CN116205865B - Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data - Google Patents

Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data Download PDF

Info

Publication number
CN116205865B
CN116205865B CN202310110004.4A CN202310110004A CN116205865B CN 116205865 B CN116205865 B CN 116205865B CN 202310110004 A CN202310110004 A CN 202310110004A CN 116205865 B CN116205865 B CN 116205865B
Authority
CN
China
Prior art keywords
dimensional
data
depth
uniformity
ssi
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
Application number
CN202310110004.4A
Other languages
Chinese (zh)
Other versions
CN116205865A (en
Inventor
王慧
张勋
池敏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University
Original Assignee
Chongqing University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing University filed Critical Chongqing University
Priority to CN202310110004.4A priority Critical patent/CN116205865B/en
Publication of CN116205865A publication Critical patent/CN116205865A/en
Application granted granted Critical
Publication of CN116205865B publication Critical patent/CN116205865B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/06—Road conditions
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0062—Adapting control system settings
    • B60W2050/0063—Manual parameter input, manual setting means, manual initialising or calibrating means
    • B60W2050/0064—Manual parameter input, manual setting means, manual initialising or calibrating means using a remote, e.g. cordless, transmitter or receiver unit, e.g. remote keypad or mobile phone
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2300/00—Indexing codes relating to the type of vehicle
    • B60W2300/36—Cycles; Motorcycles; Scooters
    • B60W2300/367—Tricycles
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2420/00—Indexing codes relating to the type of sensors based on the principle of their operation
    • B60W2420/40—Photo, light or radio wave sensitive means, e.g. infrared sensors
    • B60W2420/403—Image sensing, e.g. optical camera
    • 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
    • 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
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • 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
    • G06T2207/30248—Vehicle exterior or interior
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Transportation (AREA)
  • Automation & Control Theory (AREA)
  • Mechanical Engineering (AREA)
  • Geometry (AREA)
  • Mathematical Physics (AREA)
  • Quality & Reliability (AREA)
  • Human Computer Interaction (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data, which comprises the following steps: 1) Acquiring a two-dimensional image and three-dimensional depth data by using a three-dimensional near-ground integrated texture photographing device; 2) The upper computer carries out deformation calibration on the three-dimensional depth data to obtain a calibrated depth set Z 1 The method comprises the steps of carrying out a first treatment on the surface of the 3) Based on the gray value of the two-dimensional image data, image segmentation based on the maximum inter-class gap is carried out on the two-dimensional image, so that a plurality of connected areas are obtained; 4) Calculating kurtosis P corresponding to the connected region A Degree of deviation S A Extremely poor D A Information entropy E A Self-similarity SSI A The method comprises the steps of carrying out a first treatment on the surface of the 5) Calculating kurtosis P corresponding to the connected volume statistical set V V Degree of deviation S V Extremely poor D V Information entropy E V Self-similarity SSI V The method comprises the steps of carrying out a first treatment on the surface of the 6) The average distance index ADI, maximum distance index MDI, and road local uniformity SSI, which characterize the road uniformity, are calculated. The invention can effectively monitor the uniformity of the pavement and provide a powerful criterion for construction quality evaluation.

