CN108765376A - A kind of line scanning three-dimensional pavement data component analysis method - Google Patents
A kind of line scanning three-dimensional pavement data component analysis method Download PDFInfo
- Publication number
- CN108765376A CN108765376A CN201810415946.2A CN201810415946A CN108765376A CN 108765376 A CN108765376 A CN 108765376A CN 201810415946 A CN201810415946 A CN 201810415946A CN 108765376 A CN108765376 A CN 108765376A
- Authority
- CN
- China
- Prior art keywords
- data
- low
- dimensional
- pavement
- cross
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Length Measuring Devices With Unspecified Measuring Means (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention provides a kind of line scanning three-dimensional pavement data component analysis method, and this method includes：Low-pass filter by the design of pavement component specificity analysis suitable for three-dimensional pavement data carries out frequency domain low-pass wave to the cross-section face data of three-dimensional pavement to be detected, obtains the low-frequency component and radio-frequency component corresponding to cross-section face data respectively；By Total Variation, above-mentioned radio-frequency component is further divided into sparse features data and vibration performance data；According to the low-frequency data, the sparse features data and the vibration performance data, be conducive to analyze and identify one or more in track, crack, graticule, pit slot and the texture in the three-dimensional pavement cross section to be detected.The present invention is based on the high-precision road table three-dimensional datas of line scanning three-dimensional measurement technical limit spacing, it can realize that three-dimensional pavement data typical case's pavement component information detaches, the complex scene information in the three-dimensional data of high-precision road surface is considered, the accurate extraction of three-dimensional pavement relevant marker information is more advantageous to.
Description
Technical field
The present embodiments relate to line scanning three-dimensional data processing technology fields more particularly to a kind of line to scan three-dimensional pavement
Data component analysis method.
Background technology
With the development of line scanning three-dimensional measurement technology, the data precision that three-dimension measuring system can obtain is higher and higher, example
Such as, the cross section precision of road surface three-dimension measuring system can reach 1mm, and height accuracy can reach 0.5mm.Utilize above-mentioned high-precision road
It include more complicated information of road surface in the three-dimensional pavement altitude data that face three-dimension measuring system obtains：Macroscopic information includes road surface
Geometry, surface deformation disease (track, pit slot etc.), pavement crack, pavement strip etc.；Microscopic information includes pavement structural depth
Deng.On the other hand, there is certain relationships that influences each other between these roadway scene informational contents, for example, pavement structure is deep
The exception of degree can cause the generation of the diseases such as crack, loose, surface deformation usually different along with crack or road surface macroscopic view geometry
Often etc..
That is, three-dimensional pavement altitude data includes simultaneously multiple road information, and exist between various information
Certain degree of association.When being analyzed using the road surface a certain index of three-dimensional data road pavement, inevitably need to obtain it
The information of his index of correlation provides related prior information, more accurately to extract index and improvement method universality, such as
Three-dimensional cracking extraction is related to pavement structural depth to a certain extent, and obtaining the construction depth on different road surfaces can further increase
The accuracy of crack extract.
The existing information of road surface extracting method based on two-dimensional visual characteristic：Information of road surface inspection based on two-dimensional visual feature
Survey method mainly by optical camera and video etc. obtain road surface data, by the color characteristics of a certain disease in road surface and index with
Color distortion between other indexs of road surface carries out single index extraction, such as the brightness by crack is carried less than road surface background
The brightness of crack, graticule is taken to detect graticule higher than road surface background.
The single index detection method in the existing road surface based on three-dimensional pavement data：Such method is usually using vehicle-mounted three-dimensional
Laser scanning data, such as line scan three peacekeeping laser radars etc., and a certain disease in road surface or index are utilized in conjunction with data precision
Elevation characteristic single index is detected, using the data of lower accuracy can obtain pavement track information, using compared with
High-precision data (lateral resolution 1mm, elevation resolution ratio≤0.5mm) can detect pavement crack information.
The above-mentioned existing method for three-dimensional pavement data processing, the information for being only absorbed in a certain index mostly carry
It takes, and it is also less for the method for the decomposition of three-dimensional pavement data component information.Therefore, point of three-dimensional pavement data component information
Solution method be current industry it is urgently to be resolved hurrily need project.
Invention content
The present invention provides a kind of line and scans three-dimensional pavement data component analysis method, to solve in the prior art only for
Single index extraction information of road surface realizes the information of road surface extracting method of more robust property and universality.
The present invention provides a kind of line scanning three-dimensional pavement data component analysis method, including：Pass through pavement component characteristic point
Low-pass filter of the analysis design suitable for three-dimensional pavement data, frequency domain low-pass is carried out to the cross-section face data of three-dimensional pavement to be detected
Wave obtains low-frequency component data and radio-frequency component data corresponding to the cross-section face data of three-dimensional pavement to be detected respectively,
In, the slow deformation data on road surface cross section to be detected, the radio-frequency component data are contained in the low-frequency component data
Including sparse features data and vibration performance data, the sparse features data contain on the road surface cross section to be detected
Local acute variation information, the vibration performance data contain the texture feature information on the road surface cross section to be detected,
The slow deformation information includes track, and the acute variation information includes one or more in crack, graticule and pit slot, institute
Stating the cutoff frequency of low-pass filter influences the elevation resolution of width and the three-dimensional data according to the surface deformation to be detected
Rate obtains；By Total Variation, the radio-frequency component data are further divided into the sparse features data and described are shaken
Dynamic characteristic；According to the low-frequency component data, the sparse features data and the vibration performance data, be conducive to analyze
With it is one or more in the track, crack, graticule, pit slot and the texture that identify in the road surface cross section to be detected.
Preferably, the low-pass filter by the design of pavement component specificity analysis suitable for three-dimensional pavement data, it is right
The cross-section face data of three-dimensional pavement to be detected carries out frequency domain low-pass wave, further includes before：It is designed by pavement component specificity analysis
Suitable for the low-pass filter of three-dimensional pavement data, the cross-section face data of three-dimensional pavement to be detected is become by Fourier transformation
It changes corresponding frequency domain data into, low-pass filtering is carried out in conjunction with frequency low-pass and the frequency domain data.
