CN111242851B - Concrete structure surface crack detection method and system - Google Patents
Concrete structure surface crack detection method and system Download PDFInfo
- Publication number
- CN111242851B CN111242851B CN202010276561.XA CN202010276561A CN111242851B CN 111242851 B CN111242851 B CN 111242851B CN 202010276561 A CN202010276561 A CN 202010276561A CN 111242851 B CN111242851 B CN 111242851B
- Authority
- CN
- China
- Prior art keywords
- crack
- image
- fused
- calculating
- coordinate
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 104
- 238000000034 method Methods 0.000 claims abstract description 50
- 230000009466 transformation Effects 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 16
- 230000004927 fusion Effects 0.000 claims description 10
- 238000012935 Averaging Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000007774 longterm Effects 0.000 abstract description 3
- 230000005540 biological transmission Effects 0.000 description 20
- 238000012544 monitoring process Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 7
- 239000013589 supplement Substances 0.000 description 7
- 230000001413 cellular effect Effects 0.000 description 4
- 230000006378 damage Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 241000264877 Hippospongia communis Species 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- 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
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/32—Indexing scheme for image data processing or generation, in general involving image mosaicing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a concrete structure surface crack detection method and system, and relates to the technical field of concrete crack detection. The method comprises the following steps: acquiring area images of a plurality of areas on the surface of a concrete structure to be measured, which are shot by an industrial camera; splicing and fusing the area images to obtain a panoramic image; carrying out crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain cracks in the panoramic image; and calculating the coordinates, width and length of the crack and storing the coordinates, width and length. According to the method, the regional image of the concrete structure surface to be detected by the industrial camera is adopted, the long-distance and long-term detection of the width, the length and the distribution of the crack in a certain regional range of the structure surface is realized, the regional image acquired by the industrial camera with low resolution is utilized, the crack detection algorithm based on the seepage model is utilized to carry out crack detection on the panoramic image, the tiny cracks with the size larger than 1/5 pixels can be detected, and the crack detection resolution is greatly improved.
Description
Technical Field
The invention relates to the technical field of concrete crack detection, in particular to a method and a system for detecting a surface crack of a concrete structure.
Background
In the concrete structure disease cracks, honeycombs, pitted surfaces, pits and other modes, the cracks are one of the serious damages and the serious threats. Taking a bridge structure as an example, relevant literature data show that the breakage of the concrete structure bridge caused by cracks accounts for more than 90% of the total number of broken bridges. The cracks have great harm to the concrete bridge, and particularly easily cause safety accidents such as bridge collapse and the like. Based on this, people pay more and more attention to the regular detection and maintenance work of bridge cracks.
The most traditional method for detecting the bridge cracks is manual detection, but the manual detection mode is easily influenced by human subjective factors, the measurement result is unstable, and the precision and the efficiency are low. And for places which cannot be reached by manpower, a bridge operation detection vehicle is required to be arranged. The bridge detection operation vehicle is high in cost and low in detection efficiency.
The existing crack detection method based on the image is mainly based on the whole pixel level processing process, fine cracks at a sub-pixel level cannot be identified, and the detection precision is low. Although some crack detection methods can perform sub-pixel positioning on the edge of the crack and the crack width detection precision can reach the level of sub-pixels, the method is only suitable for cracks with larger width. If the crack width is less than 1 pixel, the method may not recognize the crack or the detection result of the crack width may become unstable even if the crack can be recognized. Therefore, the existing method for detecting the cracks has the problems of low adaptability and low detection precision.
Disclosure of Invention
The invention aims to provide a method and a system for detecting a surface crack of a concrete structure, which solve the problems of low adaptability and low detection precision of the existing method for detecting the crack.
In order to achieve the purpose, the invention provides the following scheme:
a concrete structure surface crack detection method comprises the following steps:
acquiring area images of a plurality of areas on the surface of a concrete structure to be measured, which are shot by an industrial camera; partial images of the area images of the adjacent area on the surface of the concrete structure to be detected are overlapped;
splicing and fusing the area images to obtain a panoramic image;
carrying out crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain cracks in the panoramic image;
and calculating the coordinates, the width and the length of the crack, and storing the coordinates, the width and the length of the crack.
Optionally, the splicing and fusing the area images to obtain a panoramic image specifically includes:
performing image enhancement processing on the region image by adopting a single-scale Retinex algorithm to obtain an enhanced region image;
taking the reinforced area image of any area of the surface of the concrete structure to be detected as a first image to be fused;
taking the enhanced region image adjacent to the first image to be fused as a second image to be fused;
extracting a common feature point in an overlapping pixel point of an overlapping part of the first image to be fused and the second image to be fused by adopting a scale invariant feature transformation algorithm to obtain a first coordinate of the common feature point in the first image to be fused and a second coordinate of the common feature point in the second image to be fused respectively; each enhanced region image comprises overlapped pixel points and residual pixel points;
establishing a coordinate transformation model according to the first coordinate and the second coordinate;
calculating the coordinates of the residual pixel points of the second image to be fused in the coordinate system of the first image to be fused by using the coordinate transformation model, and calculating the gray value of the overlapped pixel points by using a direct averaging method to obtain a spliced and fused image A';
and replacing the first image to be fused with the spliced and fused image A', returning to the step of taking the enhanced region image of the adjacent region of the first image to be fused as a second image to be fused, and traversing all the enhanced region images to obtain a panoramic image.
Optionally, the performing crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain cracks in the panoramic image specifically includes:
extracting cracks in the panoramic image by using a crack detection algorithm based on a seepage model to obtain a crack image;
performing expansion processing on the crack image to obtain an expansion image and a crack area in the expansion image;
thinning the expansion image by adopting a thinning algorithm, extracting a crack framework in the crack area, and obtaining the crack framework, crack framework pixel points and a thinned image;
carrying out gray value combination according to the refined image, and uniformly adding the gray value of a non-crack skeleton pixel point in the expanded image to a crack skeleton pixel point adjacent to the non-crack skeleton pixel point to obtain a gray value combination value of the crack skeleton image and the crack skeleton pixel point;
and calculating the starting point, the end point and the bifurcation point of the crack according to the crack skeleton image to obtain the crack in the panoramic image, and storing the starting point, the end point and the bifurcation point of the crack.
Optionally, the calculating the coordinates, the width and the length of the crack and storing the coordinates, the width and the length of the crack specifically include:
calculating the center of mass of the crack region, wherein the coordinate of the center of mass of the crack region is the coordinate of the crack;
acquiring a plurality of crack standard lines and standard widths of the crack standard lines;
sequentially performing expansion treatment, thinning treatment and grey value combination on the crack standard line to obtain a combined grey value of each pixel point of the crack standard line;
calculating the average value of all the merged gray values, and determining the average value as the gray merged value of the crack standard line;
calculating by using the standard width and the gray level combination value of the crack standard line through straight line fitting to obtain a crack calibration coefficient;
calculating to obtain the width of the crack by utilizing the gray combining value of the crack skeleton pixel point and the crack calibration coefficient;
calculating to obtain the length of the crack by using the crack skeleton pixel point and the crack calibration coefficient;
storing the coordinates, the width, the length, and the crack.