Description

Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data
Technical Field
The invention relates to the field of road surface uniformity analysis, in particular to a road surface uniformity analysis method based on two-dimensional and three-dimensional fusion data.
Background
The uniformity problem of the road surface obviously affects the occurrence of early diseases of the road surface and the long service life, and is mainly affected by the segregation degree in the mixing, transporting and paving processes of the mixture and the compaction uniformity degree of the road surface.
Therefore, the road surface uniformity is an important index for checking the construction quality and is also an important method for checking the early state of the road; in particular, for the case of a specific mixture composition such as regeneration, uniformity inspection is necessary.
However, the conventional construction acceptance often only detects macroscopic indexes such as flatness, and uniformity of the mixture is not considered.
Therefore, a road surface uniformity analysis method that fuses a two-dimensional image and road surface texture depth distribution information is required.
Disclosure of Invention
The invention aims to provide a pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data, which comprises the following steps of:
1) Splitting a pavement to be analyzed into a plurality of large grids, and splitting each large grid into a plurality of sub-grids; the number of specification types of these sub-grids is noted as n; each sub-grid is used as a scanning detection analysis area; n is a positive integer;
2) Acquiring a two-dimensional image and three-dimensional depth data by using a three-dimensional near-ground integrated texture photographing device;
the three-dimensional near-ground integrated texture photographing device comprises a three-wheeled trolley, a vehicle-mounted brain module, a three-dimensional photographing sensor, an encoder and a sensor clamping fixing device;
the sensor clamping position fixing device is a beam type portal frame and is provided with n clamping positions with different heights, and the n clamping positions correspond to scanning visual fields of n sub-grids with different specifications respectively;
the three-dimensional imaging sensors are fixed on the clamping positions of the sensor clamping position fixing device, respectively scan the corresponding scanning detection analysis areas, and transmit pavement image data to the vehicle-mounted brain module;
the road surface image data comprises a two-dimensional image and three-dimensional depth data;
the encoder acquires the rotation frequency of the wheel axle of the small tricycle and transmits the rotation frequency to the vehicle-mounted brain module;
the vehicle-mounted brain module drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the vehicle-mounted brain module controls the three-dimensional camera sensor and the encoder to work;
the vehicle-mounted brain module receives pavement image data and the rotation frequency of the wheel axle of the small tricycle and uploads the pavement image data and the rotation frequency to the upper computer;
3) The upper computer carries out deformation calibration on the three-dimensional depth data to obtain a calibrated depth set Z 1 ；
Calibrated three-dimensional depth information set Z 1 The following is shown:
Z＝aX+bXY+cY+dX 2 +e (5)
Z 1 ＝Z 0 -Z (6)
wherein a, b, c, d is a depth information coefficient; e is a constant; z is Z 0 A three-dimensional depth information set before calibration; z is a fitting optimization plane; x, Y is the two-dimensional coordinate of the current measuring point;
4) Based on the gray value of the two-dimensional image data, image segmentation based on the maximum inter-class gap is carried out on the two-dimensional image, so that a plurality of connected areas are obtained;
5) Calculating the number A of pixels of the connected region, and calculating the kurtosis P corresponding to the connected region based on the grading statistical data of the connected region A Degree of deviation S A Extremely poor D A Information entropy E A Self-similarity SSI A ；
6) Solving a depth threshold of the maximum inter-class gap based on the depth information;
7) Based on the depth threshold, counting the connected volumes of the protrusions and the depressions, and establishing a connected volume counting set V;
calculating kurtosis P corresponding to the connected volume statistical set V V Degree of deviation S V Extremely poor D V Information entropy E V Self-similarity SSI V ；
8) Establishing a feature set of N measuring points of a pavement to be analyzed; one measuring point corresponds to one scanning detection analysis area;
the feature set of the ith measuring point is as follows:
Fi＝{P Ai， S Ai， D Ai， E Ai， SSI Ai， P Vi， S Vi， D Vi ，E Vi ，SSI Vi }； (1)
where i=1, 2, N;
9) Calculating Euclidean distance d between every two measuring points ij ；
10 Calculating an average distance index ADI, a maximum distance index MDI and a road local uniformity condition SSI for characterizing road surface uniformity, namely:
MDI＝max i，j≤N d ij -min i，j≤N d ij (3)
wherein N is the total number of measuring points; SSI (secure Shell) i Is a local uniformity index.
Further, the three-dimensional near-ground integrated texture photographing device further comprises a trolley remote control module;
the trolley remote control module generates trolley remote control data and transmits the trolley remote control data to the vehicle-mounted brain module, so that remote control of the three-wheeled trolley is realized.
Further, the vehicle-mounted brain module is mounted on the three-wheeled trolley and comprises a trolley control system, a shooting control module, a data local storage module and a communication module;
the trolley control system drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the shooting control module is used for controlling the acquisition mode of the three-dimensional shooting sensor; the acquisition modes comprise a uniform speed acquisition mode and a speed coordination control mode based on an encoder;
the data local storage module is used for receiving and storing pavement image data acquired by the three-dimensional camera sensor;
the communication module is used for receiving the data of the trolley remote control module and transmitting the data to the trolley control system.
Further, the three-wheeled cart further comprises a flexible trailer hitch;
the flexible trailer hook is used for connecting the three-wheeled trolley with the trailer so that the trailer moves with the three-wheeled trolley;
when the trailer moves with the three-wheeled trolley, the front wheel of the three-wheeled trolley moves upwards to be retracted.
Further, the three-dimensional image pickup sensor comprises an image pickup sensor and a line laser sub-sensor;
the line laser sub-sensor is used for monitoring three-dimensional depth information of the ground;
the camera sensor is used for monitoring two-dimensional images of the ground.
Further, kurtosis P (J), skewness S (J), range D (J), information entropy E (J), and self-similarity SSI (J) are respectively as follows:
D(J)＝max J-min J (9)
E(J)＝-∑ J∈J P(j)log 2 j (10)
SSI i ＝SSI(I i ) i SSI(H i ) i (12)
wherein J represents a data matrix composed of the number A of pixels of the connected region, a data matrix composed of the connected volume statistics set V, and two-dimensional image brightness data I i Composed data matrix, three-dimensional depth data H i A data matrix is formed; m and n are numbers of corresponding subsets equally divided based on the existing data matrix; b is the number of aliquots; alpha, beta and gamma are constants; mu represents the mean value and sigma represents the standard deviation; SSI (I) i ) i Representing two-dimensional image luminance data I i Corresponding local uniformity conditions of the road; SSI (H) i Representing three-dimensional depth data H i Corresponding local uniformity conditions of the road;
