CN114812450B - Asphalt pavement construction uniformity detection and evaluation method based on machine vision - Google Patents
Asphalt pavement construction uniformity detection and evaluation method based on machine vision Download PDFInfo
- Publication number
- CN114812450B CN114812450B CN202210440015.4A CN202210440015A CN114812450B CN 114812450 B CN114812450 B CN 114812450B CN 202210440015 A CN202210440015 A CN 202210440015A CN 114812450 B CN114812450 B CN 114812450B
- Authority
- CN
- China
- Prior art keywords
- road surface
- asphalt pavement
- road
- mtd
- fractal dimension
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 69
- 239000010426 asphalt Substances 0.000 title claims abstract description 60
- 238000001514 detection method Methods 0.000 title claims abstract description 18
- 238000011156 evaluation Methods 0.000 title claims abstract description 17
- 238000003709 image segmentation Methods 0.000 claims abstract description 6
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 238000001914 filtration Methods 0.000 claims abstract description 5
- 238000012937 correction Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 38
- 238000005286 illumination Methods 0.000 claims description 14
- 238000005204 segregation Methods 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 8
- 238000011179 visual inspection Methods 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 239000002344 surface layer Substances 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000005056 compaction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000003887 surface segregation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
-
- E—FIXED CONSTRUCTIONS
- E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
- E01C—CONSTRUCTION OF, OR SURFACES FOR, ROADS, SPORTS GROUNDS, OR THE LIKE; MACHINES OR AUXILIARY TOOLS FOR CONSTRUCTION OR REPAIR
- E01C23/00—Auxiliary devices or arrangements for constructing, repairing, reconditioning, or taking-up road or like surfaces
- E01C23/01—Devices or auxiliary means for setting-out or checking the configuration of new surfacing, e.g. templates, screed or reference line supports; Applications of apparatus for measuring, indicating, or recording the surface configuration of existing surfacing, e.g. profilographs
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to a machine vision-based asphalt pavement construction uniformity detection and evaluation method, which comprises the steps of acquiring multi-angle road surface images at a certain interval in the cross section and longitudinal section directions of a detected road section through a road surface image acquisition device; preprocessing of image deformation correction, image segmentation, shadow removal and filtering noise reduction are sequentially carried out on the acquired multi-angle road surface image; carrying out 3D point cloud reverse reconstruction on the preprocessed multi-angle road table image to obtain a road table three-dimensional texture model; extracting texture distribution characteristic parameters of the three-dimensional texture model: average construction depth index MTD and fractal dimension D, and then calculating the asphalt pavement uniformity index: average constructional depth ratio U MTD And fractal dimension ratio U D Thereby evaluating the construction uniformity of the asphalt pavement. The invention can accurately and efficiently acquire the road table multi-angle images and the road table texture characteristic information; the construction uniformity of the asphalt pavement can be more accurately evaluated.
Description
Technical Field
The invention relates to the technical field of asphalt pavement construction, in particular to a machine vision-based asphalt pavement construction uniformity detection and evaluation method.
Background
The construction uniformity problem of the asphalt pavement obviously affects the occurrence of early diseases of the asphalt pavement and the long service life, is mainly affected by the segregation degree in the mixing, transporting and paving processes of the mixture and the pavement compaction uniformity degree, and is intuitively embodied on the uniformity degree of the pavement texture distribution of the newly built asphalt pavement. Therefore, the method accurately and efficiently collects the road surface textures of the newly built asphalt pavement, calculates the construction depth and uniformity indexes, and evaluates the construction depth and uniformity indexes, thereby playing an important role in controlling the construction uniformity of the asphalt pavement.
Visual inspection, sanding and laser methods are currently common methods for road surface texture detection on asphalt pavement. The visual inspection method is simple to operate, but is too strong in subjectivity and cannot be quantitatively judged; the sand paving method is simple to operate and low in cost, but has low measurement efficiency, is greatly influenced by human factors and cannot obtain complete pavement texture characteristic information; the laser method can directly obtain the road surface texture characteristic information of the asphalt pavement, but has the defects of expensive equipment and difficult large-scale application.
