | Shadow removal algorithm for aerial image of photovoltaic array unmanned aerial vehicle | |
| CN110648312A (en) | Method for identifying wool and cashmere fibers based on scale morphological characteristic analysis |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190809 |