wherein, the parameters l (m, n), c (m, n), t (m, n) are respectively as follows:
c 1 ＝(K 1 *L) 2 (17)
c 2 ＝(K 2 *L) 2 (18)
wherein, c 1 、c 2 、c 3 Is an intermediate parameter; k (K) 1 、K 2 Is a constant; l is the length of the matrix value interval. Mu (mu) m 、μ n Is the mean value; sigma (sigma) mn Is the standard deviation.
Further, euclidean distance d ij The following is shown:
wherein d ij Is the Euclidean distance between two measuring points, f ik ，f jk The value of the kth index of the feature set for points i and j.
Further, the average distance index ADI represents the overall uniformity level of the road section, the maximum distance index MDI represents the extreme non-uniformity level, and the road local uniformity condition SSI represents the local non-uniformity level;
the greater the number of ADI and MDI, the poorer the uniformity; the numerical interval of the SSI is 0-1, and the larger the numerical value is, the better the uniformity is.
Further, the step of performing image segmentation on the two-dimensional image based on the maximum inter-class gap includes:
4.1 Taking all pixel gray data in the measurement range as a statistical object, and solving the pixel gray distribution probability P corresponding to different gray levels i r (K rt )；
4.2 Calculating the inter-class variance of the pixel gray distribution corresponding to any two different gray levels;
wherein, the inter-class variance between the pixel gray level cluster corresponding to the gray level r and the pixel gray level cluster corresponding to the gray level tThe following is shown:
wherein m is G Average gray value of measurement range; m is m r (K rt )、m t (K rt ) The average gray value of the pixel gray clusters corresponding to the gray level r and the gray level t is obtained; r is not equal to t; r, t=1, 2, …, L; l is the number of gray levels;
4.3 Solving for inter-class variancesReaching a maximum gray threshold K rtmax ；
4.4 Dividing the two-dimensional image according to all gray threshold values.
Further, the step of solving the depth threshold of the maximum inter-class gap based on the depth information includes:
6.1 All depth data in the measuring range are used as statistical objects, and distribution probability P corresponding to different depth levels g is solved g (h fg )；
6.2 Calculating the inter-class variance of depth distribution corresponding to any two different depths;
wherein, depth clustering and depth level corresponding to depth level gf corresponding inter-class variance of depth clusteringThe following is shown:
wherein m is h Average depth for the measurement range; m is m f (K fg )、m g (K fg ) The average depth value of the depth clusters corresponding to the depth level f and the depth level g; f is not equal to g; f, g=1, 2, …, H; h is the number of depth levels;
6.3 Solving for inter-class variancesReaching a maximum depth threshold h fgmax 。
The method has the technical effects that the method is undoubtedly based on two-dimensional and three-dimensional road texture acquisition technology, the two-dimensional image and the road texture depth distribution information are fused and analyzed, and a nondestructive automatic detection method for road uniformity, porosity and the like is constructed.
Drawings
FIG. 1 is a schematic view of image monitoring;
FIG. 2 is a schematic diagram of a three-wheeled dolly;
fig. 3 is a schematic (gray scale) of two-dimensional luminance data Ii;
fig. 4 is a schematic representation of three-dimensional depth information data Hi;
FIG. 5 is a graph showing the image segmentation result based on the maximum inter-class gap;
FIG. 6 is a depth threshold for solving a maximum inter-class gap based on depth information.
Detailed Description
The present invention is further described below with reference to examples, but it should not be construed that the scope of the above subject matter of the present invention is limited to the following examples. Various substitutions and alterations are made according to the ordinary skill and familiar means of the art without departing from the technical spirit of the invention, and all such substitutions and alterations are intended to be included in the scope of the invention.
Example 1:
referring to fig. 1 to 6, a pavement uniformity analysis method based on two-dimensional fusion data is characterized by comprising the following steps:
1) Splitting a pavement to be analyzed into a plurality of large grids, and splitting each large grid into a plurality of sub-grids; the number of specification types of these sub-grids is noted as n; each sub-grid is used as a scanning detection analysis area;
2) Acquiring a two-dimensional image and three-dimensional depth data by using a three-dimensional near-ground integrated texture photographing device;
the three-dimensional near-ground integrated texture photographing device comprises a three-wheeled trolley, a vehicle-mounted brain module, a three-dimensional photographing sensor, an encoder and a sensor clamping fixing device;
the sensor clamping position fixing device is a beam type portal frame and is provided with n clamping positions with different heights, and the n clamping positions correspond to scanning visual fields of n sub-grids with different specifications respectively;
the three-dimensional imaging sensors are fixed on the clamping positions of the sensor clamping position fixing device, respectively scan the corresponding scanning detection analysis areas, and transmit pavement image data to the vehicle-mounted brain module;
the road surface image data comprises a two-dimensional image and three-dimensional depth data;
the encoder acquires the rotation frequency of the wheel axle of the small tricycle and transmits the rotation frequency to the vehicle-mounted brain module;
the vehicle-mounted brain module drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the vehicle-mounted brain module controls the three-dimensional camera sensor and the encoder to work;
the vehicle-mounted brain module receives pavement image data and the rotation frequency of the wheel axle of the small tricycle and uploads the pavement image data and the rotation frequency to the upper computer;
3) The upper computer carries out deformation calibration on the three-dimensional depth data to obtain a calibrated depth set Z 1 ；
4) Based on the gray value of the two-dimensional image data, image segmentation based on the maximum class gap is carried out on the two-dimensional image to obtain a plurality of connected areas, and the specific steps are as follows:
first, taking all pixel gray data in the measurement range as a statistical object, L represents a corresponding gray level, and typically between 0 and 255, the sum of the pixel numbers of the respective gray levels is the total pixel number of the image.
Secondly, solving the distribution probability P corresponding to different gray levels i i Setting K as a segmentation threshold to be solved, wherein the average gray values of the pixel numbers corresponding to two groups of gray levels with gray levels higher than K and lower than K are respectively m 1 (K) And m 2 (K) The probabilities are P respectively 1 (K) And P 2 (K) The average gray value of the whole measurement range (image range) is m G 。
Inter-class variance of Solving for->The maximum K value K is used as a gray threshold for image segmentation binarization.