Preferably, the low-pass filter by the design of pavement component specificity analysis suitable for three-dimensional pavement data, it is right
The cross-section face data of three-dimensional pavement to be detected carries out frequency domain low-pass wave, obtains the cross-section face data of three-dimensional pavement to be detected respectively
Corresponding low-frequency component data and radio-frequency component data, specifically include：By designed low-pass filter to the frequency domain
Data are filtered, and obtain filtered frequency domain data, and the free transmission range of the low-pass filter isN indicates the length of the frequency domain data, fcIndicate cutting for the low-pass filter
Only frequency；Inversefouriertransform is carried out to filtered frequency domain data, obtains filtered low-frequency component data；It will be three-dimensional cross-section
After face data subtracts the filtered low-frequency component data of acquisition, the radio-frequency component data are obtained.
Preferably, the cutoff frequency of the low-pass filter is obtained according to following formula：
Wherein, fcIndicate that the cutoff frequency of the low-pass filter, R_x indicate that the cross section of the three-dimensional data is laterally divided
Resolution, W_x indicate that the surface deformation to be detected influences width.
Preferably, described according to the low-frequency component data, the sparse features data and the vibration performance data, have
Conducive to the crack information analyzed and identified in the road surface cross section to be detected, specially：Obtain the sparse features data
Texture in mean value and the vibration performance data is averaged fluctuation amplitude, and elevation in the sparse features data is less than the institute
State the mean value of sparse features data and elevation fluctuation amplitude be more than the texture be averaged fluctuation amplitude data it is doubtful as crack
Region.
Preferably, described according to the low-frequency component data, the sparse features data and the vibration performance data, have
Conducive to the graticule information analyzed and identified in the road surface cross section to be detected, specially：
It obtains the texture in the mean value and the vibration performance data of the sparse features data to be averaged fluctuation amplitude, by institute
Elevation in sparse features data is stated to be averaged more than the texture higher than the mean value and elevation fluctuation amplitude of the sparse features data
The data of fluctuation amplitude are as graticule suspicious region.
Preferably, described according to the low-frequency component data, the sparse features data and the vibration performance data, have
Conducive to analyzing and identifying texture information in the road surface cross section to be detected, specially：According to the area of the vibration performance data
Domain mean value and variance obtain the texture information in the road surface cross section to be detected.
A kind of line provided in an embodiment of the present invention scans three-dimensional pavement data component analysis method, based on the three-dimensional survey of line scanning
The high-precision road surface elevation three-dimensional data for measuring technical limit spacing can realize three-dimensional pavement data typical case's pavement component information point
From having fully considered that complex scene information and index in the three-dimensional data of high-precision road surface influence each other relationship, are more advantageous to three
Tie up the accurate extraction of road surface relevant marker information.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 (a) is that high-precision line scans three-dimensional pavement schematic diagram data；
Fig. 1 (b) is track data, graticule data, crack data, pit slot data and the texture for including in three-dimensional pavement data
Datagram；
Fig. 1 (c) is the abstract expression schematic diagram of three-dimensional pavement data component；
Fig. 2 is the flow chart that a kind of line of the embodiment of the present invention scans three-dimensional pavement data component analysis method；
Fig. 3 is pretreated three-dimensional pavement cross section schematic diagram data；
Fig. 4 is schematic diagram of the cross-section face data of three-dimensional pavement in frequency domain；
Fig. 5 is low-pass filter schematic diagram；
Fig. 6 is the three-dimensional pavement cross section schematic diagram data after frequency domain filtering；
Fig. 7 is the low-frequency component schematic diagram data of the cross-section face data of three-dimensional pavement；
Fig. 8 is the radio-frequency component schematic diagram data of the cross-section face data of three-dimensional pavement；
Fig. 9 is the sparse features schematic diagram data of the cross-section face data of three-dimensional pavement；
Figure 10 is the vibration performance data schematic diagram of the cross-section face data of three-dimensional pavement；
Figure 11 (a) be road surface cross section to be detected include crack when three-dimensional pavement cross section data depth turn gray-scale map；
Figure 11 (b) is the sparse features schematic diagram data for including crack in road surface cross section to be detected；
Figure 11 (c) is the crack schematic diagram in road surface cross section to be detected；
Figure 12 (a) be road surface cross section to be detected include graticule when three-dimensional pavement cross section data depth turn gray-scale map；
Figure 12 (b) is the sparse features schematic diagram data for including graticule in road surface cross section to be detected；
Figure 12 (c) is the graticule schematic diagram in road surface cross section to be detected；
Figure 13 (a) is that the three-dimensional pavement cross section data depth in the road surface cross section to be detected containing track turns gray-scale map；
Figure 13 (b) is the schematic diagram of vibration performance data；
The schematic diagram of Figure 13 (c) temporal low frequency data；
Figure 13 (d) is the schematic diagram of the track information of extraction.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
The every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
In order to better illustrate the course of work of the present invention, the principle implemented first is executed to the present invention below and is explained
It is bright.
The single index detection method in the existing road surface based on three-dimensional pavement data, utilizes a certain disease in road surface or index
Elevation characteristic single index is detected, using the data of lower accuracy can obtain pavement track information, using compared with
High-precision data (lateral resolution 1mm, elevation resolution ratio≤0.5mm) can detect pavement crack information.
But (lateral resolution 1mm, elevation resolution ratio≤0.5mm), three-dimensional road when the precision of three-dimensional data is sufficiently high
More complicated roadway scene information is just contained in the altitude data of face, not only there are the letters such as surface deformation, the road surface curvature of macroscopic view
Breath, also containing the information such as microcosmic crack, graticule, repairing or even pavement structural depth also can be in the three-dimensional pavement of the precision
It is embodied in data；And in high accuracy data, different types of disease is on different road surfaces in the presence of in various degree
Influence each other relationship.Such as pavement texture fluctuation is more similar to the depth characteristic in crack in the larger road surface of construction depth,
Just with penetration of fracture feature without considering that pavement structural depth influences, it will influence the robustness and reality of crack extract method
The property used.Therefore, contain between complicated scene information and all kinds of indexs that there are mutual shadows in the data of high accuracy three-dimensional road surface
It rings, if analyzing influence of a certain index without considering other indexs only by simple elevation information, it will influence method
Robustness and universality.