A concrete structure surface crack detection system comprising:
the area image module is used for acquiring area images of a plurality of areas on the surface of the concrete structure to be measured, which are shot by an industrial camera; partial images of the area images of the adjacent area on the surface of the concrete structure to be detected are overlapped;
the panoramic image module is used for splicing and fusing the area images to obtain a panoramic image;
the crack module is used for carrying out crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain cracks in the panoramic image;
and the calculation module is used for calculating the coordinates, the width and the length of the crack and storing the coordinates, the width and the length of the crack.
Optionally, the panoramic image module specifically includes:
the enhancement unit is used for carrying out image enhancement processing on the regional image by adopting a single-scale Retinex algorithm to obtain an enhanced regional image;
the first image unit to be fused is used for taking the image of the enhanced area of any area of the surface of the concrete structure to be measured as a first image to be fused;
the second image unit to be fused is used for taking the enhanced region image of the adjacent region of the first image to be fused as a second image to be fused;
the common feature point unit is used for extracting common feature points in overlapping pixel points of the overlapping parts of the first image to be fused and the second image to be fused by adopting a scale-invariant feature transformation algorithm to obtain a first coordinate of the common feature points in the first image to be fused and a second coordinate of the common feature points in the second image to be fused; each enhanced region image comprises overlapped pixel points and residual pixel points;
the coordinate transformation model unit is used for establishing a coordinate transformation model according to the first coordinate and the second coordinate;
the splicing fusion image unit is used for calculating the coordinates of the residual pixel points of the second image to be fused in the coordinate system of the first image to be fused by using the coordinate transformation model, and calculating the gray value of the overlapped pixel points by using a direct averaging method to obtain a splicing fusion image A';
and the panoramic image unit is used for replacing the first image to be fused with the spliced and fused image A', executing a second image unit to be fused, and traversing all the images in the enhanced area to obtain a panoramic image.
Optionally, the fracture module specifically includes:
the crack image unit is used for extracting cracks in the panoramic image by using a crack detection algorithm based on a seepage model to obtain a crack image;
the expansion unit is used for performing expansion processing on the crack image to obtain an expansion image and a crack area in the expansion image;
the thinning unit is used for thinning the expansion image by adopting a thinning algorithm, extracting a crack framework in the crack area and obtaining the crack framework, crack framework pixel points and a thinned image;
the merging unit is used for performing gray value merging according to the refined image, uniformly adding the gray value of the non-crack skeleton pixel point in the expanded image to the crack skeleton pixel point adjacent to the non-crack skeleton pixel point, and obtaining the gray value merging value of the crack skeleton image and the crack skeleton pixel point;
and the calculation unit is used for calculating the starting point, the end point and the bifurcation point of the crack according to the crack skeleton image, obtaining the crack in the panoramic image and storing the starting point, the end point and the bifurcation point of the crack.
Optionally, the calculation module specifically includes:
the coordinate unit is used for calculating the center of mass of the crack region, and the coordinate of the center of mass of the crack region is the coordinate of the crack;
the standard width unit is used for acquiring a plurality of crack standard lines and the standard width of the crack standard lines;
the combined gray value unit is used for sequentially performing expansion treatment, thinning treatment and gray value combination on the crack standard line to obtain a combined gray value of each pixel point of the crack standard line;
the gray level combination value unit is used for calculating the average value of all the combination gray levels and determining the average value as the gray level combination value of the crack standard line;
the crack calibration coefficient unit is used for calculating a crack calibration coefficient through straight line fitting by using the standard width and the gray level combination value of the crack standard line;
the width unit is used for calculating the width of the crack by utilizing the gray combination value of the crack skeleton pixel point and the crack calibration coefficient;
the length unit is used for calculating the length of the crack by using the crack skeleton pixel point and the crack calibration coefficient;
a storage unit for storing the coordinates, the width, the length, and the crack.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a method and a system for detecting surface cracks of a concrete structure. The method comprises the following steps: acquiring area images of a plurality of areas on the surface of a concrete structure to be measured, which are shot by an industrial camera; partial images of regional images of adjacent regions on the surface of the concrete structure to be detected are overlapped; splicing and fusing the area images to obtain a panoramic image; carrying out crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain cracks in the panoramic image; the coordinates, width and length of the crack are calculated and stored. According to the method, the industrial camera is adopted to obtain the area image of the surface of the concrete structure to be detected, so that the long-distance and long-term detection of the width, the length and the distribution of the cracks in a certain area range of the surface of the structure is realized, the area image collected by the industrial camera with low resolution is utilized, the crack detection algorithm based on the seepage model is utilized to carry out crack detection on the panoramic image, the tiny cracks with the size larger than 1/5 pixels can be detected, and the crack detection resolution is greatly improved; meanwhile, the high-precision crack width recognition is realized by using the industrial camera with lower resolution, the crack detection cost is greatly reduced, and the method is suitable for the whole process detection and early warning of the concrete structure cracks of bridges, tunnels, dams, buildings and the like from the initiation to the expansion endangering the structure safety.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
Fig. 1 is a flowchart illustrating a method for detecting cracks on a surface of a concrete structure according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of gray level value merging provided in the embodiment of the present invention;
FIG. 3 is an image of a fracture skeleton provided by an embodiment of the present invention;
FIG. 4 is a schematic view of a fracture standard line provided by an embodiment of the present invention;
fig. 5 is a structural view of a concrete structure surface crack detecting apparatus according to an embodiment of the present invention;
fig. 6 is a structural diagram of a field image collecting and analyzing sub-apparatus according to an embodiment of the present invention;
fig. 7 is a system diagram of a system for detecting cracks on the surface of a concrete structure according to an embodiment of the present invention.
Wherein, 1, a field image acquisition and analysis sub-device; 2. a 4G base station; 3. a 4G cellular network; 4. the internet; 5. a monitoring center; 6. a light supplement lamp; 7. an embedded industrial personal computer; 8. a two-dimensional electric pan-tilt; 9. the 4G wireless data remote transmission module; 10. an industrial camera; 20. a power supply module; 11. a first slit; 12. a second split; 13. and a third crack.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a method and a system for detecting a surface crack of a concrete structure, which solve the problems of low adaptability and low detection precision of the existing method for detecting the crack.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
The present embodiment provides a method for detecting a crack on a surface of a concrete structure, and fig. 1 is a flowchart of the method for detecting a crack on a surface of a concrete structure according to the present embodiment. Referring to fig. 1, the method for detecting cracks on the surface of a concrete structure includes:
And 102, splicing and fusing the area images to obtain a panoramic image, wherein in order to ensure the detection resolution, the area range of the surface of the concrete structure to be detected is 2m × 2m, and the panoramic image obtained through splicing and fusing is the surface image of the concrete structure to be detected, wherein the surface image is not more than 4 square meters.