With the development of machine vision technology, the technology is gradually applied to asphalt pavement texture detection, and aims to collect pavement texture characteristic information and evaluate asphalt pavement construction uniformity accurately and efficiently at low cost. However, the conventional machine vision technology directly extracts road surface texture feature information from a single-angle picture or extracts road surface texture feature information after reconstructing a three-dimensional model based on the single-angle picture, so that the problem of loss of road surface texture feature details exists, and the obtained construction depth and the asphalt pavement uniformity index have certain deviation, so that the construction uniformity of the asphalt pavement cannot be reasonably estimated. In addition, the conventional method only calculates the section depth MPD, namely the one-dimensional space index, after three-dimensional reconstruction, and cannot quantitatively evaluate the construction uniformity of the asphalt pavement.
Disclosure of Invention
Aiming at the problems, the patent provides a machine vision-based asphalt pavement construction uniformity detection and evaluation method, which can extract road surface texture distribution characteristic parameters more efficiently and accurately, avoid the loss of three-dimensional texture distribution characteristic details caused by single-angle pictures, obtain more accurate asphalt pavement uniformity indexes by utilizing the road surface texture distribution characteristic parameters extracted from the three-dimensional texture model, and analyze and evaluate the asphalt pavement construction uniformity more accurately.
A machine vision-based asphalt pavement construction uniformity detection and evaluation method comprises the following steps:
step 1: multi-angle road surface image acquisition
A plurality of acquisition points are arranged in the cross section and longitudinal section directions of the detection road section at certain intervals, and multi-angle road surface images of the acquisition points are obtained through a road surface image acquisition device;
the road surface image acquisition device comprises an image acquisition module, a main control computer for reading information of the image acquisition module, an illumination module for providing illumination for the image acquisition module and a support frame for fixing the image acquisition module and the illumination module.
The support frame comprises a panel and a base for supporting and adjusting the height of the panel, and the image acquisition module and the illumination module are fixed on one side of the panel, which is close to the road surface;
the image acquisition module comprises a plurality of cameras which are uniformly distributed at intervals in a ring shape on the same horizontal height, and one end of each camera far away from the panel is inclined towards the direction close to the center of the ring shape;
preferably, the angle between each camera and the horizontal plane is 45 degrees.
Step 2: multi-angle road table image preprocessing
Preprocessing of image deformation correction, image segmentation, shadow removal and filtering noise reduction are sequentially carried out on the acquired multi-angle road surface image;
step 3: three-dimensional texture model for obtaining road table through reverse reconstruction of 3D point cloud
Accurately calculating the position of a camera through the two-dimensional space coordinates of a single-angle picture by the preprocessed multi-angle road table image, realizing characteristic point matching and point cloud reconstruction among the multi-angle images, and carrying out triangular mesh division to obtain a road table three-dimensional texture model;
the specific process of the feature point matching is as follows:
assume that the effective focal length of the first camera is f 1 And in world coordinate system O-xyz, its position is marked as O 1 -x 1 y 1 z 1 The acquired road surface image is positioned in a two-dimensional coordinate system O-XY, and the position of the road surface image is marked as O 1 -X 1 Y 1 The method comprises the steps of carrying out a first treatment on the surface of the The effective focal length of the second camera is f 2 Also in world coordinate system O-xyz, its position is marked as O 2 -x 2 y 2 z 2 The acquired road surface image is positioned in a two-dimensional coordinate system O-XY, and the position of the road surface image is marked as O 2 -X 2 Y 2 Representing O by a space conversion matrix M according to the perspective transformation basic principle 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 The interrelationship between the two coordinate systems is as follows, and alignment of the two coordinate systems is realized:
wherein ,
represents O 1 -x 1 y 1 z 1 and Q2 -x 2 y 2 z 2 A matrix of azimuth rotations in between;
represents O 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 A variation vector between them;
the point cloud reconstruction is obtained through the corresponding relation between the two camera image surface points, and is specifically as follows:
wherein ρ is the two-dimensional coordinates of the spatial point of the target road surface area in the rest angle images;
thus, the three-dimensional coordinate mathematical relationship of the spatial point of the target road surface area is obtained as follows,
and thereby creating a road table three-dimensional texture model.