Therefore, in order to find the relationship that influences each other between various information of road surface, below by by crack in road pavement,
The data characteristicses such as graticule, pit slot, texture and track are analyzed, and Fig. 1 (a) is that high-precision line scans the signal of three-dimensional pavement data
Figure, Fig. 1 (b) are track data, graticule data, crack data, pit slot data and the data texturing for including in three-dimensional pavement data
Schematic diagram, Fig. 1 (c) be three-dimensional pavement data component abstract expression schematic diagram, from figure 1 it appears that crack is in three-dimensional
More sharp downward thorn-like characteristic is presented in the cross section of road surface, and graticule then shows more regular elevation step protrusion spy
Property, the edge of pit slot is also generally configured with the characteristic drastically declined, and from the perspective of frequency domain, these three ingredients all have high frequency spy
Property, and have sparse characteristic, therefore, crack, graticule and pit slot can be summarized as sparse features data.
Slowly deformation and road surface nominal contour etc. all do not include radio-frequency component usually to pavement track etc., in three-dimensional cross section
Belong to low-frequency component in data.
And pavement texture shows the quick fluctuation characteristic in particular range in three-dimensional pavement cross section, is also high frequency spy
Property, compared to pavement crack, graticule etc., pavement texture fluctuation does not have sparse characteristic.
In addition, in the cross-section face data of three-dimensional pavement, from the point of view of spatial domain, each constituents are typically mutual aliasing, example
Such as, three-dimensional cracking extraction usually requires to consider the influence of pavement texture background；And it is also required to remove when assessing pavement structural depth
It is mixed in the influence in the crack in pavement texture.
The above analysis, for the cross-section face data y of any three-dimensional pavement, can centainly be divided into radio-frequency component data and
Low-frequency component data, and radio-frequency component data include sparse features data and vibration performance data, that is to say, that it can be three-dimensional
Road surface cross section data representation is as follows：
F=f+x+t,
Wherein, y indicates the three-dimensional pavement cross section altitude data of input, the length of N；F indicates low-frequency component data, can
To characterize the information such as track, nominal contour, slowly varying deformation disease；X is road surface sparse features data, can characterize and split
Seam, pit slot disease or artificial graticule edge etc. have catastrophe characteristics and account for smaller information in the cross section of road surface；T is road
Surface vibration characteristic can characterize the fluctuation information of pavement texture.
F is low-frequency component, and compared to x and t, frequency content is extremely low, can obtain f first by low-pass filter.By t
It is modeled as meeting the signal of statistics white Gaussian noise characteristic, sets its variance as σ2.And x has sparse characteristic, it is poor to combine
Equation is divided to be expressed.
According to the above analytic process, it can be seen that the cross-section face data of any one three-dimensional pavement can be expressed as low frequency
Compositional data and radio-frequency component data include sparse features data and vibration performance data again in radio-frequency component data.Based on this
Conclusion is below described specific carry into execution a plan of the embodiment of the present invention, and Fig. 2 is that one kind of the embodiment of the present invention is filtered based on frequency domain
The flow chart for involving the line scanning three-dimensional pavement data component analysis method of Total Variation, as shown in Fig. 2, this method includes：
Low-pass filter by the design of pavement component specificity analysis suitable for three-dimensional pavement data, to three-dimensional pavement cross section to be detected
Data carry out frequency domain low-pass wave, obtain the low-frequency component data corresponding to the cross-section face data of three-dimensional pavement to be detected respectively
With radio-frequency component data, wherein the slow deformation data on road surface cross section to be detected is contained in the low-frequency component data,
The radio-frequency component data include sparse features data and vibration performance data, and the sparse features data contain described to be checked
The acute variation information on the cross section of road surface is surveyed, the vibration performance data contain the line on the road surface cross section to be detected
Characteristic information is managed, the slow deformation information includes track, and the acute variation information includes one in crack, graticule and pit slot
Kind is a variety of, and the cutoff frequency of the low-pass filter influences width and the three-dimensional data according to the surface deformation to be detected
Elevation resolution ratio obtain；
By Total Variation, the radio-frequency component data are further divided into the sparse features data and described are shaken
Dynamic characteristic；
According to the low-frequency component data, the sparse features data and the vibration performance data, be conducive to analysis and
It identifies one or more in track, crack, graticule, pit slot and the texture in the road surface cross section to be detected.
Before this, it is also necessary to carry out following steps：First, the cross-section face data of three-dimensional pavement to be detected is obtained, it is to be detected
The cross-section face data of three-dimensional pavement is to measure obtained data to road surface to be detected by line scanning three-dimensional measurement sensor,
The three-dimensional measurement sensor obtains measured object surface relative elevation situation based on principle of triangulation measurement, and to be detected the three of acquisition
The dimension cross-section face data in road surface can reflect the elevation information on measured object surface.
Line scanning three-dimensional measurement sensor can realize same posture, synchronization profiled outline synchro measure, acquisition
Mode include two ways：First, three-dimensional measurement sensor is mounted on fixing bracket, in three-dimensional measurement sensor measurement model
In enclosing, testee passes through measured zone with certain speed, in testee motion process, realizes to testee profile
The acquisition of three-dimensional data；Second, three-dimensional measurement sensor is mounted on motion carrier, during measuring carrier movement, to quilt
The three-dimensional data for surveying contour of object is acquired.
Three-dimensional measurement sensor is mounted on motion carrier in data acquisition, it is right during measuring carrier movement
Testee three-D profile carries out data acquisition.
Due to the interference of measuring environment, such as road surface is water stain, oil stain or tested region have foreign matter, collected data may
There are a small amount of extraordinary noise (zero points), and therefore, it is necessary to be located in advance to the collected cross-section face data of three-dimensional pavement to be detected
Reason, specific pre-treatment step are：Abnormality value removing and data scaling processing are carried out to the cross-section face data of three-dimensional pavement to be detected.
Since line scanning three-dimensional measurement sensor is made of area array cameras with the mode that laser line generator is combined, camera
Distortion at center is minimum, and collected road surface cross section three-dimensional data is stablized the most near section central point, and the present invention is real
Applying example utilizes the non-abnormal sample point close to section central area to replace extraordinary noise point, obtains image space profile data.