Step 102 specifically includes: and performing image enhancement processing on the regional image by adopting a Single Scale Retinex (SSR) algorithm to obtain an enhanced regional image. The SSR algorithm removes illumination components in the area image so as to reduce the influence of uneven illumination on the detection precision.
Taking the enhanced area image of any area of the surface of the concrete structure to be detected as a first image to be fused.
And taking the enhanced region image adjacent to the first image to be fused as a second image to be fused.
Extracting common feature points in overlapped pixel points of the overlapped parts of the first image to be fused and the second image to be fused by adopting a Scale-invariant feature transform (SIFT) algorithm to obtain a first coordinate of the common feature points in the first image to be fused and a second coordinate of the common feature points in the second image to be fused; each enhanced region image comprises an overlapped part and a non-overlapped part, pixel points of the overlapped part are overlapped pixel points, and the rest pixel points are pixel points except the overlapped pixel points in the enhanced region image, namely each enhanced region image comprises the overlapped pixel pointsAnd the public characteristic points are characteristic pixel points in the enhanced region image overlapping pixel points of any two adjacent regions extracted by the SIFT algorithm. Enhanced region image A with 1 st region1And enhanced region image A of 2 nd region2The description is given for the sake of example: respectively extracting enhanced region images A by adopting SIFT algorithm1And enhanced region image A2Of the overlapping portion of1、T2、…、TjAnd j represents the number of common feature points in the enhanced region image A1The first coordinate of (A) isx 1，y 1）、（x 2，y 2）、…、（x j，y j) Common feature point in enhanced region image A2The second coordinate of (A) is
、
、…、
。
And establishing a coordinate transformation model according to the first coordinate and the second coordinate. The method specifically comprises the following steps:
and (3) establishing a coordinate transformation model according to the first coordinate and the second coordinate by adopting a formula (1):
in the formula (1), the first and second groups,m i （i=0, 1, …, 7) represents an element to be solved; (x，y) And
respectively represent the same common feature point in the enhanced region image A1And enhanced region image A2The first coordinate and the second coordinate.
Will coordinate (a)x 1，y 1）、（x 2，y 2）、…、（x j，y j) And
、
、…、
respectively substituted into the formula (1) to obtain an equation set, solving the equation set by adopting a Levenberg-Marquardt (L evenberg-Marquardt, L-M) method, and calculating to obtain the equation setm i （iValue of =0, 1, …, 7), according tom i （iValues of =0, 1, …, 7) to obtain a model of the established coordinate transformation.
And calculating the coordinates of the residual pixel points of the second image to be fused in the coordinate system of the first image to be fused by using a coordinate transformation model, and calculating the gray value of the overlapped pixel points by using a direct averaging method to obtain a spliced and fused image A'. The method specifically comprises the following steps:
and converting the coordinates of the residual pixel points in the second image to be fused into the coordinates in the coordinate system of the first image to be fused according to the established coordinate transformation model, averaging the gray value of the overlapped pixel points in the second image to be fused and the gray value of the overlapped pixel points in the first image to be fused by using a direct averaging method to obtain the gray value of the overlapped pixel points, and further obtaining the spliced fused image A'. The principle of fusion by direct averaging is: and directly adding the gray values of the two images and dividing by 2 to obtain the gray value of the spliced fusion image. And coordinates of the overlapped pixel points and the residual pixel points are positions of the pixel points in the enhanced region image, and a coordinate system of the enhanced region image is established according to the positions of the pixel points.
Replacing the first image to be fused with the spliced and fused image A', returning to the step of taking the enhanced region image adjacent to the first image to be fused as the second image to be fused, and traversing all the enhanced region imagesAnd splicing and fusing all the enhanced region images to obtain a panoramic image. In particular to enhance the regional image A1And enhanced region image A2Splicing and fusing the image A ' into a spliced and fused image A ', and then combining the spliced and fused image A ' with the enhanced region image A of the 3 rd region3Splicing and fusing are carried out, and the spliced and fused image A' and the enhanced region image A are used3The spliced and fused image replaces and updates the spliced and fused image A ', and the content of the updated spliced and fused image A' comprises the enhanced region image A1Enhanced region image A2And enhanced region image A3Then the updated spliced fusion image A' and the enhanced area image A of the 4 th area are used4And performing splicing fusion, and repeating the steps until all the images in the enhancement area are spliced and fused into a panoramic image. The 1 st area and the 2 nd area are two adjacent areas, the 3 rd area and the 1 st area (or the 2 nd area) are two adjacent areas, and the 4 th area and the 1 st area (or the 2 nd area or the 3 rd area) are two adjacent areas.
And 103, carrying out crack detection on the panoramic image by using a crack detection algorithm based on the seepage model to obtain cracks in the panoramic image.
Step 103 specifically comprises: and extracting the cracks in the panoramic image by using a crack detection algorithm based on a seepage model to obtain a crack image.
And performing expansion processing on the crack image to obtain an expansion image and a crack area in the expansion image. The method specifically comprises the following steps:
scanning each pixel point in the crack image by using a 3 × template, performing or calculating the pixel points covered by the 3 × template and the 3 × template to obtain the expanded pixel point of each pixel point, and obtaining an expanded image according to the expanded pixel point of each pixel point, wherein the expanded pixel point has the value of 0 or the 1.3 × template is a region of 3 pixels × pixels, and performing or calculating 9 pixel points in the 3 × template to obtain the expanded pixel point of each pixel point, wherein P = P1 | | P2 | | P3 | | P4 | | | P5| | | | P6 | P7 | P8 | P9, P represents the expanded pixel point, P1, P2, P3, P4, P5, P6, P7, P8 and P9 respectively represent 9 pixel points in 3 crack processing, so that a crack in a circle of the expansion region is extracted, and a crack is slightly expanded in the crack processing process, and a crack is partially lost in a crack is compensated.
And thinning the crack region in the expansion image by adopting a thinning algorithm, extracting the crack skeleton in the crack region, and obtaining the crack skeleton, crack skeleton pixel points and a thinned image. The method specifically comprises the following steps:
and extracting the skeleton of each crack region in the expansion image by adopting a thinning algorithm to obtain crack skeleton pixel points, and deleting non-crack skeleton pixel points to obtain crack skeletons, crack skeleton pixel points and thinned images after thinning treatment. The crack skeleton is the axis of the crack. Each location on the central axis is only one pixel wide.