Step 4: road table texture distribution characteristic parameter extraction
Extracting texture distribution characteristic parameters through a three-dimensional texture model: the average construction depth index MTD, the fractal dimension D and the relation between the MTD and the fractal dimension D are specifically as follows:
the average construction depth index MTD is calculated as shown in the following formula:
V＝∫∫ Q [P 0 -P(x，y)]dxdy
wherein ,P0 -a road surface plane being the space of the target road surface area;
p (x, y) -is the area surface formed by the space pavement elevation points of the target pavement area;
q-is the integration region;
v-is P 0 A volume enclosed by the road surface;
a-is the area of region Q;
the process of calculating the fractal dimension D is as follows: after fast Fourier transformation is carried out on the road surface texture coordinate elevation data of the target road surface area, a power spectrum S (omega) taking omega as power is obtained, then linear fitting is carried out on lg S (omega) and lg omega, and the relation between the fitting slope alpha and the fractal dimension D is as follows:
S(ω)∝ω -(5-2D)
and fitting the average construction depth index MTD and the fractal dimension D to obtain the relation between the MTD and the fractal dimension D.
Step 5: calculating uniformity index of asphalt pavement
Calculating the average construction depth ratio U of the road section to be detected MTD And fractal dimension ratio U D The specific process is as follows:
firstly, determining a region with excellent pavement construction uniformity by a visual inspection method, and measuring an average construction depth index MTD (maximum Transmission distance) 0 Fractal dimension D 0 Then detecting the road section to be detected through the road surface image acquisition device to obtain the MTD n and Dn Thereby calculating and obtaining the uniformity index of the asphalt pavement: average constructional depth ratio U MTD And fractal dimension ratio U D Tool for cleaning and cleaningThe body has the following formula:
wherein ,MTDn -constructing a depth for the average of the road segments to be detected;
MTD 0 -an average construction depth for the target road segment;
D n -a fractal dimension for the road segment to be detected;
D 0 -fractal dimension for the target road segment.
Step 6: evaluation of Bituminous pavement construction uniformity
By fractal dimension ratio U D Judging the segregation degree of the asphalt pavement, and evaluating the construction uniformity of the asphalt pavement, wherein the concrete process is as follows: based on the asphalt pavement segregation degree determination method specified in the existing standard NCHRP Report441, the asphalt pavement segregation degree determination method is implemented by using the average construction depth ratio U MTD Instead of the structural ratio, the structural depth ratio U is determined by the critical average MTDC Calculating critical fractal dimension ratio U DC In step 1-step 5, the U of the road section to be detected is calculated D Then, judge U D Ratio of critical fractal dimension U DC And (3) judging the segregation degree of the asphalt surface layer so as to evaluate the construction uniformity of the asphalt pavement.
Compared with the existing evaluation method, the evaluation method has the following beneficial effects:
(1) The invention provides a road surface image acquisition device which can accurately and efficiently acquire road surface multi-angle images;
(2) The invention realizes the three-dimensional reconstruction of the road surface texture of the asphalt pavement based on the machine vision technology, and can obtain the road surface texture characteristic information more accurately;
(3) The invention provides an asphalt pavement uniformity index U based on texture feature information extracted from a road surface three-dimensional model MTD and UD The construction uniformity of the asphalt pavement can be more accurately and quantitatively analyzed and evaluated.
Drawings
FIG. 1 is a schematic diagram of a road surface image acquisition device according to the present patent;
FIG. 2 is a schematic view showing the bottom angle structure of the road surface image capturing device according to the embodiment 1;
FIG. 3 is the angle of the road surface image captured by the first camera of the road surface image capturing device in embodiment 1;
FIG. 4 is the angle of the road surface image captured by the second camera of the road surface image capturing device in example 1;
FIG. 5 is the angle of the road surface image captured by the third camera of the road surface image capturing device in embodiment 1;
FIG. 6 is a schematic diagram of a three-dimensional texture model of the road surface obtained from the road surface images collected in FIGS. 3-5 in example 1;
FIG. 7 is a construction depth MTD of the sanding method in example 1 p And a correlation graph of the construction depth MTD of the road table three-dimensional texture model;
fig. 8 is a correlation graph of the construction depth MTD and the fractal dimension D.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention for achieving the intended purpose, the following detailed description will refer to the specific implementation, structure, characteristics and effects according to the present invention with reference to the accompanying drawings and preferred embodiments.
Example 1
A machine vision-based asphalt pavement construction uniformity detection and evaluation method comprises the following steps:
step 1: multi-angle road surface image acquisition
The road surface image acquisition device 1 acquires multi-angle road surface images at a certain interval in the cross section and longitudinal section directions of the detected road section.