Area array cameras is with the road surface three-dimension measuring system of high-power laser line generator composition, and there is sensor established angles
Degree, laser rays collimation, the unequal systematic error of laser intensity distribution.These systematic errors will weaken road surface interesting target
Feature, it is therefore desirable to the data of three-dimensional measurement sensor acquisition are corrected by demarcating file, while image space data being turned
Change object space data y into.
Just because of on frequency domain, the cross-section face data of three-dimensional pavement to be detected can be divided into radio-frequency component data and low frequency
Therefore compositional data first passes through Fourier transformation, the collected cross-section face data of three-dimensional pavement to be detected is transformed into frequency
Then domain is again filtered frequency domain data by low-pass filter, the low frequency in the cross-section face data of three-dimensional pavement to be detected
Compositional data and radio-frequency component data separation come.
The most important problem of low-pass filter is that the cutoff frequency of low-pass filter how is arranged,
The present embodiments relate to system acquired in the cross-section face data of three-dimensional pavement to be detected, the wave of data component
It moves in addition to related to the fluctuation of ingredient itself, it is also related to the resolution ratio of data.Used by data of the embodiment of the present invention
The resolution ratio of the cross-section face data of road surface three-dimensional pavement is known (profile data length N=2048, lateral resolution R_x=
1mm, elevation resolution ratio R_z=0.1mm), this link mainly in combination with the data component itself under fixed resolution fluctuation, with
Obtain the cutoff frequency f of the low-frequency component and radio-frequency component differentiation of the cross-section face data of three-dimensional pavementc。
Under normal conditions, it is contemplated that the factors such as road drainage, there are certain radians for bituminous paving；On the other hand, pitch
Road surface usually there will be variation slowly but lateral extent is larger, and track of the depth more than 10mm can influence traffic safety to need
It is detected, the width W_r that in addition unilateral track influences is generally 0.5m-1m.
T=W_r/R_x；
T_m=min (T)；
fc=1/T_m；
When the influence width of track is 0.5m-1m, it is believed that the minimum period T_m in the data of R_x=1mm is 500.
So for the cutoff frequency f of the low-frequency component allowed in sectionc=1/500=0.002.
Fourier transformation, frequency domain ideal low pass filtered are utilized in conjunction with cutoff frequency for the profile data that data length is N
Wave and inversefouriertransform carry out low-pass filtering to data, for the cross-section face data y of three-dimensional pavement to be detected to be carried out low frequency
The separation of compositional data f and radio-frequency component data, radio-frequency component data include sparse features data x and vibration performance data w.
Therefore, the cross-section face data of three-dimensional pavement to be detected is filtered by low-pass filter, and according to filtered
Frequency domain data obtains low-frequency component data and radio-frequency component data, the specific steps are：
(1) face data y cross-section to three-dimensional pavement to be detected carries out Fourier transformation：Utilize the quick of discrete Fourier transform
Algorithm carries out Fourier transformation to the discrete cross-section face data y of three-dimensional pavement to be detected, and the length of y is N, by the Fourier of acquisition
Transform sequence is denoted as Y, and the length of Y is also N (wherein [N/2,1] corresponding frequency range is [0,1]), indicates that three-dimensional pavement to be detected is horizontal
The frequency distribution information of profile data y.In conjunction with Fourier transformation correlation theories knowledge it is found that Y is centrosymmetric, centre is corresponding
For the low-frequency component of signal y, both sides correspond to the radio-frequency component of signal y.
(2) frequency domain ideal low-pass filter：For the sequence Y that above-mentioned steps obtain, data characteristic and road are utilized in conjunction with above-mentioned
The acquired low-frequency component cutoff frequency f of face low-frequency component analysisc, low-pass filtering is carried out to sequence in frequency domain.So for length
Degree is the sequence Y of N, retains Y intermediate points or so each fc* the frequency of N number of point is (i.e. intermediate's
Frequency amplitude retains), and the frequency amplitude of Y other parts is all set to 0, it is Y_ by the sequence mark after frequency domain low-pass filtering
LF。
(3) inversefouriertransform is carried out to frequency sequence Y_LF：Inversefouriertransform is carried out to frequency sequence Y_LF and is taken
Its real part obtains its corresponding low-frequency component data, is y_lf by the low-frequency component data markers of acquisition, and y_lf is to wait for
Detect the low-frequency component data f of the cross-section face data y of three-dimensional pavement.
After the low-frequency component data f for obtaining the cross-section face data y of three-dimensional pavement to be detected, (y-f) is remainder
Radio-frequency component data h, that is, the sum of sparse features data x and vibration performance data w in model.
After solving low-frequency component data and radio-frequency component data, then by Total Variation, further by high frequency at
Divided data is divided into sparse features data and vibration performance data.
For containing the Noise reducing of data problem of sparse characteristic and sparse derivative (e.g. piece-wise constants) characteristic, full variation
(Total Variation Denoising, abbreviation TVD) has generally acknowledged stick signal details and effectively removes the spy of noise
Point.For noise-containing sparse signal, classical TVD utilizes lagrange's method of multipliers by building sparse optimization object function
Conditional extremum is obtained, converts sparse Solve problems to signal model energy functional minimization problem, resolving is as follows：
For discrete-time series, the definition of first differential matrix D (N-1) * N is：
X is sparse derivative (sparse derivative) signal, and w is that meet variance be σ2White Gaussian noise signal, when
When h=x+w, according to classical Total Variation, have
arg min||Dx||1L1, norm regularization meets sparsity.
Constraints：Two norms, the air line distances of two vectors in space.
It can convert above-mentioned minimization problem to following object function by lagrange's method of multipliers：
Using above formula, by radio-frequency component data h sparse features data x and vibration performance data w solve, wherein
Parameter lambda is regularization parameter, and the weight shared in optimization of two parts for adjusting composition object function, value should meet
λ>0, value setting is generally proportional to the standard deviation of oscillating component in signal, and λ takes 1.2 more suitable (roads in this application
The ranging from 1-2mm of face macrostructure depth).
Finally, according to low-frequency component data, sparse features data and vibration performance data, the identification road surface to be detected is horizontal
It is one or more in track, crack, graticule, pit slot and texture in section.