And combining the gray values according to the thinned image, and uniformly adding the gray values of the non-crack skeleton pixel points in the crack area in the expanded image to the crack skeleton pixel points adjacent to the non-crack skeleton pixel points to obtain the gray value combination values of the crack skeleton image and the crack skeleton pixel points. The method specifically comprises the following steps:
according to a crack skeleton in a refined image and a crack region in an expanded image, scanning each non-crack skeleton pixel point in each crack region in a panoramic image by using a 3 × template to obtain a gray value of the non-crack skeleton pixel point, and uniformly dispersing the gray value of the non-crack skeleton pixel point to an adjacent crack skeleton pixel point of the non-crack skeleton pixel point to obtain a gray value combination value of the crack skeleton pixel point and a crack skeleton image fig. 2 is a gray value combination schematic diagram provided by the embodiment of the invention, referring to fig. 2, uniformly dispersing the gray value of the non-crack skeleton pixel point P5 to adjacent crack skeleton pixel points P6, P7, P8 and P9, before gray value combination, the gray values of the pixel points P5, P6, P7, P8 and P9 are 120, 180, 70, 100 and 200 respectively, after gray value combination, the gray values of the pixel points P5, P6, P7, P8 and P9 are 0, 210, 100, 130, P230, P1, P3 and the gray values of the non-crack skeleton pixel points are 3 and 3 in the crack skeleton template 3 and the central point 3 and the non-crack skeleton pixel point 3 are respectively, 3 and the central point 36.
The crack starting point, the crack ending point and the branch point are calculated according to the crack skeleton image, the crack in the panoramic image is obtained, the crack starting point, the crack ending point and the branch point are stored in a crack database, the branch point is calculated by scanning a refined image through a 3 × 3 template, the number of pixels with the median value of 1 in 9 scanned pixels is judged, if the number of the pixels with the median value of 1 is larger than 4, the pixels of the center points of corresponding areas of the 9 scanned pixels are the branch point, fig. 3 is the crack skeleton image provided by the embodiment of the invention, by taking fig. 3 as an example, referring to fig. 3, the starting point of a first crack 11 is a pixel point A, the ending point is a pixel point B, the starting point of a second crack 12 is a pixel point B, the ending point is a pixel point C, the starting point of a third crack 13 is a pixel point B, the ending point is a pixel point D, in a 3 × 3 template with the pixel point B as the center point, the number of the pixels with the median value of 1 is larger than 4, and the pixel point B is a second crack skeleton image, and the starting point or the ending point is 3 × 3, and the value of each scanned pixel point is defined as another pixel point or the starting point.
And 104, calculating the coordinates, the width and the length of the crack, and storing the coordinates, the width and the length of the crack.
Step 104 specifically includes:
and calculating the center of mass of the crack region, wherein the coordinate of the center of mass of the crack region is the coordinate of the crack. In the embodiment, the centroid of the crack region in the expansion image is calculated, and the coordinate (x) of the centroid in the expansion image is calculatedc，yc) The crack coordinates.
Obtaining a plurality of crack standard lines and standard widths of the crack standard lines: and attaching the crack calibration standard line to the surface of the concrete structure to be detected. Fig. 4 is a schematic view of crack standard lines provided in an embodiment of the present invention, where the number of the crack calibration standard lines in the embodiment is 15, the number of the crack calibration standard lines in the embodiment is shown in fig. 4, and the standard widths of the crack standard lines in fig. 4 are a1=0.05mm、a2=0.06mm、a3=0.07mm、a4=0.08mm、a5=0.09mm、a6=0.1mm、a7=0.2mm、a8=0.3mm、a9=0.4mm、a10=0.5mm、a11=0.6mm、a12=0.7mm、a13=0.8mm、a14=0.9mm、a15=1.00mm, and the lengths are all 40 mm. Images of a plurality of crack standard lines photographed by an industrial camera are acquired.
And sequentially performing expansion treatment, thinning treatment and grey value combination on the crack standard line to obtain a combined grey value of each pixel point of the crack standard line. The method specifically comprises the following steps: and sequentially performing expansion treatment, thinning treatment and gray value combination on the shot image comprising the plurality of crack standard lines to obtain a combined gray value of each pixel point of each crack standard line.
Calculating the average value of all the merged gray values, and determining the average value as the gray merged value of the crack standard line; the method specifically comprises the following steps: calculating an average value of all the merged gray values of one crack standard line, determining the average value as a gray merged value of the crack standard line, and calculating the gray merged values of all the crack standard lines according to the above steps, wherein the gray merged value of 15 crack standard lines is b1、b2、…、b15。
And calculating to obtain the fracture calibration coefficient by straight line fitting according to the standard width and the gray level combination value of the fracture standard line. The method specifically comprises the following steps: combining the standard width of each crack standard line with the gray level combination value of the corresponding crack standard line to form a group of data to obtain (a)1，b1）、（a2，b2）、…、（a15，b15) And (3) performing linear fitting on 15 groups of data according to a formula (2) to obtain the slope of the straight line, namely the crack calibration coefficient k.
a=kb （2）；
In the formula (2), a represents a standard width; k represents a crack calibration coefficient; and b represents a gray level merged value of the crack standard line. The crack standard line is only used for calibrating a k value, namely, is used for straight line fitting; the more the number of the crack standard lines is, the more the data used for straight line fitting is, and the more accurate the crack calibration coefficient k is.
And calculating to obtain the width of the crack by utilizing the gray combined value and the crack calibration coefficient of the crack skeleton pixel point. The width = the gray-scale combination value of the crack skeleton pixel point and the crack calibration coefficient k, and the width of the position of each crack skeleton pixel point on the crack can be obtained by multiplying the gray-scale combination value of each crack skeleton pixel point by the crack calibration coefficient k, so that the widths of different positions of the crack can be obtained.
And calculating to obtain the length of the crack by using the crack skeleton pixel point and the crack calibration coefficient. The crack length = total number of pixels in the crack skeleton and crack calibration coefficient k, and the length of one crack is equal to the product of the total number of pixels in the crack skeleton of the crack and the crack calibration coefficient.
The coordinates, width, length, and crack are stored. The method specifically comprises the following steps: the coordinates, width, length, start point of the fracture, end point of the fracture and bifurcation point of the fracture are stored in a fracture database.
The method for detecting the surface crack of the concrete structure can be used for monitoring whether the crack on the surface of the concrete structure is expanded or whether the surface of the concrete structure is newly cracked, and specifically comprises the following steps:
and updating the area images of a plurality of areas on the surface of the concrete structure to be detected, which are shot by the industrial camera, and updating the cracks according to the concrete structure surface crack detection method, wherein the updated cracks comprise the coordinates, the width and the length of the cracks.
Comparing the updated crack with the crack stored in the crack database, and judging whether the crack is expanded or whether a new crack exists, specifically comprising the following steps:
and matching the updated crack with the stored crack to obtain the stored crack corresponding to the updated crack.
And respectively subtracting the updated width and length of the crack from the corresponding stored width and length of the crack, judging whether the crack is expanded, if the difference is greater than a preset expansion threshold, judging that the crack is expanded beyond the limit, and performing expansion over-limit alarm. Or directly comparing the updated width and length of the crack with a preset width threshold value and a preset length threshold value, if the updated width of the crack is larger than the preset width threshold value or the updated length of the crack is larger than the preset length threshold value, judging that the crack is expanded beyond the limit, and performing expansion over-limit alarm.