As a specific scheme, as shown in fig. 1 and 2, the road surface image acquisition device 1 includes an image acquisition module 1-1, a main control computer 1-2 for reading information of the image acquisition module 1-1, an illumination module 1-3 for providing illumination for the image acquisition module 1-1, and a support frame 1-4 for fixing the image acquisition module 1-1 and the illumination module 1-3, where the support frame 1-4 includes a panel 1-41 and a base 1-42 for supporting and adjusting the height of the panel 1-41, and the base 1-42 can be adjusted in height by means of threads, telescopic rods, and the like, and is not limited herein.
The image acquisition module 1-1 includes a plurality of cameras, and a plurality of cameras are fixed in the panel 1-41 and are close to one side on road surface, and a plurality of cameras are annular interval equipartition and arrange on the panel 1-41, and here interval equipartition refers to that the contained angle gamma that forms between arbitrary two adjacent cameras and the annular center is equal, like fig. 2, and preferably, the camera number of selecting in this patent is three, and the camera model can be: the arrangement track of the Basler acA1300 industrial CCD cameras on the panel forms a circle, and an included angle gamma formed between two adjacent cameras and the circle center is 120 degrees. Furthermore, the end of each camera facing away from the panels 1-41 is inclined towards the centre of the ring and forms an angle β with the horizontal plane, as is preferred in fig. 1, the angle β between each camera and the horizontal plane in the present invention being 45 °.
The built-in API programming of the main control computer 1-2 triggers a plurality of cameras to work simultaneously in real time, so that multi-angle image acquisition is completed simultaneously and stored in the main control computer 1-2. The configuration of the main control computer 1-2 can be i5-9700CPU, GTX2060 2G and 16GB RAM. The road surface image acquisition device 1 can acquire the multi-angle road surface images of the same area at a certain time point, so that the acquired images can more accurately represent the current road surface form.
The illumination module 1-3 is fixed on the panel 1-41 of the support frame 1-4 to provide illumination for the image acquisition module 1-1, and can be an LED shadowless lamp with the power of 60w, so that the image is more clear and visible, and the acquired road surface image is not influenced by natural illumination intensity.
The invention sets image sampling points in the cross section direction of the detection road section according to the 1m interval, sets a plurality of image sampling points in the longitudinal section direction according to the 20m interval, and preferably, 8 image sampling points are arranged in the detection road section; the API call is completed through MATLAB to acquire multi-angle road table images of each acquisition point, one of the acquisition points marked as T01 is taken as an example, and the road table images of different angles acquired by the three cameras are respectively shown in figures 3-5.
Step 2: multi-angle road table image preprocessing
And carrying out pretreatment of image deformation correction, image segmentation, shadow removal, filtering and noise reduction on the acquired multi-angle road surface image in sequence.
The invention adopts MATLAB to carry out distortion elimination and unification of brightness and contrast of the image; image segmentation is carried out by adopting a Graph Cut interactive image segmentation algorithm, and the foreground is segmented from the background by utilizing the difference between the background and foreground pixel points, namely, a target image is extracted from the background; although the lighting module 1-3 of the invention can prevent shadow generation to the greatest extent, the situation that individual images are blocked by shadows still exists, and MATLAB is utilized to call a Gaussian smoothing function and a brightness compensation command to remove shadows of the images; and filtering and denoising the image by using a wavelet threshold denoising algorithm.
Step 3: three-dimensional texture model for obtaining road table through reverse reconstruction of 3D point cloud
And reversely reconstructing the preprocessed multi-angle road table image through a 3D point cloud to obtain a three-dimensional texture model of the road table, wherein the 3D point cloud reversely reconstructing is used for precisely calculating the position of a camera through the two-dimensional space coordinates of the single-angle image and realizing characteristic point matching and point cloud reconstruction among the multi-angle images, and then carrying out triangular grid division to finally obtain the three-dimensional texture model of the road table, as shown in fig. 6. The specific process of feature point matching is as follows:
first, assume that the effective focal length of the first camera in the present invention is f 1 And in world coordinate system O-xyz, its position is marked as O 1 -x 1 y 1 z 1 The acquired road surface image is positioned in a two-dimensional coordinate system O-XY, and the position of the road surface image is marked as O 1 -X 1 Y 1 The method comprises the steps of carrying out a first treatment on the surface of the The effective focal length of the second camera is f 2 Also in world coordinate system O-xyz, its position is marked as O 2 -x 2 y 2 z 2 The acquired road surface image is positioned in a two-dimensional coordinate system O-XY, and the position of the road surface image is marked as O 2 -X 2 Y 2 According to the perspective transformation basic principle, a spatial transformation matrix M can be used to represent O 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 Interrelationship between:
wherein ,represents O 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 Orientation rotation matrix between->Represents O 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 A variation vector between them;
thus, alignment of the two coordinate systems can be achieved by the space transformation matrix M.