The cross-section face data of three-dimensional pavement is decomposed into low frequency component, sparse component and shaken by binding model of the embodiment of the present invention
Dynamic component can obtain the low-frequency component of three-dimensional pavement, sparse features data and shake respectively after being spliced each component
Dynamic characteristic, for the accurate extraction of all kinds of indexs in road surface and defect information.
The practical application requests such as combining road Defect inspection and maintenance, mainly using crack and graticule as example index come
Verify the sparse features data of model decomposition；Using there are the cross-section face datas of the three-dimensional pavement of different configuration depth to refer to as example
The vibration performance data of mark verification model decomposition；And the accuracy of low-frequency component, then combine referenced patent 201710861318.2
Disclosed envelope method verifies the accuracy of the low-frequency component of model decomposition.
For the crack information in sparse component, usually less than normal road surface.According to this feature, by sparse features data
Mean value is x_aver, for can be with the fluctuation amplitude t_a of statistic texture, by sparse features in the radio-frequency component data h of acquisition
It is less than mean value or less in data and fluctuating range is more than the region point of t_a, i.e., elevation is less than x_aver-t_ in sparse features data
The point set of a is as crack suspicious region.Each adjacent sections can obtain crack information after splicing.For graticule edge
It takes similar mode, graticule to be usually above normal road surface, mean value or more will be higher than in sparse features data and fluctuating range is big
In the region point of t_a, i.e., point set of the elevation higher than x_aver+t_a is as graticule suspicious region in sparse features data.
And for the vibration performance data that model obtains, the only regional average value of vibration performance data in the embodiment of the present invention
The size of oscillating component is characterized with variance.
In order to verify the validity and reliability of said program, the embodiment of the present invention is with the asphalt road containing crack, graticule
For face three-dimensional data and bituminous paving three-dimensional data containing different configuration depth, describe based on line scanning three-dimensional measurement
Bituminous pavement data component analyzing method.
Due to the interference (road surface is water stain, oil stain or tested region have foreign matter) of measuring environment, collected data may deposit
Extraordinary noise (zero point) in part, this patent utilize the non-abnormal sample point close to section central area to replace extraordinary noise
Point；Using demarcating file, correct in the object profiled outline of three-dimensional measurement sensor measurement because of sensor installation, laser rays radian
And systematic error caused by light distribution unevenness, while by image space data conversion at object space data.Simultaneously by pretreated one
Serial section is spliced along direction of traffic, obtains the cross-section face data of pitch three-dimensional pavement.
For common road surface Indexs measure application, the cross-section face data y of three-dimensional pavement is modeled as to the ingredient of three types：
Y=f+x+t,
Wherein, y indicates the cross-section face data of three-dimensional pavement of input, the length of N；F is road surface low-frequency component data, can be with
Characterize the slow deformation informations such as pavement track；X is road surface sparse features data, can characterize crack, pit slot disease or artificial
Graticule edge etc. has catastrophe characteristics and accounts for smaller information in the cross section of road surface；T is road vibration characteristic.f,x,
The length of t is N.
For acquired three-dimensional pavement cross-section face data y, data length N, data lateral resolution is set as R_
X, elevation resolution ratio are R_z, and the slow deformation effect width in road surface is W_r.Pavement Evaluation length is T, road surface arc cutoff frequency
For fc。
T=W_r/R_x；
T_m=min (T),
fc=1/T_m,
In conjunction with actual conditions, W_r is generally 0.5m-1m.It is analyzed using above formula, is the data of 1mm for resolution ratio, section
Only frequency fc=0.002.
For the cross-section face data y of three-dimensional pavement, in conjunction with cutoff frequency fc, utilize Fourier transformation, frequency domain ideal low pass filtered
Wave and inversefouriertransform carry out low-pass filtering to data, for the cross-section face data y of three-dimensional pavement to be carried out low-frequency component number
According to the separation of f and radio-frequency component data, radio-frequency component data include sparse features data x and vibration performance data w.
The process embodiments of the cross-section face data low frequency of three-dimensional pavement, radio-frequency component separation that are carried out based on frequency domain low-pass wave
As shown, Fig. 3 is pretreated three-dimensional pavement cross section schematic diagram data, the pretreated three-dimensional pavement cross section number
It is indicated according to y；Fig. 4 is schematic diagram of the cross-section face data of three-dimensional pavement in frequency domain, should be in the cross-section face data of three-dimensional pavement of frequency domain
It is indicated with Y；Fig. 5 is low-pass filter schematic diagram, which is to utilize fcThe frequency domain ideal low-pass filter L of design,
Its cutoff frequency is fc；Fig. 6 is the three-dimensional pavement cross section schematic diagram data after frequency domain filtering, after being filtered to Y using L
The frequency distribution Y_LF of acquisition；Fig. 7 is the low-frequency component schematic diagram data of the cross-section face data of three-dimensional pavement, as shown in fig. 7, to frequency
Rate sequence Y_LF carries out inversefouriertransform, and takes its real part, obtains its corresponding low-frequency component, as signal y's is low
Frequency compositional data f；Fig. 8 is the radio-frequency component schematic diagram data of the cross-section face data of three-dimensional pavement, as shown in figure 8, by three-dimensional pavement
After the y removals low-frequency component data f of cross-section face data, remaining radio-frequency component data h, including vibration performance data t and dilute
Dredge characteristic x.
Using full variation solving model by h sparse features data x and vibration performance data t solve, embodiment
As shown in Figure 9 and Figure 10, Fig. 9 is the sparse features schematic diagram data of the cross-section face data of three-dimensional pavement；Figure 10 is that three-dimensional pavement is horizontal
The vibration performance data schematic diagram of profile data.
Using the low-frequency component data f of above-mentioned acquisition, sparse features data x, vibration performance data t, each ingredient is by spelling
Ingredient spliced map is obtained after connecing.Accordingly for the sparse features data of acquisition, this link utilizes in sparse features data
Crack information and graticule information are verified.And for the vibration performance data that model obtains, only Vibration parameter in this patent
According to regional average value and variance characterize the size of vibration performance data.Figure 11 is sparse to test using the cross-section face data of three-dimensional pavement
The validity of characteristic, Figure 11 (a) be road surface cross section to be detected include crack when three-dimensional pavement cross section data depth
Turn gray-scale map, Figure 11 (b) is the sparse features schematic diagram data for including crack in road surface cross section to be detected, and Figure 11 (c) is to wait for
Detect the crack schematic diagram in the cross section of road surface.