Based on the gray threshold, image segmentation is performed on the two-dimensional image.
5) Calculating the number A of pixels of the connected region, and calculating the kurtosis P corresponding to the connected region based on the grading statistical data of the connected region A Degree of deviation S A Extremely poor D A Information entropy E A Self-similarity SSI A ；
6) Solving a depth threshold of the maximum inter-class gap based on the depth information; the method comprises the following specific steps:
first, all depth data in a measurement range are taken as statistical objects, H represents a corresponding depth value distribution interval, and the sum of pixel numbers of each depth level is the total pixel number of an image.
Secondly, solving different depths hCorresponding distribution probability P h Setting h as a depth plane segmentation threshold to be solved, wherein the flat depths of the pixel numbers corresponding to two groups of gray scales higher than h and lower than h are respectively m 1 (h) And m 2 (h) The probabilities are P respectively 1 (h) And P 2 (h) The average depth value of the whole measurement range (image range) is m h 。
Inter-class variance of Solving for->The largest h value h is taken as the depth threshold of the image segmentation binarization.
7) Based on the depth threshold, counting the connected volumes of the protrusions and the depressions, and establishing a connected volume counting set V;
calculating kurtosis P corresponding to the connected volume statistical set V V Degree of deviation S V Extremely poor D V Information entropy E V Self-similarity SSI V ；
8) Establishing a feature set of N measuring points of a pavement to be analyzed; one measuring point corresponds to one scanning detection analysis area.
The feature set of the ith point is as follows:
Fi＝{P Ai， S Ai， D Ai， E Ai， SSI Ai， P Vi， S Vi， D Vi ，E Vi ，SSI Vi }； (1)
where i=1, 2, N;
9) Calculating Euclidean distance d between every two measuring points ij ；
10 Calculating an average distance index ADI, a maximum distance index MDI and a road local uniformity condition SSI for characterizing road surface uniformity, namely:
MDI＝max i，j≤N d ij -min i，j≤N d ij (3)
wherein N is the total number of measuring points; SSI (secure Shell) i Is a local uniformity index.
The three-dimensional near-ground integrated texture photographing device further comprises a trolley remote control module;
the trolley remote control module generates trolley remote control data and transmits the trolley remote control data to the vehicle-mounted brain module, so that remote control of the three-wheeled trolley is realized.
The vehicle-mounted brain module is mounted on the three-wheeled trolley and comprises a trolley control system, a photographing control module, a data local storage module and a communication module;
the trolley control system drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the shooting control module is used for controlling the acquisition mode of the three-dimensional shooting sensor; the acquisition mode comprises a uniform speed acquisition mode and an encoder-based speed coordination control mode. Setting an acquisition rate in a uniform acquisition mode; and coordinating an acquisition mode and setting a scanning sampling interval.
The data local storage module is used for receiving and storing pavement image data acquired by the three-dimensional camera sensor;
the communication module is used for receiving the data of the trolley remote control module and transmitting the data to the trolley control system.
The three-wheeled trolley further comprises a flexible trailer hook;
the flexible trailer hook is used for connecting the three-wheeled trolley with the trailer so that the trailer moves with the three-wheeled trolley;
when the trailer moves with the three-wheeled trolley, the front wheel of the three-wheeled trolley moves upwards to be retracted.
The three-dimensional camera sensor comprises a camera sensor and a line laser sub-sensor;
the line laser sub-sensor is used for monitoring three-dimensional depth information of the ground;
the camera sensor is used for monitoring two-dimensional images of the ground.
In step 3), the calibrated three-dimensional depth information set Z 1 The following is shown:
Z＝aX+bXY+cY+dX 2 +e (5)
Z 1 ＝Z 0 -Z (6)
wherein a, b, c, d is a coefficient; e is a constant; z is Z 0 A three-dimensional depth information set before calibration; z is a fitting optimization plane; x, Y is the current measurement point two-dimensional coordinate.
Kurtosis P (J), skewness S (J), range D (J), information entropy E (J), and self-similarity SSI (J) are shown below, respectively:
D(J)＝max J-min J (9)
E(J)＝-∑ j∈J P(j)log 2 j (10)
SSI i ＝SSI(I i ) i SSI(H i ) i (12)
wherein J represents a data matrix composed of the number A of pixels of the connected region, a data matrix composed of the connected volume statistics set V, and two-dimensional image brightness data I i Composed data matrix, three-dimensional depth data H i A data matrix is formed; m and n are numbers of corresponding subsets equally divided based on the existing data matrix; b is an equal fractionAn amount of; alpha, beta and gamma are constants; μ represents the mean and σ represents the standard deviation.
Wherein, the parameters l (m, n), c (m, n), t (m, n) are respectively as follows:
c 1 ＝(K 1 *L) 2 (17)
c 2 ＝(K 2 *L) 2 (18)
wherein, c 1 、c 2 、c 3 Is an intermediate parameter.
When J represents a data matrix composed of the number a of pixels of the connected region, α=0, β=1, γ=1;
when J represents the data matrix composed of the connected volume statistics set V, α=1, β=1, γ=1, respectively.
Euclidean distance d ij The following is shown:
wherein d ij Is the Euclidean distance between two measuring points, f ik ，f jk The value of the kth index of the feature set for points i and j.
The average distance index ADI represents the overall uniformity level of the road section, the maximum distance index MDI represents the extreme non-uniformity level, and the road local uniformity condition SSI represents the local non-uniformity level;
the greater the number of ADI and MDI, the poorer the uniformity; the numerical interval of the SSI is 0-1, and the larger the numerical value is, the better the uniformity is.
Example 2:
referring to fig. 1 to 6, a road surface uniformity analysis method based on two-dimensional fusion data includes the following steps:
1) The collected pavement data is split into grids of 1m multiplied by 1m, each cross section at least comprises 3 grids, the vertical section samples at least 10 grids based on fixed intervals, usually 30 grids, and the grid number is set to be N.
2) Each 1 square meter grid was split into nine equal parts of 3X 3, and each sub-grid was a scanning test object to be analyzed, and was set to 30cm X30 cm,20cm X20 cm, and 10cm X10 cm.
3) And acquiring data by adopting a three-dimensional near-ground integrated texture photogrammetry remote control trolley.
The composition and the data acquisition range are shown in the figure, and the three-dimensional imaging device comprises 5 parts of a three-wheeled trolley, a vehicle-mounted brain module, a three-dimensional imaging sensor, an encoder and a sensor clamping fixing device.
The three-wheeled trolley comprises a transmission device, remote control equipment, a motor and other equipment, and a flexible trailer hook arranged at the front end for pulling; the internal light source shooting channel is used for projecting laser and preventing the acquisition of shielding data; the front wheels have the function of upward movement and contraction, and are retracted in the situation of a trailer;