And (3) making a difference between the updated cracks and the stored cracks, wherein the coordinates of the cracks are the mass center of the crack area, so that each crack only has one coordinate, calculating the number of the updated coordinates, making a difference between the number of the updated coordinates and the number of the stored crack coordinates to obtain the number of the new cracks, and if the number of the new cracks is more than or equal to 1, judging that the new cracks appear. After matching the updated fracture with the stored fractures, the remaining updated fractures are determined to be new fractures.
The updated fracture is stored to a fracture database.
And if the number of the new cracks is larger than a preset new crack threshold value, judging that the new cracks are out of limit, and alarming the new cracks.
The present embodiment provides a concrete structure surface crack detection apparatus, fig. 5 is a structural diagram of the concrete structure surface crack detection apparatus provided in the embodiment of the present invention, and referring to fig. 5, the concrete structure surface crack detection apparatus includes: the system comprises a field image acquisition and analysis sub-device 1, a data remote transmission sub-device and a monitoring center 5.
The field image acquisition and analysis sub-device 1 is connected with a data remote transmission sub-device, and the data remote transmission sub-device is connected with a monitoring center 5. And the field image acquisition and analysis sub-device transmits the data to be analyzed and detected to the monitoring center through the data remote transmission sub-device. The number of the on-site image acquisition and analysis sub-device 1 and the number of the data remote transmission sub-devices can be multiple.
The field image acquisition and analysis sub-device is used for acquiring an area image of the surface of the concrete structure and carrying out crack detection on the area image to obtain detection data.
Fig. 6 is a structural diagram of a field image collecting and analyzing sub-apparatus according to an embodiment of the present invention, and arrows in fig. 5 and 6 indicate a data transmission direction. Referring to fig. 6, the field image collecting and analyzing sub-apparatus includes: the device comprises an industrial camera 10, a light supplement lamp 6, a mobile platform, a control module, a data transmission module and a power supply module 20. The mobile platform adopts a two-dimensional electric pan-tilt 8, the control module adopts an embedded industrial personal computer 7, and the data transmission module adopts a 4G wireless data remote transmission module 9.
The industrial camera is used for shooting area images of a plurality of areas on the surface of the concrete structure to be measured.
The industrial camera 10 is electrically connected to a control module, and the control module is configured to output an instruction to control the industrial camera to capture an area image and acquire the captured area image.
The light filling lamp is used for illuminating the surface of the concrete structure to be detected.
The light supplement lamp 6 is electrically connected with the control module, and the control module is used for outputting an instruction to control the light supplement lamp to illuminate the surface of the concrete structure to be tested.
Light filling lamp 6 and industrial camera 10 all are fixed in on the moving platform, and light filling lamp 6 is fixed in around the industrial camera 10, and preferred light filling lamp 6 and industrial camera 10 are fixed in on the electronic cloud platform 8 of two-dimentional side by side.
The two-dimensional electric holder 8 is electrically connected with the control module and used for changing the horizontal angle and the pitching angle according to the instruction of the control module and driving the light supplementing lamp and the industrial camera to be aligned to any region of the surface of the concrete structure to be detected. The control module is used for outputting an instruction to control the two-dimensional electric holder to change the horizontal angle and the pitching angle.
The embedded industrial personal computer 7 is also used for detecting cracks of the shot area images by adopting a concrete structure surface crack detection method to obtain detection data, and sending the detection data to the data remote transmission sub-device through the 4G wireless data remote transmission module so as to transmit the detection data to the monitoring center. The inspection data includes crack information of the surface of the concrete structure, the crack information including coordinates, width and length of the crack. The embedded industrial personal computer 7 is also used for updating detection data according to the detection instruction, monitoring whether cracks on the surface of the concrete structure expand or whether new cracks on the surface of the concrete structure grow according to the updated detection data, and transmitting a monitoring result to the monitoring center. The monitoring result comprises extension overrun alarm information and new crack overrun alarm information, wherein the extension overrun alarm information is that the difference value of updated detection data and correspondingly stored detection data is larger than a preset extension threshold value, or the updated detection data is larger than a preset width threshold value and a preset length threshold value; the new crack overrun alarm information is that the number of new cracks in the updated detection data is larger than a preset new threshold value.
The power module 20 is respectively connected with the embedded industrial personal computer 7, the two-dimensional electric cradle head 8, the light supplement lamp, the industrial camera 10 and the 4G wireless data remote transmission module 9 and is used for supplying power to the embedded industrial personal computer, the two-dimensional electric cradle head, the light supplement lamp, the industrial camera and the 4G wireless data remote transmission module.
The data remote transmission sub-device comprises: a 4G base station 2, a 4G cellular network 3 and the Internet 4, and is used for transmitting the data of the field image acquisition and analysis sub-device to a monitoring center. The data remote transmission sub-device can realize remote wireless transmission of data.
The 4G wireless data remote transmission module 9 is connected with the 4G base station 2, the 4G base station 2 is connected with the 4G cellular network 3, the 4G cellular network 3 is connected with the Internet 4, and the Internet 4 is connected with the monitoring center 5. The number of the 4G base stations 2 can be multiple, and the 4G wireless data remote transmission module of each field image acquisition and analysis sub-device is connected with one 4G base station.
The monitoring center 5 includes: a network server, a client and a database. The internet 4 is connected to a web server, which is connected to the client and the database, respectively.
The client is used for establishing communication with each field image acquisition and analysis sub-device and receiving data transmitted by the field image acquisition and analysis sub-devices; the embedded industrial personal computer is also used for establishing connection with a user, receiving a regular or irregular detection instruction input by the user and transmitting the detection instruction to the embedded industrial personal computer. The client is also used for receiving the monitoring result and giving an alarm to a preset user according to the monitoring result. The database is used for storing the data transmitted by the field image acquisition and analysis sub-device.
The working process of the concrete structure surface crack detection device of the embodiment is as follows: the client-side sends the detection instruction to the field image acquisition and analysis sub-device regularly or irregularly according to the regular or irregular detection instruction input by the user. The field image acquisition and analysis sub-device controls the industrial camera and the light supplement lamp to shoot the regional images; the embedded industrial personal computer obtains the area image, carries out crack detection on the area image to obtain detection data, stores the detection data into a database, and simultaneously sends the detection data to a monitoring center through a 4G wireless data remote transmission module. And the client stores the received detection data to a database of the local server. And the user can perform statistics, comparison, summarization and printout of corresponding crack characteristic information on the detection data through the client.