Figure 12 (a) be road surface cross section to be detected include graticule when three-dimensional pavement cross section data depth turn gray-scale map,
Figure 12 (b) is the sparse features schematic diagram data for including graticule in road surface cross section to be detected, and Figure 12 (c) is that road surface to be detected is horizontal
Graticule schematic diagram in section.
From in Figure 11 and Figure 12 as can be seen that solved by model come sparse features data in contain it is more complete
Whole crack information and graticule information.
In addition to the validity of verification low-frequency component data, passes through the above-mentioned cross-section face data of the three-dimensional pavement containing track
It verifies, Figure 13 (a) is that the three-dimensional pavement cross section data depth in road surface cross section to be detected containing track turns gray-scale map, is schemed
13 (b) is the schematic diagram of vibration performance data, and the schematic diagram of Figure 13 (c) low-frequency component data, Figure 13 (d) is the track letter of extraction
The schematic diagram of breath.It can be used for carrying for the slow deformation information such as track in the low-frequency component data extracted as can be seen from Figure 13
It takes.
To sum up, the embodiment of the present invention is by pretreatment, to the road surface section profile of three-dimensional measurement sensor measurement because measuring
The abnormal zero noise spot in part caused by environmental disturbances is handled, and image space profiled outline is obtained；Using demarcating file, effective school
It is caused by sensor installation, laser rays radian and light intensity unevenness in the road surface section profile of positive three-dimensional measurement sensor measurement
System error, and conversion of the image space to object space is carried out, the true object space profiled outline information for being tested road surface is obtained, is subsequent mark
Line detects and information extraction provides good data input.
The embodiment of the present invention utilizes the certain ingredients of characteristic and road surface comprising Multiple components in three-dimensional pavement scene that can use
The characteristics such as sparsity, vibratility and low frequency characterize respectively, to by sparse characteristic by the crack of roadway scene, graticule
Edge, this kind of part in pit slot edge information jumpy are sparse ingredient；Pavement texture is fluctuated by vibration characteristics
Characteristic is abstracted as vibration component, and the characteristic of road surface smooth variation is abstracted as low-frequency component by low frequency characteristic, by being abstracted table
Three-dimensional pavement typical composition type after reaching includes：Low-frequency information ingredient, sparse ingredient and vibration component.
The embodiment of the present invention utilizes taken three-dimensional data resolution character and length characteristic, combining road slowly to deform
Scoped features, determine that the cutoff frequency that low-frequency component and radio-frequency component are distinguished in three-dimensional pavement data, construction frequency domain are ideal
Low-pass filter, and the cross-section face data of three-dimensional pavement is decomposed into low-frequency component and radio-frequency component.
Combination of embodiment of the present invention Total Variation for solution have sparse characteristic noisy acoustical signal have it is good
The noise removal capability for keeping signal detail characteristic, contain while above-mentioned steps are obtained the high frequency of vibration component and sparse ingredient at
Point, sparse ingredient and vibration component are obtained by full Variational Decomposition model.
Combination decomposition model of the embodiment of the present invention, by road surface three-dimensional data decompose obtain sparse ingredient, low-frequency component with
And vibration component, and split using threshold method acquisition after the sparse ingredient of model acquisition using the data containing crack, graticule
Seam and graticule region, and be compared with labeled data, the validity of illustration method.
The apparatus embodiments described above are merely exemplary, wherein the unit illustrated as separating component can
It is physically separated with being or may not be, the component shown as unit may or may not be physics list
Member, you can be located at a place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of module achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case of, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be expressed in the form of software products in other words, should
Computer software product can store in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, it will be understood by those of ordinary skill in the art that：It still may be used
With technical scheme described in the above embodiments is modified or equivalent replacement of some of the technical features；
And these modifications or replacements, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (7)
1. a kind of line scans three-dimensional pavement data component analysis method, which is characterized in that including：
Low-pass filter by the design of pavement component specificity analysis suitable for three-dimensional pavement data, to three-dimensional pavement to be detected cross
Profile data carries out frequency domain low-pass wave, obtains the low-frequency component corresponding to the cross-section face data of three-dimensional pavement to be detected respectively
Data and radio-frequency component data, wherein the slow deformation on road surface cross section to be detected is contained in the low-frequency component data
Information, the radio-frequency component data include sparse features data and vibration performance data, and the sparse features data contain institute
The local acute variation information on road surface cross section to be detected is stated, it is horizontal that the vibration performance data contain the road surface to be detected
Texture feature information on section, the slow deformation information includes track, the acute variation information include crack, graticule and
One or more in pit slot, the cutoff frequency of the low-pass filter influences width and institute according to the surface deformation to be detected
The elevation resolution ratio for stating three-dimensional data obtains；
By Total Variation, the radio-frequency component data are further divided into the sparse features data and the vibration is special
Levy data；
According to the low-frequency component data, the sparse features data and the vibration performance data, is conducive to analyze and identify
It is one or more in track, crack, graticule, pit slot and texture in the road surface cross section to be detected.
2. method according to claim 1, which is characterized in that described to be suitable for three-dimensional by the design of pavement component specificity analysis
The low-pass filter of road surface data carries out frequency domain low-pass wave to the cross-section face data of three-dimensional pavement to be detected, further includes before：
Low-pass filter by the design of pavement component specificity analysis suitable for three-dimensional pavement data, by Fourier transformation by institute
It states the cross-section face data of three-dimensional pavement to be detected and is transformed into corresponding frequency domain data, in conjunction with frequency low-pass and the frequency domain number
According to progress low-pass filtering.