the vehicle-mounted brain module comprises a control system for controlling the trolley to run according to the track, a photographic control module for obtaining the data of the wheel axle encoder and controlling the acquisition mode of the camera sensor, a data local storage module and a communication module for remote control (remote controller);
the three-dimensional camera sensor adopts a laser shooting triangulation principle and consists of a centrally controlled camera and line laser sub-sensor;
the encoder is used for obtaining the rotation frequency of the wheel shaft;
the sensor clamping fixing device is mainly used for fixing a three-dimensional camera sensor, is a beam type portal frame, and is provided with three fixed heights which respectively correspond to three scanning fields of view of 30cm multiplied by 30cm,20cm multiplied by 20cm and 10cm multiplied by 10cm.
4) Two-dimensional brightness data and three-dimensional depth information data are stored simultaneously.
5) Performing deformation calibration on depth information
The driving direction of the collected single-measuring point image is L, the scanning direction is T, and the depth information direction is Z.
First, a surface fit is performed after removing the Z of the extreme projections and depressions of the area window 10×10.
Fitting an optimized plane Z, wherein the original depth set is Z 0 The depth set after calibration is Z 1 。
Z＝aX+bXY+cY+dX 2 +e
Z 1 ＝Z 0 -Z
6) Image segmentation based on the maximum inter-class gap is performed based on the gray values of the gray map, and data is obtained as shown in the following map.
Counting the number A of pixels of the connected region in the measuring range (namely, each test grid), and calculating and obtaining the corresponding kurtosis P based on the grading statistical data of the connected region A Degree of deviation S A Extremely poor D A Information entropy E A Self-similarity SSI A 。
The depth threshold for the maximum inter-class gap is solved based on the depth information, and the obtained data is shown in the following graph.
7) The connected volume statistics set V of the bulge and the recess is carried out based on the threshold surface, the distribution statistics of the volume data are obtained, and the corresponding kurtosis P is obtained through calculation V Degree of deviation S V Extremely poor D V Information entropy E V Self-similarity SSI V 。
8) The calculation formula of each index of the corresponding measuring point is as follows, wherein J represents corresponding A, V and I i ,H i Is a data matrix of the data processor.
D(J)＝maxJ-minJ
Wherein m and n are numbers of corresponding subsets based on equipartition of the existing data matrix, B is equal dividing number, equipartition ensures that the actual space x and y corresponding to the size of each sub-matrix is not less than 50mm, and the number of B is not less than 4.
c 1 ＝(K 1 *L) 2
c 2 ＝(K 2 *L) 2
Generally k1=0.01, k2=0.03, and l=matrix numerical interval range, which is a dynamic range of pixel values or depth values, is 255 for a gray scale image.
When the image is a gray image, alpha=0, beta=1, and gamma=1;
when the depth matrix is, α=1, β=1, γ=1;
local uniformity index SSI i ＝SSI(I i ) i SSI(H i ) i
9) The characteristic set of N measuring points belonging to the same road section is obtained, and the characteristic set of the ith point is represented by the following formula:
Fi＝{P Ai ，S Ai ，D Ai ，E Ai ，SSI Ai ，P Vi ，S Vi ，D Vi ，E Vi ，SSI Vi }，
where i=1 to N.
10 Calculating the Euclidean distance d between every two measuring points based on a set of 10 uniform statistical characteristic indexes ij 。
Wherein d ij Is the Euclidean distance between two measuring points, f ik ，f jk The value of the kth index of the feature set for points i and j.
Test road data set d= { D ij }，i＝1～N，j＝1～N。
11 The average distance index ADI, the maximum distance index MDI and the road local uniformity SSI are calculated.
Where N is the total number of measurement points, typically 30, allowing for an increase or decrease of not less than 10;
ADI represents the overall uniformity level of the road section, MDI represents the extreme non-uniformity level, SSI represents the local non-uniformity level, and the three are comprehensively used for evaluating the road surface uniformity, and the larger the values of ADI and MDI are, the worse the uniformity is; the numerical interval of the SSI is 0-1, and the larger the numerical value is, the better the uniformity is.
Example 3:
a pavement uniformity analysis method based on two-dimensional fusion data comprises the following steps:
1) Splitting a pavement to be analyzed into a plurality of large grids, and splitting each large grid into a plurality of sub-grids; the number of specification types of these sub-grids is noted as n; each sub-grid is used as a scanning detection analysis area; n is a positive integer;
2) Acquiring a two-dimensional image and three-dimensional depth data by using a three-dimensional near-ground integrated texture photographing device;
the three-dimensional near-ground integrated texture photographing device comprises a three-wheeled trolley, a vehicle-mounted brain module, a three-dimensional photographing sensor, an encoder and a sensor clamping fixing device;
the sensor clamping position fixing device is a beam type portal frame and is provided with n clamping positions with different heights, and the n clamping positions correspond to scanning visual fields of n sub-grids with different specifications respectively;
the three-dimensional imaging sensors are fixed on the clamping positions of the sensor clamping position fixing device, respectively scan the corresponding scanning detection analysis areas, and transmit pavement image data to the vehicle-mounted brain module;
the road surface image data comprises a two-dimensional image and three-dimensional depth data;
the encoder acquires the rotation frequency of the wheel axle of the small tricycle and transmits the rotation frequency to the vehicle-mounted brain module;
the vehicle-mounted brain module drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the vehicle-mounted brain module controls the three-dimensional camera sensor and the encoder to work;
the vehicle-mounted brain module receives pavement image data and the rotation frequency of the wheel axle of the small tricycle and uploads the pavement image data and the rotation frequency to the upper computer;
3) The upper computer carries out deformation calibration on the three-dimensional depth data to obtain a calibrated depth set Z 1 ；
Calibrated three-dimensional depth information set Z 1 The following is shown:
Z＝aX+bXY+cY+dX 2 +e (5)
Z 1 ＝Z 0 -Z (6)
wherein a, b, c, d is a depth information coefficient; e is a constant; z is Z 0 A three-dimensional depth information set before calibration; z is a fitting optimization plane; x, Y is the two-dimensional coordinate of the current measuring point;
4) Based on the gray value of the two-dimensional image data, image segmentation based on the maximum inter-class gap is carried out on the two-dimensional image, so that a plurality of connected areas are obtained;
5) Calculating the number A of pixels of the connected region, and calculating the kurtosis P corresponding to the connected region based on the grading statistical data of the connected region A Degree of deviation S A Extremely poor D A Information entropy E A Self-similarity SSI A ；
6) Solving a depth threshold of the maximum inter-class gap based on the depth information;
7) Based on the depth threshold, counting the connected volumes of the protrusions and the depressions, and establishing a connected volume counting set V;
calculating kurtosis P corresponding to the connected volume statistical set V V Degree of deviation S V Extremely poor D V Information entropy E V Self-similarity SSI V ；