When the monitoring distance between the industrial camera and the surface of the concrete structure is 6m, the crack detection resolution of the concrete structure surface crack detection device is 0.25 mm/pixel, and the main technical indexes of the concrete structure surface crack detection device are as follows:
1) the monitoring area range of the industrial camera is less than or equal to 2m × 2 m;
2) the recognizable minimum crack width is 0.05 mm;
3) when the crack width ranges from 0.1 mm to 0.38mm, the detection precision of the crack width is less than or equal to +/-0.02 mm;
4) the detection precision of the crack length is less than or equal to +/-10 percent.
Preferably, the protection grade of the concrete structure surface crack detection device is IP 65; the working temperature is between minus 40 ℃ and plus 65 ℃.
According to the concrete structure surface crack detection method and the concrete structure surface crack detection device, the width, the length and the distribution of the crack in a certain area range on the surface of the concrete structure can be remotely and long-term detected and monitored through the area image shot by the industrial camera, the whole process that the crack is initiated and expanded to endanger the structure safety is monitored and early-warned, and the method and the device are suitable for detecting and monitoring the crack of structures such as bridges, tunnels, dams, buildings and the like; the crack width recognition with higher precision can be achieved by using the industrial camera with lower resolution, and the implementation cost of the concrete structure surface crack detection method and the cost of the detection device are greatly reduced. The concrete structure surface crack detection method can extract fine cracks larger than 1/5 pixels, can realize sub-pixel identification of the width of the concrete structure surface crack, and greatly improves crack detection resolution.
Fig. 7 is a system diagram of a concrete structure surface crack detection system according to an embodiment of the present invention. Referring to fig. 7, the system for detecting cracks on the surface of a concrete structure includes:
the area image module 201 is used for acquiring area images of a plurality of areas on the surface of the concrete structure to be measured, which are shot by an industrial camera; partial images of the area images of the adjacent area on the surface of the concrete structure to be measured are overlapped. The number of regions is N, the number of region images is N, and the overlapping area of two adjacent region images is preferably at least 1/4 of the region image area.
And the panoramic image module 202 is configured to perform splicing and fusion processing on the area images to obtain a panoramic image.
The panoramic image module 202 specifically includes:
and the enhancement unit is used for carrying out image enhancement processing on the regional image by adopting a single-scale Retinex algorithm to obtain an enhanced regional image.
And the first image unit to be fused is used for taking the image of the enhanced area of any area of the surface of the concrete structure to be measured as a first image to be fused.
And the second image unit to be fused is used for taking the enhanced region image of the adjacent region of the first image to be fused as a second image to be fused.
The common feature point unit is used for extracting common feature points in overlapping pixel points of the overlapping parts of the first image to be fused and the second image to be fused by adopting a Scale-invariant feature transform (SIFT) algorithm to obtain a first coordinate of the common feature points in the first image to be fused and a second coordinate of the common feature points in the second image to be fused; each enhancement region image comprises an overlapping part and a non-overlapping part, pixel points of the overlapping part are overlapping pixel points, and the remaining pixel points are pixel points except the overlapping pixel points in the enhancement region image, namely each enhancement region image comprises the overlapping pixel points and the remaining pixel points, and the common feature points are feature pixel points in the overlapping pixel points of the enhancement region images of any two adjacent regions extracted by the SIFT algorithm.
And the coordinate transformation model unit is used for establishing a coordinate transformation model according to the first coordinate and the second coordinate.
And the splicing fusion image unit is used for calculating the coordinates of the residual pixel points of the second image to be fused in the coordinate system of the first image to be fused by using a coordinate transformation model, and calculating the gray value of the overlapped pixel points by using a direct average method to obtain a splicing fusion image A'.
And the panoramic image unit is used for replacing the first image to be fused with the spliced and fused image A', executing the second image unit to be fused, traversing all the images in the enhanced area, and splicing and fusing all the images in the enhanced area to obtain the panoramic image.
And the crack module 203 is configured to perform crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain a crack in the panoramic image.
The crack module 203 specifically includes:
and the crack image unit is used for extracting cracks in the panoramic image by using a crack detection algorithm based on the seepage model to obtain a crack image.
And the expansion unit is used for performing expansion processing on the crack image to obtain an expansion image and a crack area in the expansion image.
And the thinning unit is used for thinning the crack region in the expansion image by adopting a thinning algorithm, extracting the crack skeleton in the crack region and obtaining the crack skeleton, the crack skeleton pixel points and the thinned image.
And the merging unit is used for merging the gray values according to the thinned image, uniformly adding the gray values of the non-crack skeleton pixel points in the crack area in the expanded image to the crack skeleton pixel points adjacent to the non-crack skeleton pixel points, and obtaining the gray value merging value of the crack skeleton image and the crack skeleton pixel points.
And the calculation unit is used for calculating the starting point, the end point and the bifurcation point of the crack according to the crack skeleton image, obtaining the crack in the panoramic image and storing the starting point, the end point and the bifurcation point of the crack.
And the calculation module 204 is used for calculating the coordinates, the width and the length of the crack and storing the coordinates, the width and the length of the crack.
The calculating module 204 specifically includes:
and the coordinate unit is used for calculating the mass center of the crack region, and the coordinate of the mass center of the crack region is the coordinate of the crack.
And the standard width unit is used for acquiring a plurality of crack standard lines and the standard width of the crack standard lines.
And the combined gray value unit is used for sequentially performing expansion treatment, thinning treatment and gray value combination on the crack standard line to obtain a combined gray value of each pixel point of the crack standard line. The method is specifically used for: and sequentially performing expansion treatment, thinning treatment and gray value combination on the shot image comprising the plurality of crack standard lines to obtain a combined gray value of each pixel point of each crack standard line.
The gray level combination value unit is used for calculating the average value of all the combined gray levels and determining the average value as the gray level combination value of the crack standard line; the method is specifically used for: and calculating the average value of all the combined gray values of one crack standard line, determining the average value as the gray combined value of the crack standard line, and calculating the gray combined values of all the crack standard lines according to the gray combined value unit.
And the crack calibration coefficient unit is used for calculating the crack calibration coefficient through straight line fitting by utilizing the standard width and the gray level combination value of the crack standard line.
And the width unit is used for calculating the width of the crack by utilizing the gray combination value and the crack calibration coefficient of the crack skeleton pixel point. Width = gray scale value of crack skeleton pixel point × -crack calibration coefficient k.
And the length unit is used for calculating the length of the crack by utilizing the crack skeleton pixel point and the crack calibration coefficient. Crack length = total number of crack skeleton pixels x crack calibration coefficient k.
And the storage unit is used for storing the coordinates, the width, the length and the cracks.
The concrete structure surface crack detection system includes: and a monitoring module.
And the monitoring module is used for monitoring whether the cracks on the surface of the concrete structure are expanded or whether the cracks on the surface of the concrete structure are newly generated. The monitoring module specifically includes:
and the updating unit is used for updating the area images of the plurality of areas on the surface of the concrete structure to be detected, which are shot by the industrial camera, and updating the cracks according to the concrete structure surface crack detection method, wherein the updated cracks comprise the coordinates, the width and the length of the cracks.