The point cloud reconstruction is mainly obtained through a corresponding relation formula of two camera image surface points, and is as follows:
wherein ρ is the two-dimensional coordinates of the spatial point of the target road surface area in the rest angle images.
Thus, the three-dimensional coordinate mathematical relationship of the spatial points of the target pavement area can be obtained, and the road surface three-dimensional texture model can be obtained, as shown in fig. 6.
wherein ,x1 y 1 z 1 Is a three-dimensional space coordinate; x is X 1 Y 1 Is a two-dimensional space coordinate; f (f) 1 f 2 Focal length for two phases; r is (r) 1 -r 9 Is the element of the azimuth rotation matrix R, t x 、t y and tz Is an element in T.
Step 4: road table texture distribution characteristic parameter extraction
The texture distribution characteristic parameters comprise an average construction depth index MTD and a fractal dimension D of the texture distribution characteristic parameters, wherein the calculation process of the average construction depth index MTD is as follows:
V＝∫∫ Q [P 0 -P(x，y)]dxdy
wherein ,P0 -a road surface plane being the space of the target road surface area;
p (x, y) -is the area surface formed by the space pavement elevation points of the target pavement area;
q-is the integration region;
v-is P 0 A volume enclosed by the road surface;
a-is the area of region Q.
Construction depth MTD measured by establishing a sanding method p And the relation between MTDs obtained in the invention, the validity of the road table three-dimensional texture model in the invention is verified, and the results of 8 image acquisition points set in the invention are shown in FIG. 7: taking an average construction depth index MTD obtained by a road surface three-dimensional texture model as an abscissa, and measuring the construction depth MTD by a sand paving method p The ordinate is taken as the ordinate, and the fitting goodness R of the two is obtained 2 0.96, it is demonstrated that the method provided by the invention can well simulate the real pavement texture condition.
On the basis, the pavement texture coordinate elevation data of the target pavement area is subjected to fast Fourier transformation to obtain a power spectrum S (omega) taking omega as power, then, lg S (omega) and lg omega are subjected to linear fitting, and the relation between a fitting slope alpha and a fractal dimension D is shown as follows:
S(ω)∝ω -(5-2D)
the average construction depth index MTD is taken as an abscissa, the fractal dimension D is taken as an ordinate, the relation between the MTD and the D is established, the results of the 8 image acquisition points arranged in the invention are shown in figure 8, the straight line fitting formula between the average construction depth index MTD and the fractal dimension D is y= -2.532x+1.5152, and the fitting goodness R of the average construction depth index MTD and the fractal dimension D is equal to the fitting goodness R of the average construction depth index MTD and the fractal dimension D 2 0.97.
Step 5: calculating uniformity index of asphalt pavement
Firstly, determining an area with excellent road construction uniformity in a target road section by a visual inspection method, and measuring an average construction depth index MTD (maximum Transmission device) 0 Fractal dimension D 0 。
Performing the processes of step 1-step 4 on the road section to be detected to obtain the MTD n and Dn And then calculating to obtain the uniformity index of the section of asphalt pavement: average constructional depth ratio U MTD And fractal dimension ratio U D The calculation process is as follows:
wherein ,MTDn -constructing a depth for the average of the road segments to be detected;
MTD 0 -an average construction depth for the target road segment;
D n -for the detectionFractal dimension of road segments;
D 0 -fractal dimension for the target road segment.