8) Establishing a feature set of N measuring points of a pavement to be analyzed; one measuring point corresponds to one scanning detection analysis area;
the feature set of the ith measuring point is as follows:
Fi＝{P Ai ，S Ai ，D Ai ，E Ai ，SSI Ai ，P Vi ，S Vi ，D Vi ，E Vi ，SSI Vi }； (1
where i=1, 2, N;
9) Calculating Euclidean distance between every two measuring pointsd ij ；
10 Calculating an average distance index ADI, a maximum distance index MDI and a road local uniformity condition SSI for characterizing road surface uniformity, namely:
MDI＝max i，j≤N d ij -min i，j≤N d ij (3)
wherein N is the total number of measuring points; SSI (secure Shell) i Is a local uniformity index.
Example 4:
the main content of the pavement uniformity analysis method based on the two-dimensional fusion data is shown in the embodiment 3, wherein the three-dimensional near-ground integrated texture photographing device further comprises a trolley remote control module;
the trolley remote control module generates trolley remote control data and transmits the trolley remote control data to the vehicle-mounted brain module, so that remote control of the three-wheeled trolley is realized.
Example 5:
the main content of the pavement uniformity analysis method based on the two-dimensional fusion data is shown in the embodiment 3, wherein the vehicle-mounted brain module is mounted on a three-wheeled trolley and comprises a trolley control system, a shooting control module, a data local storage module and a communication module;
the trolley control system drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the shooting control module is used for controlling the acquisition mode of the three-dimensional shooting sensor; the acquisition modes comprise a uniform speed acquisition mode and a speed coordination control mode based on an encoder;
the data local storage module is used for receiving and storing pavement image data acquired by the three-dimensional camera sensor;
the communication module is used for receiving the data of the trolley remote control module and transmitting the data to the trolley control system.
Example 6:
the main content of the pavement uniformity analysis method based on the two-dimensional fusion data is shown in the embodiment 3, wherein the three-wheeled trolley further comprises a flexible trailer hook;
the flexible trailer hook is used for connecting the three-wheeled trolley with the trailer so that the trailer moves with the three-wheeled trolley;
when the trailer moves with the three-wheeled trolley, the front wheel of the three-wheeled trolley moves upwards to be retracted.
Example 7:
the main content of the pavement uniformity analysis method based on the two-dimensional fusion data is shown in the embodiment 3, wherein the three-dimensional camera sensor comprises a camera sensor and a line laser sub-sensor;
the line laser sub-sensor is used for monitoring three-dimensional depth information of the ground;
the camera sensor is used for monitoring two-dimensional images of the ground.
Example 8:
the main content of the pavement uniformity analysis method based on the two-dimensional fusion data is shown in the embodiment 3, wherein kurtosis P (J), skewness S (J), extremely poor D (J), information entropy E (J) and self-similarity SSI (J) are respectively shown as follows:
D(J)＝max J-min J (9)
E(J)＝-∑ j∈J P(j)log 2 j (10)
SSI i ＝SSI(I i ) i SSI(H i ) i (12)
wherein J represents a data matrix composed of the number A of pixels of the connected region, a data matrix composed of the connected volume statistics set V, and two-dimensional image brightness data I i Composed data matrix, three-dimensional depth data H i A data matrix is formed; m and n are numbers of corresponding subsets equally divided based on the existing data matrix; b is the number of aliquots; alpha, beta and gamma are constants; mu represents the mean value and sigma represents the standard deviation; SSI (I) i ) i Representing two-dimensional image luminance data I i Corresponding local uniformity conditions of the road; SSI (H) i Representing three-dimensional depth data H i Corresponding local uniformity conditions of the road;
wherein, the parameters l (m, n), c (m, n), t (m, n) are respectively as follows:
c 1 ＝(K 1 *L) 2 (17)
c 2 ＝(K 2 *L) 2 (18)
wherein, c 1 、c 2 、c 3 Is an intermediate parameter; k (K) 1 、K 2 Is a constant; l is the length of the matrix value interval. Mu (mu) m 、μ n Is the mean value; sigma (sigma) mn Is the standard deviation.
Example 9:
a road surface uniformity analysis method based on two-dimensional and three-dimensional fusion data is disclosed in example 3, wherein the Euclidean distance d ij The following is shown:
wherein d ij Is the Euclidean distance between two measuring points, f ik ，f jk The value of the kth index of the feature set for points i and j.
Example 10:
the main content of the road surface uniformity analysis method based on two-dimensional fusion data is shown in the embodiment 3, wherein an average distance index ADI represents the overall uniformity level of the road section, a maximum distance index MDI represents an extremely uneven level, and a road local uniformity condition SSI represents a local uneven level;
the greater the number of ADI and MDI, the poorer the uniformity; the numerical interval of the SSI is 0-1, and the larger the numerical value is, the better the uniformity is.
Example 11:
the main content of the pavement uniformity analysis method based on the two-dimensional fusion data is shown in the embodiment 3, wherein the step of performing image segmentation on the two-dimensional image based on the maximum class gap comprises the following steps:
4.1 Taking all pixel gray data in the measurement range as a statistical object, and solving the pixel gray distribution probability P corresponding to different gray levels i r (K rt )；
4.2 Calculating the inter-class variance of the pixel gray distribution corresponding to any two different gray levels;
wherein, the inter-class variance between the pixel gray level cluster corresponding to the gray level r and the pixel gray level cluster corresponding to the gray level tThe following is shown:
wherein m is G Average gray value of measurement range; m is m r (K rt )、m t (K rt ) The average gray value of the pixel gray clusters corresponding to the gray level r and the gray level t is obtained; r is not equal to t; r, t=1, 2, …, L; l is the number of gray levels;
4.3 Solving for inter-class variancesReaching a maximum gray threshold K rtmax ；
4.4 Dividing the two-dimensional image according to all gray threshold values.
Example 12:
the main content of the pavement uniformity analysis method based on the two-dimensional fusion data is shown in the embodiment 3, wherein the step of solving the depth threshold value of the maximum inter-class gap based on the depth information comprises the following steps:
6.1 All depth data in the measuring range are used as statistical objects, and distribution probability P corresponding to different depth levels g is solved g (h fg )；
6.2 Calculating the inter-class variance of depth distribution corresponding to any two different depths;
wherein, the depth cluster corresponding to the depth level g and the depth cluster corresponding to the depth level f have inter-class variancesThe following is shown:
wherein m is h Average depth for the measurement range; m is m f (K fg )、m g (K fg ) The average depth value of the depth clusters corresponding to the depth level f and the depth level g; f is not equal to g; f, g=1, 2, …, H; h is the number of depth levels;
6.3 Solving for inter-class variancesReaching a maximum depth threshold h fgmax 。/>