The comparison unit is used for comparing the updated crack with the cracks stored in the crack database and judging whether the crack is expanded or whether a new crack exists, and specifically comprises the following steps:
and the matching subunit is used for matching the updated crack with the stored crack to obtain the stored crack corresponding to the updated crack.
And the crack expanding subunit is used for respectively subtracting the updated width and length of the crack from the corresponding stored width and length of the crack to judge whether the crack is expanded.
And the expansion overrun alarm subunit is used for judging that the crack expands and overrun and carrying out expansion overrun alarm if the difference value is greater than a preset expansion threshold value. Or directly comparing the updated width and length of the crack with a preset width threshold value and a preset length threshold value, if the updated width of the crack is larger than the preset width threshold value or the updated length of the crack is larger than the preset length threshold value, judging that the crack is expanded beyond the limit, and performing expansion over-limit alarm.
And the new crack subunit is used for subtracting the updated cracks from the stored cracks, the coordinates of the cracks are the mass center of the crack region, each crack has only one coordinate, the number of the updated coordinates is calculated, the number of the updated coordinates is subtracted from the number of the stored crack coordinates to obtain the number of the new cracks, and if the number of the new cracks is greater than or equal to 1, the new cracks are judged to appear. After matching the updated fracture with the stored fractures, the remaining updated fractures are determined to be new fractures.
And the updating storage unit is used for storing the updated crack to the crack database.
And the new crack over-limit alarm unit is used for judging that the new cracks are over-limited and alarming the new cracks if the number of the new cracks is larger than a preset new threshold value.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (4)
1. A concrete structure surface crack detection method is characterized by comprising the following steps:
acquiring area images of a plurality of areas on the surface of a concrete structure to be measured, which are shot by an industrial camera; partial images of the area images of the adjacent area on the surface of the concrete structure to be detected are overlapped;
splicing and fusing the area images to obtain a panoramic image;
the splicing and fusing processing of the area images to obtain a panoramic image specifically comprises the following steps:
performing image enhancement processing on the region image by adopting a single-scale Retinex algorithm to obtain an enhanced region image;
carrying out crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain cracks in the panoramic image;
the method for detecting the cracks of the panoramic image by using the crack detection algorithm based on the seepage model to obtain the cracks in the panoramic image specifically comprises the following steps:
extracting cracks in the panoramic image by using a crack detection algorithm based on a seepage model to obtain a crack image;
performing expansion processing on the crack image to obtain an expansion image and a crack area in the expansion image;
thinning the expansion image by adopting a thinning algorithm, extracting a crack framework in the crack area, and obtaining the crack framework, crack framework pixel points and a thinned image;
carrying out gray value combination according to the refined image, and uniformly adding the gray value of a non-crack skeleton pixel point in the expanded image to a crack skeleton pixel point adjacent to the non-crack skeleton pixel point to obtain a gray value combination value of the crack skeleton image and the crack skeleton pixel point;
calculating a starting point, an end point and a bifurcation point of the crack according to the crack skeleton image to obtain the crack in the panoramic image, and storing the starting point, the end point and the bifurcation point of the crack;
calculating the coordinates, the width and the length of the crack, and storing the coordinates, the width and the length of the crack;
the calculating the coordinates, the width and the length of the crack and storing the coordinates, the width and the length of the crack specifically comprise:
calculating the center of mass of the crack region, wherein the coordinate of the center of mass of the crack region is the coordinate of the crack;
acquiring a plurality of crack standard lines and standard widths of the crack standard lines;
sequentially performing expansion treatment, thinning treatment and grey value combination on the crack standard line to obtain a combined grey value of each pixel point of the crack standard line;
calculating the average value of all the merged gray values, and determining the average value as the gray merged value of the crack standard line;
calculating by using the standard width and the gray level combination value of the crack standard line through straight line fitting to obtain a crack calibration coefficient;
calculating to obtain the width of the crack by utilizing the gray combining value of the crack skeleton pixel point and the crack calibration coefficient;
calculating to obtain the length of the crack by using the crack skeleton pixel point and the crack calibration coefficient;
storing the coordinates, the width, the length, and the crack.
2. The method for detecting cracks on the surface of a concrete structure according to claim 1, wherein the splicing and fusing of the area images to obtain a panoramic image further comprises:
taking the reinforced area image of any area of the surface of the concrete structure to be detected as a first image to be fused;
taking the enhanced region image adjacent to the first image to be fused as a second image to be fused;
extracting a common feature point in an overlapping pixel point of an overlapping part of the first image to be fused and the second image to be fused by adopting a scale invariant feature transformation algorithm to obtain a first coordinate of the common feature point in the first image to be fused and a second coordinate of the common feature point in the second image to be fused respectively; each enhanced region image comprises overlapped pixel points and residual pixel points;
establishing a coordinate transformation model according to the first coordinate and the second coordinate;
calculating the coordinates of the residual pixel points of the second image to be fused in the coordinate system of the first image to be fused by using the coordinate transformation model, and calculating the gray value of the overlapped pixel points by using a direct averaging method to obtain a spliced and fused image A';
and replacing the first image to be fused with the spliced and fused image A', returning to the step of taking the enhanced region image of the adjacent region of the first image to be fused as a second image to be fused, and traversing all the enhanced region images to obtain a panoramic image.
3. A concrete structure surface crack detection system, comprising:
the area image module is used for acquiring area images of a plurality of areas on the surface of the concrete structure to be measured, which are shot by an industrial camera; partial images of the area images of the adjacent area on the surface of the concrete structure to be detected are overlapped;
the panoramic image module is used for splicing and fusing the area images to obtain a panoramic image;
the panoramic image module specifically comprises:
the enhancement unit is used for carrying out image enhancement processing on the regional image by adopting a single-scale Retinex algorithm to obtain an enhanced regional image;
the crack module is used for carrying out crack detection on the panoramic image by using a crack detection algorithm based on a seepage model to obtain cracks in the panoramic image;
the fracture module specifically includes:
the crack image unit is used for extracting cracks in the panoramic image by using a crack detection algorithm based on a seepage model to obtain a crack image;
the expansion unit is used for performing expansion processing on the crack image to obtain an expansion image and a crack area in the expansion image;
the thinning unit is used for thinning the expansion image by adopting a thinning algorithm, extracting a crack framework in the crack area and obtaining the crack framework, crack framework pixel points and a thinned image;
the merging unit is used for performing gray value merging according to the refined image, uniformly adding the gray value of the non-crack skeleton pixel point in the expanded image to the crack skeleton pixel point adjacent to the non-crack skeleton pixel point, and obtaining the gray value merging value of the crack skeleton image and the crack skeleton pixel point;
the calculation unit is used for calculating a starting point, an end point and a bifurcation point of the crack according to the crack skeleton image to obtain the crack in the panoramic image and storing the starting point, the end point and the bifurcation point of the crack;
the calculation module is used for calculating the coordinates, the width and the length of the crack and storing the coordinates, the width and the length of the crack;
the calculation module specifically includes:
the coordinate unit is used for calculating the center of mass of the crack region, and the coordinate of the center of mass of the crack region is the coordinate of the crack;
the standard width unit is used for acquiring a plurality of crack standard lines and the standard width of the crack standard lines;
the combined gray value unit is used for sequentially performing expansion treatment, thinning treatment and gray value combination on the crack standard line to obtain a combined gray value of each pixel point of the crack standard line;
the gray level combination value unit is used for calculating the average value of all the combination gray levels and determining the average value as the gray level combination value of the crack standard line;
the crack calibration coefficient unit is used for calculating a crack calibration coefficient through straight line fitting by using the standard width and the gray level combination value of the crack standard line;
the width unit is used for calculating the width of the crack by utilizing the gray combination value of the crack skeleton pixel point and the crack calibration coefficient;
the length unit is used for calculating the length of the crack by using the crack skeleton pixel point and the crack calibration coefficient;
a storage unit for storing the coordinates, the width, the length, and the crack.