Step 6: evaluation of Bituminous pavement construction uniformity
The present invention uses the average construction depth ratio U based on the asphalt pavement segregation degree determination method proposed in NCHRP Report441 as shown in Table 1 MTD Instead of the formation ratio (formation depth MTD at non-uniformity) Isolation Mean depth of construction MTD Average of ) Judging the segregation degree of the asphalt pavement surface layer, wherein the specific process is as follows:
asphalt surface segregation degree determination criteria in Table 1 NCHRP Report441
The invention calculates the critical average construction depth ratio U according to the relation between the MTD and the D obtained in the step 4 MTDC Corresponding critical fractal dimension ratio U DC The results are shown in Table 2:
TABLE 2 machine vision based asphalt pavement segregation degree criterion
wherein ：
carrying out the processes of the steps 1-5 on the road section to be detected to obtain the fractal dimension ratio U D And substituting the asphalt pavement segregation degree into the table 2 for judgment, so that the asphalt pavement segregation degree of the road section to be tested can be obtained, and the construction uniformity of the asphalt pavement is evaluated.
The present invention is not limited to the above embodiments, but is capable of modification and variation in detail, and other modifications and variations can be made by those skilled in the art without departing from the scope of the present invention.
Claims (6)
1. The machine vision-based asphalt pavement construction uniformity detection and evaluation method is characterized by comprising the following steps of:
step 1: multi-angle road surface image acquisition
A plurality of acquisition points are arranged in the cross section and longitudinal section directions of the detection road section at certain intervals, and multi-angle road surface images of the acquisition points are acquired through a road surface image acquisition device (1);
step 2: multi-angle road table image preprocessing
Preprocessing of image deformation correction, image segmentation, shadow removal and filtering noise reduction are sequentially carried out on the acquired multi-angle road surface image;
step 3: three-dimensional texture model for obtaining road table through reverse reconstruction of 3D point cloud
Accurately calculating the position of a camera through the two-dimensional space coordinates of a single-angle picture by the preprocessed multi-angle road table image, realizing characteristic point matching and point cloud reconstruction among the multi-angle images, and carrying out triangular mesh division to obtain a road table three-dimensional texture model;
the specific process of feature point matching in the step 3 is as follows:
assume that the effective focal length of the first camera is f 1 And in world coordinate system O-xyz, its position is marked as O 1 -x 1 y 1 z 1 The acquired road surface image is positioned in a two-dimensional coordinate system O-XY, and the position of the road surface image is marked as O 1 -X 1 Y 1 The method comprises the steps of carrying out a first treatment on the surface of the The effective focal length of the second camera is f 2 Also in world coordinate system O-xyz, its position is marked as O 2 -x 2 y 2 z 2 The acquired road surface image is positioned in a two-dimensional coordinate system O-XY, and the position of the road surface image is marked as O 2 -X 2 Y 2 Representing O by a space conversion matrix M according to the perspective transformation basic principle 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 The interrelationship between the two coordinate systems is as follows, and alignment of the two coordinate systems is realized:
wherein ,
represents O 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 A matrix of azimuth rotations in between;
represents O 1 -x 1 y 1 z 1 and O2 -x 2 y 2 z 2 A variation vector between them;
step 4: road table texture distribution characteristic parameter extraction
Extracting texture distribution characteristic parameters through a three-dimensional texture model: the average construction depth index MTD and the fractal dimension D and the relation between the MTD and the fractal dimension D;
the specific process of the step 4 is as follows:
the average construction depth index MTD is calculated as shown in the following formula:
V＝∫∫ Q [P 0 -P(x,y)dxdy
wherein ,P0 -a road surface plane being the space of the target road surface area;
p (x, y) -is the area surface formed by the space pavement elevation points of the target pavement area;
q-is the integration region;
v-is P 0 A volume enclosed by the road surface;
a-is the area of region Q;
the process of calculating the fractal dimension D is as follows: after fast Fourier transformation is carried out on the road surface texture coordinate elevation data of the target road surface area, a power spectrum S (omega) taking omega as power is obtained, and then straight line fitting is carried out on lgS (omega) and lgomega, wherein the relation between the fitting slope alpha and the fractal dimension D is as follows:
S(ω)∝ω -(5-2D)
fitting the average construction depth index MTD and the fractal dimension D to obtain a relation between the MTD and the fractal dimension D;
step 5: calculating uniformity index of asphalt pavement
Calculating the average construction depth ratio U of the road section to be detected MTD And fractal dimension ratio U D ；
The specific process of the step 5 is as follows:
firstly, determining a region with excellent pavement construction uniformity by a visual inspection method, and measuring an average construction depth index MTD (maximum Transmission distance) 0 Fractal dimension D 0 Then detecting the road section to be detected through the road surface image acquisition device (1) to obtain the MTD n and Dn Thereby calculating and obtaining the uniformity index of the asphalt pavement: average constructional depth ratio U MTD And fractal dimension ratio U D The specific formula is as follows:
wherein ,MTDn -constructing a depth for the average of the road segments to be detected;
MTD 0 -an average construction depth for the target road segment;
D n -a fractal dimension for the road segment to be detected;
D 0 -fractal dimension for the target road segment;
step 6: evaluation of Bituminous pavement construction uniformity
By fractal dimension ratio U D Judging the segregation degree of the asphalt pavement and evaluating the construction uniformity of the asphalt pavement;
the specific process for evaluating the uniformity of the asphalt pavement in the step 6 is as follows: based on the asphalt pavement segregation degree determination method specified in the existing standard NCHRPReport441, the asphalt pavement segregation degree determination method is implemented by the average construction depth ratio U MTD Instead of the structural ratio, the structural depth ratio U is determined by the critical average MTDC Calculating critical fractal dimension ratio U DC In step 1-step 5, the U of the road section to be detected is calculated D Then, judge U D Ratio of critical fractal dimension U DC And (3) judging the segregation degree of the asphalt surface layer so as to evaluate the construction uniformity of the asphalt pavement.
2. The machine vision-based asphalt pavement construction uniformity detection and evaluation method according to claim 1, wherein the road surface image acquisition device (1) comprises an image acquisition module (1-1), a main control computer (1-2) for reading information of the image acquisition module (1-1), an illumination module (1-3) for providing illumination for the image acquisition module (1-1), and a support frame (1-4) for fixing the image acquisition module (1-1) and the illumination module (1-3).
3. The machine vision-based asphalt pavement construction uniformity detection and evaluation method according to claim 2, wherein the support frame (1-4) comprises a panel (1-41) and a base (1-42) for supporting and adjusting the height of the panel (1-41), and the image acquisition module (1-1) and the illumination module (1-3) are fixed on one side of the panel (1-41) close to the pavement.
4. The machine vision-based asphalt pavement construction uniformity detection and evaluation method according to claim 3, wherein the image acquisition module (1-1) comprises a plurality of cameras which are uniformly distributed and arrayed at the same horizontal height at annular intervals, and one end of each camera far away from the panel (1-41) is inclined towards the direction close to the annular center.
5. The method for detecting and evaluating the construction uniformity of the asphalt pavement based on the machine vision according to claim 4, wherein the included angle between each camera and the horizontal plane is 45 degrees.
6. The machine vision-based asphalt pavement construction uniformity detection and evaluation method according to claim 1, wherein in the step 3, the point cloud reconstruction is obtained through a correspondence relationship between two camera image surface points, and specifically comprises the following steps:
wherein ρ is the two-dimensional coordinates of the spatial point of the target road surface area in the rest angle images;
thus, the three-dimensional coordinate mathematical relationship of the spatial point of the target road surface area is obtained as follows,
and thereby creating a road table three-dimensional texture model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210440015.4A CN114812450B (en) | 2022-04-25 | 2022-04-25 | Asphalt pavement construction uniformity detection and evaluation method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210440015.4A CN114812450B (en) | 2022-04-25 | 2022-04-25 | Asphalt pavement construction uniformity detection and evaluation method based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114812450A CN114812450A (en) | 2022-07-29 |
| CN114812450B true CN114812450B (en) | 2023-10-20 |
Family
ID=82506774