Claims (10)

1. The pavement uniformity analysis method based on the two-dimensional fusion data is characterized by comprising the following steps of:
1) Splitting a pavement to be analyzed into a plurality of large grids, and splitting each large grid into a plurality of sub-grids; the number of specification types of these sub-grids is noted as n; each sub-grid is used as a scanning detection analysis area; n is a positive integer;
2) Acquiring a two-dimensional image and three-dimensional depth data by using a three-dimensional near-ground integrated texture photographing device;
the three-dimensional near-ground integrated texture photographing device comprises a three-wheeled trolley, a vehicle-mounted brain module, a three-dimensional photographing sensor, an encoder and a sensor clamping fixing device;
the sensor clamping position fixing device is a beam type portal frame and is provided with n clamping positions with different heights, and the n clamping positions correspond to scanning visual fields of n sub-grids with different specifications respectively;
the three-dimensional imaging sensors are fixed on the clamping positions of the sensor clamping position fixing device, respectively scan the corresponding scanning detection analysis areas, and transmit pavement image data to the vehicle-mounted brain module;
the road surface image data comprises a two-dimensional image and three-dimensional depth data;
the encoder acquires the rotation frequency of the wheel axle of the small tricycle and transmits the rotation frequency to the vehicle-mounted brain module;
the vehicle-mounted brain module drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the vehicle-mounted brain module controls the three-dimensional camera sensor and the encoder to work;
the vehicle-mounted brain module receives pavement image data and the rotation frequency of the wheel axle of the small tricycle and uploads the pavement image data and the rotation frequency to the upper computer;
3) Upper computerPerforming deformation calibration on the three-dimensional depth data to obtain a calibrated three-dimensional depth information set Z 1 ；
Calibrated three-dimensional depth information set Z 1 The following is shown:
Z＝aX+bXY+cY+dX 2 +e (5)
Z 1 ＝Z 0 -Z (6)
wherein a, b, c, d is a depth information coefficient; e is a constant; z is Z 0 A three-dimensional depth information set before calibration; z is a fitting optimization plane; x, Y is the two-dimensional coordinate of the current measuring point;
4) Based on the gray value of the two-dimensional image data, image segmentation based on the maximum inter-class gap is carried out on the two-dimensional image, so that a plurality of connected areas are obtained;
5) Calculating the number A of pixels of the connected region, and calculating the kurtosis P corresponding to the connected region based on the grading statistical data of the connected region A Degree of deviation S A Extremely poor D A Information entropy EA and self-similarity SSI A ；
6) Solving a depth threshold of the maximum inter-class gap based on the depth information;
7) Based on the depth threshold, counting the connected volumes of the protrusions and the depressions, and establishing a connected volume counting set V;
calculating kurtosis P corresponding to the connected volume statistical set V V Degree of deviation S V Extremely poor D V Information entropy E V Self-similarity SSI V ；
8) Establishing a feature set of N measuring points of a pavement to be analyzed; one measuring point corresponds to one scanning detection analysis area;
the feature set of the ith measuring point is as follows:
Fi＝{P Ai ，S Ai ，D Ai ，E Ai ，SSI Ai ，P Vi ，S Vi ，D Vi ，E Vi ，SSI Vi }；(1)
wherein i=1, 2, …, N;
9) Calculating Euclidean distance d between every two measuring points ij ；
10 Calculating an average distance index ADI, a maximum distance index MDI and a road local uniformity condition SSI for characterizing road surface uniformity, namely:
MDI＝max i，j≤N d ij -min i，j≤N d ij (3)
wherein N is the total number of measuring points; SSI (secure Shell) i Is a local uniformity index.
2. The pavement uniformity analysis method based on two-dimensional fusion data according to claim 1, wherein the three-dimensional near-ground integrated texture photographing device further comprises a trolley remote control module;
the trolley remote control module generates trolley remote control data and transmits the trolley remote control data to the vehicle-mounted brain module, so that remote control of the three-wheeled trolley is realized.
3. The pavement uniformity analysis method based on the two-dimensional fusion data according to claim 2, wherein the vehicle-mounted brain module is mounted on a three-wheeled trolley and comprises a trolley control system, a photographing control module, a data local storage module and a communication module;
the trolley control system drives the three-wheeled trolley to run on a road surface to be analyzed and controls the running track of the three-wheeled trolley;
the shooting control module is used for controlling the acquisition mode of the three-dimensional shooting sensor; the acquisition modes comprise a uniform speed acquisition mode and a speed coordination control mode based on an encoder;
the data local storage module is used for receiving and storing pavement image data acquired by the three-dimensional camera sensor;
the communication module is used for receiving the data of the trolley remote control module and transmitting the data to the trolley control system.
4. The method for analyzing the road surface uniformity based on the two-dimensional fusion data according to claim 1, wherein the three-wheeled trolley further comprises a flexible trailer hook;
the flexible trailer hook is used for connecting the three-wheeled trolley with the trailer so that the trailer moves with the three-wheeled trolley;
when the trailer moves with the three-wheeled trolley, the front wheel of the three-wheeled trolley moves upwards to be retracted.
5. The pavement uniformity analysis method based on two-dimensional fusion data according to claim 1, wherein the three-dimensional image sensor comprises an image sensor and a line laser sub-sensor;
the line laser sub-sensor is used for monitoring three-dimensional depth information of the ground;
the camera sensor is used for monitoring two-dimensional images of the ground.
6. The method for analyzing the road surface uniformity based on the two-dimensional fusion data according to claim 1, wherein kurtosis P (J), skewness S (J), range D (J), information entropy E (J) and self-similarity SSI (J) are respectively as follows:
D(J)＝max J-min J (9)
E(J)＝-∑ j'∈J P(j’)log 2 j’ (10)
SSI i ＝SSI(I i ) i SSI(H i ) i (12)
wherein J represents a data matrix composed of the number A of pixels of the connected region, a data matrix composed of the connected volume statistics set V, and two-dimensional image brightness data I i Composed data matrix, three-dimensional depth data H i A data matrix is formed; m and n are numbers of corresponding subsets equally divided based on the existing data matrix; b is the number of aliquots; alpha, beta and gamma are constants; mu represents the mean value and sigma represents the standard deviation; SSI (I) i ) i Representing two-dimensional image luminance data I i Corresponding local uniformity conditions of the road; SSI (H) i Representing three-dimensional depth data H i Corresponding local uniformity conditions of the road;
wherein, the parameters l (m, n), c (m, n), t (m, n) are respectively as follows:
c 1 ＝(K 1 *L) 2 (17)
c 2 ＝(K 2 *L) 2 (18)
wherein, c 1 、c 2 、c 3 Is an intermediate parameter; k (K) 1 、K 2 Is a constant; l is the length of the matrix numerical interval; mu (mu) m 、μ n Is the mean value; sigma (sigma) mn Is the standard deviation.
7. The method for analyzing the road surface uniformity based on the two-dimensional fusion data according to claim 1, wherein the Euclidean distance d ij The following is shown:
wherein d ij Is the Euclidean distance between two measuring points, f ik ，f jk The value of the kth index of the feature set for points i and j.
8. The method for analyzing the road surface uniformity based on two-dimensional fusion data according to claim 1, wherein the average distance index ADI represents the overall uniformity level of the road section, the maximum distance index MDI represents the extreme non-uniformity level, and the road local uniformity condition SSI represents the local non-uniformity level;
the greater the number of ADI and MDI, the poorer the uniformity; the numerical interval of the SSI is O-1, and the larger the numerical value is, the better the uniformity is.
9. The method for analyzing the road surface uniformity based on the two-dimensional fusion data according to claim 1, wherein the step of performing the image segmentation based on the maximum inter-class gap on the two-dimensional image comprises the steps of:
4.1 Taking all pixel gray data in the measurement range as a statistical object, and solving the pixel gray distribution probability P corresponding to different gray levels r (K rt )；
4.2 Calculating the inter-class variance of the pixel gray distribution corresponding to any two different gray levels;
wherein, the inter-class variance between the pixel gray level cluster corresponding to the gray level r and the pixel gray level cluster corresponding to the gray level tThe following is shown:
wherein m is G Average gray value of measurement range; m is m r (K rt )、m t (K rt ) The average gray value of the pixel gray clusters corresponding to the gray level r and the gray level t is obtained; r is not equal to t; r, t=1, 2, …, L; l is the number of gray levels;
4.3 Solving for inter-class variancesReaching a maximum gray threshold K rtmax ；
4.4 Dividing the two-dimensional image according to all gray threshold values.
10. The method for analyzing the road surface uniformity based on the two-dimensional fusion data according to claim 1, wherein the step of solving the depth threshold of the maximum inter-class gap based on the depth information comprises the steps of:
6.1 All depth data in the measuring range are used as statistical objects, and distribution probability P corresponding to different depth levels g is solved g (h fg )；
6.2 Calculating the inter-class variance of depth distribution corresponding to any two different depths;
wherein, the depth cluster corresponding to the depth level g and the depth cluster corresponding to the depth level f have inter-class variancesThe following is shown:
wherein m is h Average depth for the measurement range; m is m f (h fg )、m g (h fg ) The average depth value of the depth clusters corresponding to the depth level f and the depth level g; f is not equal to g; f, g=1, 2, …, H; h is the number of depth levels;
6.3 Solving for inter-class variancesReaching a maximum depth threshold h fgmax 。
CN202310110004.4A 2023-02-14 2023-02-14 Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data Active CN116205865B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310110004.4A CN116205865B (en) 2023-02-14 2023-02-14 Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310110004.4A CN116205865B (en) 2023-02-14 2023-02-14 Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data

Publications (2)

Publication Number Publication Date
CN116205865A CN116205865A (en) 2023-06-02
CN116205865B true CN116205865B (en) 2023-09-26

Family

ID=86510738

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310110004.4A Active CN116205865B (en) 2023-02-14 2023-02-14 Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data

Country Status (1)

Country Link
CN (1) CN116205865B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102359056A (en) * 2011-07-20 2012-02-22 东南大学 Detection method of bituminous pavement data
CN102706880A (en) * 2012-06-26 2012-10-03 哈尔滨工业大学 Road information extraction device based on two-dimensional image and depth information and road crack information detection method based on same
CN107288016A (en) * 2017-06-07 2017-10-24 长安大学 A kind of bituminous paving milling quality evaluating method based on three-dimensional laser detection technique
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
CN114812450A (en) * 2022-04-25 2022-07-29 山东省路桥集团有限公司 Machine vision-based asphalt pavement construction uniformity detection and evaluation method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10789487B2 (en) * 2018-04-05 2020-09-29 Here Global B.V. Method, apparatus, and system for determining polyline homogeneity

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102359056A (en) * 2011-07-20 2012-02-22 东南大学 Detection method of bituminous pavement data
CN102706880A (en) * 2012-06-26 2012-10-03 哈尔滨工业大学 Road information extraction device based on two-dimensional image and depth information and road crack information detection method based on same
CN107288016A (en) * 2017-06-07 2017-10-24 长安大学 A kind of bituminous paving milling quality evaluating method based on three-dimensional laser detection technique
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
CN114812450A (en) * 2022-04-25 2022-07-29 山东省路桥集团有限公司 Machine vision-based asphalt pavement construction uniformity detection and evaluation method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于PQI评价沥青路面施工质量的有效检测频率;熊学堂;张肖宁;虞将苗;苏尚武;;中外公路(03);全文 *

Also Published As

Publication number Publication date
CN116205865A (en) 2023-06-02

Similar Documents

Publication Publication Date Title
JP6454109B2 (en) Road surface state management device and road surface state management program
CN110838128B (en) Image method aggregate stacking void ratio prediction method and system
She et al. Feasibility study of asphalt pavement pothole properties measurement using 3D line laser technology
CN105528891A (en) Traffic flow density detection method and system based on unmanned aerial vehicle monitoring
CN109297428A (en) A kind of high-precision deformation based on unmanned plane patrols survey technology method
CN113465541B (en) Contact line abrasion measuring method and system
CN111783638A (en) System and method for detecting number of vehicle axles and identifying vehicle type
CN111896549A (en) Building crack monitoring system and method based on machine learning
CN115808324B (en) Light safety management monitoring method and system for small and medium span bridges
CN116448773A (en) Pavement disease detection method and system with image-vibration characteristics fused
CN112184707A (en) Method and system for judging earth and stone load of muck vehicle based on point cloud data
CN116399302B (en) Method for monitoring dynamic compaction settlement in real time based on binocular vision and neural network model
CN110926426B (en) Real-time detection device and detection method for residual muck volume of subway shield segment
CN116205865B (en) Pavement uniformity analysis method based on two-dimensional and three-dimensional fusion data
CN113884492A (en) Ringelmann blackness calibration and detection method and device for motor vehicle exhaust
CN111275753A (en) Method for measuring volume of sand and stone in carriage
CN114494845A (en) Artificial intelligence hidden danger troubleshooting system and method for construction project site
CN117169121B (en) Traffic road disease detection system and method based on cloud edge end architecture
CN114910046B (en) Pavement three-dimensional detection system and method based on bidirectional line structured light
CN114104894B (en) Multi-parameter detection method for quality of elevator guide rail
CN113569491B (en) Wheel set size detection data analysis and correction method and device
CN115821880A (en) Tamping settlement measuring method based on TOF camera
CN114919079A (en) Method and system for evaluating working state of double-wheel milling machine based on image recognition
CN117218089B (en) Asphalt pavement structure depth detection method
CN118656773B (en) Roadbed compaction quality intelligent monitoring method, device and system

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