3. method according to claim 2, which is characterized in that described to be suitable for three-dimensional by the design of pavement component specificity analysis
The low-pass filter of road surface data carries out frequency domain low-pass wave, respectively described in acquisition to the cross-section face data of three-dimensional pavement to be detected
Low-frequency component data corresponding to the cross-section face data of three-dimensional pavement to be detected and radio-frequency component data, specifically include：
The frequency domain data is filtered by designed low-pass filter, obtains filtered frequency domain data, it is described low
The free transmission range of bandpass filter is [N/2-fc*N/2,N/2+fc* N/2], N indicates the length of the frequency domain data, fcDescribed in expression
The cutoff frequency of low-pass filter；
Inversefouriertransform is carried out to filtered frequency domain data, obtains filtered low-frequency component data；
After the filtered low-frequency component data that three-dimensional cross-section face data is subtracted to acquisition, the radio-frequency component number is obtained
According to.
4. according to claim 1 or 3 the methods, which is characterized in that the cutoff frequency of the low-pass filter is according to following public
Formula obtains：
Wherein, fcIndicate that the cutoff frequency of the low-pass filter, R_x indicate the cross section lateral resolution of the three-dimensional data,
W_x indicates that the surface deformation to be detected influences width.
5. method according to claim 1, which is characterized in that described according to the low-frequency component data, the sparse features
Data and the vibration performance data are conducive to analyze and identify the crack information in the road surface cross section to be detected, specifically
For：
The texture in the mean value and the vibration performance data of the sparse features data is obtained to be averaged fluctuation amplitude, it will be described dilute
Elevation in characteristic is dredged averagely to fluctuate more than the texture less than the mean value and elevation fluctuation amplitude of the sparse features data
The data of amplitude are as crack suspicious region.
6. method according to claim 1, which is characterized in that described according to the low-frequency component data, the sparse features
Data and the vibration performance data are conducive to analyze and identify the graticule information in the road surface cross section to be detected, specifically
For：
The texture in the mean value and the vibration performance data of the sparse features data is obtained to be averaged fluctuation amplitude, it will be described dilute
Elevation in characteristic is dredged averagely to fluctuate more than the texture higher than the mean value and elevation fluctuation amplitude of the sparse features data
The data of amplitude are as graticule suspicious region.
7. method according to claim 1, which is characterized in that described according to the low-frequency component data, the sparse features
Data and the vibration performance data are conducive to analyze and identify texture information in the road surface cross section to be detected, specially：
According to the regional average value and variance of the vibration performance data, the texture letter in the road surface cross section to be detected is obtained
Breath.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810415946.2A CN108765376B (en) | 2018-05-03 | 2018-05-03 | Linear scanning three-dimensional pavement data component analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810415946.2A CN108765376B (en) | 2018-05-03 | 2018-05-03 | Linear scanning three-dimensional pavement data component analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108765376A true CN108765376A (en) | 2018-11-06 |
| CN108765376B CN108765376B (en) | 2022-03-22 |
Family
ID=64009660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810415946.2A Active CN108765376B (en) | 2018-05-03 | 2018-05-03 | Linear scanning three-dimensional pavement data component analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108765376B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110473187A (en) * | 2019-08-08 | 2019-11-19 | 武汉武大卓越科技有限责任公司 | A kind of line scanning three-dimensional pavement crack extract method of object-oriented |
| CN110675392A (en) * | 2019-09-29 | 2020-01-10 | 山东科技大学 | Rut fine three-dimensional feature extraction method based on pavement continuous laser point cloud |
| CN111882664A (en) * | 2020-07-08 | 2020-11-03 | 武汉武大卓越科技有限责任公司 | Multi-window accumulated difference crack extraction method |
| CN111968079A (en) * | 2020-07-28 | 2020-11-20 | 武汉武大卓越科技有限责任公司 | Three-dimensional pavement crack extraction method based on section local extreme value and segmentation sparsity |
| CN116718132A (en) * | 2023-05-06 | 2023-09-08 | 深圳大学 | Pavement three-dimensional measurement method and system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3266833B2 (en) * | 1997-07-25 | 2002-03-18 | 本田技研工業株式会社 | Vehicle with contrasting object recognition function and road surface determination method |
| JP5175549B2 (en) * | 2004-06-09 | 2013-04-03 | コグネックス・テクノロジー・アンド・インベストメント・コーポレーション | Method and apparatus for visual detection and inspection of objects |
| CN103473745A (en) * | 2013-09-16 | 2013-12-25 | 东南大学 | Low-dosage CT image processing method based on distinctive dictionaries |
| CN104504721A (en) * | 2015-01-08 | 2015-04-08 | 中国科学院合肥物质科学研究院 | Unstructured road detecting method based on Gabor wavelet transformation texture description |
| CN105113375A (en) * | 2015-05-15 | 2015-12-02 | 南京航空航天大学 | Pavement cracking detection system and method based on line structured light |
| CN105678770A (en) * | 2016-01-07 | 2016-06-15 | 潘燕 | Wall crack detection device excellent in profile identification and filtering performances |
| US20160305783A1 (en) * | 2008-08-29 | 2016-10-20 | Sony Corporation | Velocity calculation device, velocity calculation method, and navigation device |
| CN107201713A (en) * | 2016-03-16 | 2017-09-26 | 南京理工技术转移中心有限公司 | Pavement crack Defect inspection method and system |
| CN107610094A (en) * | 2017-08-02 | 2018-01-19 | 长安大学 | Based on the three-dimensional crack 3 D detection method characterized of ellipsoid |
| CN107871133A (en) * | 2017-12-27 | 2018-04-03 | 中公高科养护科技股份有限公司 | The recognition methods of the optimization method, pavement disease of rim detection network and system |
-
2018