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210440015.4A Active CN114812450B (en) | 2022-04-25 | 2022-04-25 | Asphalt pavement construction uniformity detection and evaluation method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114812450B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116205865B (en) * | 2023-02-14 | 2023-09-26 | 重庆大学 | Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data |
| CN117218089B (en) * | 2023-09-18 | 2024-04-19 | 中南大学 | Asphalt pavement structure depth detection method |
| CN117854835B (en) * | 2024-01-05 | 2024-05-31 | 广东新亚光电缆股份有限公司 | Fireproof flame-retardant cable |
| CN118274735A (en) * | 2024-04-26 | 2024-07-02 | 同济大学 | Automatic testing system and method for asphalt pavement construction depth based on multi-vision |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003105675A2 (en) * | 2002-06-18 | 2003-12-24 | Lifespan Biosciences, Inc. | Computerized image capture of structures of interest within a tissue sample |
| CN104294745A (en) * | 2014-10-16 | 2015-01-21 | 武汉科技大学 | Detection method of skid resistance of asphalt pavement based on three-dimensional fractal dimension |
| CN112927366A (en) * | 2021-05-10 | 2021-06-08 | 中南大学 | Asphalt pavement structure depth measuring method |
| CN113051647A (en) * | 2021-03-30 | 2021-06-29 | 阜阳师范大学 | Pavement aggregate distribution uniformity evaluation method and system based on three-dimensional texture model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11645769B2 (en) * | 2020-03-14 | 2023-05-09 | Purdue Research Foundation | Pavement macrotexture determination using multi-view smartphone images |
-
2022
- 2022-04-25 CN CN202210440015.4A patent/CN114812450B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003105675A2 (en) * | 2002-06-18 | 2003-12-24 | Lifespan Biosciences, Inc. | Computerized image capture of structures of interest within a tissue sample |
| CN104294745A (en) * | 2014-10-16 | 2015-01-21 | 武汉科技大学 | Detection method of skid resistance of asphalt pavement based on three-dimensional fractal dimension |
| CN113051647A (en) * | 2021-03-30 | 2021-06-29 | 阜阳师范大学 | Pavement aggregate distribution uniformity evaluation method and system based on three-dimensional texture model |
| CN112927366A (en) * | 2021-05-10 | 2021-06-08 | 中南大学 | Asphalt pavement structure depth measuring method |
Non-Patent Citations (1)
| Title |
|---|
| 基于数字图像方法的沥青混合料离析的判定及标准;吴洪平;筑路机械与施工机械化;第35卷(06);77-82 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114812450A (en) | 2022-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114812450B (en) | Asphalt pavement construction uniformity detection and evaluation method based on machine vision | |
| Chen et al. | Real-time identification system of asphalt pavement texture based on the close-range photogrammetry | |
| Ouyang et al. | Pavement cracking measurements using 3D laser-scan images | |
| CN109632033B (en) | Volume measurement device and method | |
| Chang et al. | Detection of pavement distresses using 3D laser scanning technology | |
| CN106856002B (en) | Unmanned aerial vehicle shooting image quality evaluation method | |
| Yu et al. | 3D reconstruction of road surfaces using an integrated multi-sensory approach | |
| CN105279372B (en) | A kind of method and apparatus of determining depth of building | |
| Lin et al. | Applications of computer vision on tile alignment inspection | |
| CN104502990A (en) | Geological survey method of tunnel face based on digital image | |
| Xue et al. | Novel SfM-DLT method for metro tunnel 3D reconstruction and Visualization | |
| CN112215820A (en) | Tunnel face analysis method based on image data | |
| CN108665468B (en) | Device and method for extracting tangent tower insulator string | |
| Mathavan et al. | Pavement raveling detection and measurement from synchronized intensity and range images | |
| CN102297660A (en) | Measuring method of shield tunnel lining segment seam open width and apparatus thereof | |
| CN110503637B (en) | Road crack automatic detection method based on convolutional neural network | |
| CN109166125A (en) | A kind of three dimensional depth image partitioning algorithm based on multiple edge syncretizing mechanism | |
| CN114881987B (en) | Hot-pressing light guide plate defect visual detection method based on improvement YOLOv5 | |
| Sarker et al. | Multi-scale 3D roughness quantification of concrete interfaces and pavement surfaces with a single-camera set-up | |
| Dan et al. | Application of deep learning-based image recognition technology to asphalt–aggregate mixtures: Methodology | |
| Wang et al. | Automated pavement distress survey: a review and a new direction | |
| CN105072330A (en) | Linear array camera automatic focusing method | |
| CN102682435B (en) | Multi-focus image edge detection method based on space relative altitude information | |
| Jung et al. | Fast and non-invasive surface crack detection of press panels using image processing | |
| Zheng et al. | Three-dimensional translucent segregation table (3D-TST) test for soil particle size and shape distribution |
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 |