4. The concrete structure surface crack detection system of claim 3, wherein the panoramic image module further comprises:
the first image unit to be fused is used for taking the image of the enhanced area of any area of the surface of the concrete structure to be measured as a first image to be fused;
the second image unit to be fused is used for taking the enhanced region image of the adjacent region of the first image to be fused as a second image to be fused;
the common feature point unit is used for extracting common feature points in overlapping pixel points of the overlapping parts of the first image to be fused and the second image to be fused by adopting a scale-invariant feature transformation algorithm to obtain a first coordinate of the common feature points in the first image to be fused and a second coordinate of the common feature points in the second image to be fused; each enhanced region image comprises overlapped pixel points and residual pixel points;
the coordinate transformation model unit is used for establishing a coordinate transformation model according to the first coordinate and the second coordinate;
the splicing fusion image unit is used for calculating the coordinates of the residual pixel points of the second image to be fused in the coordinate system of the first image to be fused by using the coordinate transformation model, and calculating the gray value of the overlapped pixel points by using a direct averaging method to obtain a splicing fusion image A';
and the panoramic image unit is used for replacing the first image to be fused with the spliced and fused image A', executing a second image unit to be fused, and traversing all the images in the enhanced area to obtain a panoramic image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010276561.XA CN111242851B (en) | 2020-04-10 | 2020-04-10 | Concrete structure surface crack detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010276561.XA CN111242851B (en) | 2020-04-10 | 2020-04-10 | Concrete structure surface crack detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111242851A CN111242851A (en) | 2020-06-05 |
| CN111242851B true CN111242851B (en) | 2020-08-04 |
Family
ID=70875546
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010276561.XA Active CN111242851B (en) | 2020-04-10 | 2020-04-10 | Concrete structure surface crack detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111242851B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111696098B (en) * | 2020-06-16 | 2021-11-26 | 乐清市凡山电器有限公司 | Concrete member detection system and method based on big data |
| CN112348804A (en) * | 2020-11-25 | 2021-02-09 | 浙江大成工程项目管理有限公司 | Method, system and device for monitoring crack caused by foundation pit displacement and storage medium |
| CN112560587B (en) * | 2020-11-27 | 2022-04-08 | 贵州中建建筑科研设计院有限公司 | Dynamic early warning method and system for convolutional neural network slope crack change |
| CN112630223B (en) * | 2020-12-07 | 2023-12-26 | 杭州申昊科技股份有限公司 | Tunnel crack detection system and method |
| CN112906562B (en) * | 2021-02-19 | 2022-08-02 | 内蒙古科技大学 | Safety early warning method for side plate crack of trolley of chain grate |
| CN113610060B (en) * | 2021-09-29 | 2022-01-04 | 北京雷图科技有限公司 | Structure crack sub-pixel detection method |
| CN115184378B (en) * | 2022-09-15 | 2024-03-29 | 北京思莫特科技有限公司 | Concrete structure disease detection system and method based on mobile equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103528527A (en) * | 2013-10-15 | 2014-01-22 | 北京交通大学长三角研究院 | Area selection-based automatic crack size measurement method |
| CN104700395A (en) * | 2014-11-11 | 2015-06-10 | 长安大学 | Method and system for detecting appearance crack of structure |
| JP6380054B2 (en) * | 2014-11-28 | 2018-08-29 | Dic株式会社 | Concrete crack detecting agent and concrete crack detecting method |
-
2020
- 2020-04-10 CN CN202010276561.XA patent/CN111242851B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111242851A (en) | 2020-06-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111242851B (en) | Concrete structure surface crack detection method and system | |
| KR102121958B1 (en) | Method, system and computer program for providing defect analysis service of concrete structure | |
| CN103778681B (en) | A kind of vehicle-mounted highway cruising inspection system and data acquisition and disposal route | |
| KR100726009B1 (en) | System and method for measuring displacement of structure | |
| CN109712148A (en) | Segment joint position automatic identifying method based on shield tunnel image | |
| CN113192063B (en) | Bridge line-shaped monitoring system and bridge line-shaped monitoring method | |
| CN110874861B (en) | Three-dimensional digital image acquisition method and device | |
| CN111610193A (en) | System and method for inspecting structural defects of subway tunnel segment by adopting multi-lens shooting | |
| CN112802004B (en) | Portable intelligent video detection device for health of power transmission line and pole tower | |
| CN113537016B (en) | Method for automatically detecting and early warning road damage in road patrol | |
| CN113554667B (en) | Three-dimensional displacement detection method and device based on image recognition | |
| CN110260785A (en) | Rock Tunnel area surface analysis based on 3 D laser scanning feeds back integrated system | |
| CN110560376B (en) | Product surface defect detection method and device | |
| CN112528979B (en) | Transformer substation inspection robot obstacle distinguishing method and system | |
| CN106225702A (en) | Fracture width detection apparatus and method | |
| CN112000124B (en) | Unmanned aerial vehicle inspection method applied to power grid | |
| KR20130133596A (en) | Method and apparatus for measuring slope of poles | |
| CN114926415A (en) | Steel rail surface detection method based on PCNN and deep learning | |
| CN109827515A (en) | A kind of the screw steel wire area of bed detection system and method for separate type | |
| EP3852356A1 (en) | Composition processing system, composition processing device, and composition processing method | |
| CN112781518B (en) | House deformation monitoring method and system | |
| KR100887353B1 (en) | System and method for measuring outdoor advertisement article | |
| CN114812403A (en) | Large-span steel structure hoisting deformation monitoring method based on unmanned aerial vehicle and machine vision | |
| CN113313760A (en) | Corrosion detection method, device and system for array type metal connecting piece | |
| CN114419080B (en) | Curtain wall inspection system and 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 |