- 2018-05-03 CN CN201810415946.2A patent/CN108765376B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3266833B2 (en) * | 1997-07-25 | 2002-03-18 | 本田技研工業株式会社 | Vehicle with contrasting object recognition function and road surface determination method |
| JP5175549B2 (en) * | 2004-06-09 | 2013-04-03 | コグネックス・テクノロジー・アンド・インベストメント・コーポレーション | Method and apparatus for visual detection and inspection of objects |
| US20160305783A1 (en) * | 2008-08-29 | 2016-10-20 | Sony Corporation | Velocity calculation device, velocity calculation method, and navigation device |
| CN103473745A (en) * | 2013-09-16 | 2013-12-25 | 东南大学 | Low-dosage CT image processing method based on distinctive dictionaries |
| CN104504721A (en) * | 2015-01-08 | 2015-04-08 | 中国科学院合肥物质科学研究院 | Unstructured road detecting method based on Gabor wavelet transformation texture description |
| CN105113375A (en) * | 2015-05-15 | 2015-12-02 | 南京航空航天大学 | Pavement cracking detection system and method based on line structured light |
| CN105678770A (en) * | 2016-01-07 | 2016-06-15 | 潘燕 | Wall crack detection device excellent in profile identification and filtering performances |
| CN107201713A (en) * | 2016-03-16 | 2017-09-26 | 南京理工技术转移中心有限公司 | Pavement crack Defect inspection method and system |
| CN107610094A (en) * | 2017-08-02 | 2018-01-19 | 长安大学 | Based on the three-dimensional crack 3 D detection method characterized of ellipsoid |
| CN107871133A (en) * | 2017-12-27 | 2018-04-03 | 中公高科养护科技股份有限公司 | The recognition methods of the optimization method, pavement disease of rim detection network and system |
Non-Patent Citations (2)
| Title |
|---|
| IVANW. SELESNICK 等: "Simultaneous Low-Pass Filtering and Total Variation Denoising", 《IEEE TRANSACTIONS ON SIGNAL PROCESSING》 * |
| 李清泉 等: "利用高精度三维测量技术进行路面破损检测", 《武汉大学学报·信息科学版》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110473187A (en) * | 2019-08-08 | 2019-11-19 | 武汉武大卓越科技有限责任公司 | A kind of line scanning three-dimensional pavement crack extract method of object-oriented |
| CN110473187B (en) * | 2019-08-08 | 2022-03-22 | 武汉光谷卓越科技股份有限公司 | Object-oriented line scanning three-dimensional pavement crack extraction method |
| CN110675392A (en) * | 2019-09-29 | 2020-01-10 | 山东科技大学 | Rut fine three-dimensional feature extraction method based on pavement continuous laser point cloud |
| CN110675392B (en) * | 2019-09-29 | 2022-04-22 | 山东科技大学 | Rut fine three-dimensional feature extraction method based on pavement continuous laser point cloud |
| CN111882664A (en) * | 2020-07-08 | 2020-11-03 | 武汉武大卓越科技有限责任公司 | Multi-window accumulated difference crack extraction method |
| CN111968079A (en) * | 2020-07-28 | 2020-11-20 | 武汉武大卓越科技有限责任公司 | Three-dimensional pavement crack extraction method based on section local extreme value and segmentation sparsity |
| CN111968079B (en) * | 2020-07-28 | 2023-11-17 | 武汉光谷卓越科技股份有限公司 | Three-dimensional pavement crack extraction method based on local extremum of section and segmentation sparsity |
| CN116718132A (en) * | 2023-05-06 | 2023-09-08 | 深圳大学 | Pavement three-dimensional measurement method and system |
| CN116718132B (en) * | 2023-05-06 | 2024-03-26 | 深圳大学 | Pavement three-dimensional measurement method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108765376B (en) | 2022-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108765376A (en) | A kind of line scanning three-dimensional pavement data component analysis method | |
| AU2016385541B2 (en) | Object surface deformation feature extraction method based on line scanning three-dimensional point Cloud | |
| CN105809668B (en) | The body surface deformation behaviour extracting method of three-dimensional point cloud is scanned based on line | |
| Radopoulou et al. | Patch detection for pavement assessment | |
| Li et al. | Integrated processing of image and GPR data for automated pothole detection | |
| Jahanshahi et al. | Unsupervised approach for autonomous pavement-defect detection and quantification using an inexpensive depth sensor | |
| Wang et al. | Pavement crack width measurement based on Laplace's equation for continuity and unambiguity | |
| Zalama et al. | Road crack detection using visual features extracted by Gabor filters | |
| Ouyang et al. | Pavement cracking measurements using 3D laser-scan images | |
| Zhang et al. | Automatic road-marking detection and measurement from laser-scanning 3D profile data | |
| Ma et al. | DEM refinement by low vegetation removal based on the combination of full waveform data and progressive TIN densification | |
| Wang et al. | Automated pavement distress survey: a review and a new direction | |
| CN115797910A (en) | Long-tail target identification method, computer-readable storage medium and vehicle | |
| Xu et al. | Vision-based systems for structural deformation measurement: Case studies | |
| CN113408519B (en) | Method and system for pointer instrument reading based on template rotation matching | |
| Bian et al. | Lidar based edge-detection for bridge defect identification | |
| Jiang | A crack detection and diagnosis methodology for automated pavement condition evaluation | |
| Laureshyn et al. | Automated video analysis as a tool for analysing road user behaviour | |
| Alayat et al. | Pavement Surface Distress Detection Using Digital Image Processing Techniques | |
| Chen | PTZ visibility detection based on image luminance changing tendency | |
| Ghanta et al. | A Hessian-based methodology for automatic surface crack detection and classification from pavement images | |
| Bosurgi et al. | Automatic crack detection results using a novel device for survey and analysis of road pavement condition | |
| Yang et al. | Effect of 3D laser point spacing on cement concrete crack width measurement | |
| Al-Mistarehi | An approach for automated detection and classification of pavement cracks | |
| Li et al. | Histogram-Based Road Markings Extraction by Using 3D Laser Imaging |
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 | ||
| CB02 | Change of applicant information |
Address after: 430223 No.6, 4th Road, Wuda Science Park, Donghu high tech Zone, Wuhan City, Hubei Province Applicant after: Wuhan Optical Valley excellence Technology Co.,Ltd. Address before: 430223 No.6, 4th Road, Wuda Science Park, Donghu high tech Zone, Wuhan City, Hubei Province Applicant before: Wuhan Wuda excellence Technology Co.,Ltd. Address after: 430223 No.6, 4th Road, Wuda Science Park, Donghu high tech Zone, Wuhan City, Hubei Province Applicant after: Wuhan Wuda excellence Technology Co.,Ltd. Address before: 430223 No.6, 4th Road, Wuda Science Park, Donghu high tech Zone, Wuhan City, Hubei Province Applicant before: WUHAN WUDA ZOYON SCIENCE AND TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |