阅读说明：本技术 路面三维纹理形貌的检测方法、检测系统及存储介质 (Detection method and detection system for three-dimensional texture morphology of pavement and storage medium ) 是由 王元元 郭诗言 薛金顺 成羽 周飞 程宸 于 2019-09-30 设计创作，主要内容包括：本发明公开了一种路面三维纹理形貌的检测方法、检测系统及存储介质,路面三维纹理形貌的检测方法包括：获取同一待测路面的第一图像和第二图像,标记线将第一图像和第二图像分割成多个子区域；识别标记线；根据标记线和半全局匹配算法,对第一图像和第二图像中的每一子区域的边界和像素点进行匹配,获得多个匹配子区域及其匹配视差；根据匹配视差获得待测路面的三维重构结果。本发明中,通过标记线的约束作用,获得第一图像和第二图像中高度重合的对应子区域,继而分割并提取第一图像和第二图像中的每一匹配子区域,计算每一匹配子区域的匹配视差,获得匹配子区域的空间坐标,便利于重构待测路面整体的三维纹理形貌,提高检测结果的准确度。(The invention discloses a detection method, a detection system and a storage medium for three-dimensional texture morphology of a road surface, wherein the detection method for the three-dimensional texture morphology of the road surface comprises the following steps: acquiring a first image and a second image of the same road surface to be detected, and dividing the first image and the second image into a plurality of sub-areas by using marking lines; identifying the marking line; matching the boundary and the pixel point of each subregion in the first image and the second image according to the mark line and a semi-global matching algorithm to obtain a plurality of matching subregions and matching parallax thereof; and obtaining a three-dimensional reconstruction result of the road surface to be detected according to the matching parallax. According to the method, the corresponding sub-regions which are highly overlapped in the first image and the second image are obtained through the constraint effect of the marking lines, then each matching sub-region in the first image and each matching sub-region in the second image are segmented and extracted, the matching parallax of each matching sub-region is calculated, the space coordinates of the matching sub-regions are obtained, the reconstruction of the whole three-dimensional texture morphology of the road surface to be detected is facilitated, and the accuracy of the detection result is improved.)

1. A detection method for three-dimensional texture morphology of a pavement is characterized by comprising the following steps:

acquiring a first image and a second image which are shot at different angles on the same road surface to be detected, wherein the first image and the second image are provided with marking lines, and the marking lines divide the first image and the second image into a plurality of sub-areas which correspond to each other one by one;

identifying the marker lines of the first and second images, respectively;

matching the boundaries and pixel points of each corresponding subregion in the first image and the second image according to the identified marking line and a preset semi-global matching algorithm to obtain a plurality of matching subregions and matching parallax of each matching subregion;

and obtaining a three-dimensional reconstruction result of the road surface to be detected according to the matching parallax.

2. The method for detecting the three-dimensional texture topography of the road surface according to claim 1, wherein the step of acquiring a first image and a second image which are taken from the same road surface to be detected at different angles, wherein the first image and the second image are both provided with marking lines, and the step of dividing the first image and the second image into a plurality of sub-areas which correspond to each other one by the marking lines comprises the steps of:

controlling a laser generating device to project a laser beam on a road surface to be detected so that the laser beam can divide the road surface to be detected into a plurality of sections;

and controlling a binocular camera to shoot the road surface to be detected so as to obtain a first image and a second image, wherein the laser beam and the plurality of segments respectively form a marking line and a plurality of sub-regions on the first image and the second image correspondingly.

3. The method for detecting the three-dimensional texture topography of the road surface according to claim 2, wherein before the step of controlling the binocular camera to shoot the road surface to be detected so as to obtain a first image and a second image, wherein the laser beam and the plurality of segments respectively form a marking line and a plurality of sub-regions on the first image and the second image, the method further comprises:

controlling the work of the reinforcing light source to adjust the illumination of the binocular camera to be 150 LUX-300 LUX;

and calibrating the binocular camera according to a Zhang Yoghuang chessboard calibration method, and obtaining camera parameters of the binocular camera, wherein the camera parameters comprise internal parameters, external parameters and distortion parameters.

4. The method for detecting the three-dimensional texture topography of the pavement according to claim 3, wherein after the step of calibrating the binocular cameras according to the jargon checker marking method and obtaining the camera parameters of the binocular cameras, wherein the camera parameters include intrinsic parameters, extrinsic parameters and distortion parameters, and before the step of identifying the marking lines of the first image and the second image respectively, further comprising:

correspondingly correcting the distortion of the first image and the second image according to the camera parameters;

according to the distribution conditions of the red, green and blue color components in the first image and the second image, correspondingly and roughly separating a target area and a background area in the first image and the second image;

and sequentially performing expansion, filling and corrosion morphological processing on the target area and the background area in the first image and the second image which are roughly separated preliminarily, correspondingly removing the background area, and obtaining the target area.

5. The method for detecting the three-dimensional texture topography of a road surface according to claim 1, wherein said step of identifying said marking lines of each of said first and second images respectively comprises:

and respectively performing decorrelation stretching processing on target areas in the first image and the second image, and correspondingly identifying the mark lines of the first image and the second image according to the difference of red and green color components in the first image and the second image after the decorrelation stretching processing.

6. The method for detecting the three-dimensional texture morphology of the road surface according to claim 1, wherein the step of obtaining the three-dimensional reconstruction result of the road surface to be detected according to the matching parallax comprises the following steps:

combining and superposing the matching parallaxes of the matching subregions to obtain a parallax matrix of a target region;

and obtaining the space coordinate of the target area according to the preset corresponding relation among the camera parameters, the parallax matrix and the space coordinate of the target area, and forming a three-dimensional reconstruction result of the road surface to be measured.

7. The method for detecting the three-dimensional texture morphology of the road surface according to claim 6, wherein the preset corresponding relationship is as follows:

wherein, (X, Y, Z) is the space coordinate of the target area, D is the parallax matrix, B is the base line distance between the left camera and the right camera in the binocular camera, and (u)₀,v₀) As the center coordinates of the binocular camera, f_xAnd f_yEffective focal lengths of the binocular camera in the X direction and the Y direction are respectively; (u, v) are pixel coordinates of the target region in the first image.

8. A system for detecting a three-dimensional texture topography of a road surface, comprising a control device, wherein the control device comprises a memory, a processor and a program for detecting a three-dimensional texture topography of a road surface, which is stored in the memory and can be run on the processor, and wherein the program for detecting a three-dimensional texture topography of a road surface is configured to implement the steps of the method for detecting a three-dimensional texture topography of a road surface according to any one of claims 1 to 7.

9. The system for detecting the three-dimensional texture topography of the road surface according to claim 8, further comprising a device body including a base, a laser generating device and a binocular camera, wherein:

the base body is provided with an installation area for loading a road surface to be tested;

the laser generating device comprises a plurality of laser generators which are arranged at intervals, the plurality of laser generators are movably arranged on the base body and project laser beams towards the mounting area;

the binocular camera is close to and keeps away from the direction of installing zone movable mounting in the pedestal.

10. A storage medium, characterized in that the storage medium stores thereon a detection program of a three-dimensional texture topography of a road surface, which when executed by a processor implements the steps of the detection method of a three-dimensional texture topography of a road surface according to any one of claims 1 to 7.

Technical Field

The invention relates to the technical field of road engineering detection, in particular to a detection method, a detection system and a storage medium for three-dimensional texture morphology of a road surface.

Background

The road engineering detection comprises the detection of the three-dimensional texture morphology of the road surface so as to better evaluate and analyze the skid resistance characteristics of the road surface. The skid resistance of the road surface is an important guarantee of driving safety, is a key control index in the road construction process, and directly influences the problems of quality acceptance of the road, road surface condition performance evaluation, decision of maintenance period and the like.

The existing method for detecting the texture morphology of the pavement mainly comprises the following steps: a sand paving method, an annular shape testing method, a laser structure depth method and the like. Among them, the sand-laying method has low efficiency, poor repeatability and is greatly influenced by human factors. The annular shape testing method and the laser structure depth rule only can finish the detection of the elevation information of a single measuring point due to the limited testing resolution, and cannot reflect the wavelength distribution and the shape structure characteristics of the texture shape of the whole pavement area to be tested, namely the testing result is a discontinuous two-dimensional index and cannot reflect the characteristics of the three-dimensional texture shape of the pavement.

Disclosure of Invention

The invention mainly aims to provide a detection method, a detection system and a storage medium for three-dimensional texture morphology of a road surface, and aims to solve the technical problem that the accuracy of a detection result is low due to the fact that the three-dimensional texture morphology information of the road surface is difficult to reflect comprehensively in the existing method for detecting the texture morphology of the road surface.

In order to achieve the purpose, the invention provides a detection method of a three-dimensional texture morphology of a pavement, which comprises the following steps:

acquiring a first image and a second image which are shot at different angles on the same road surface to be detected, wherein the first image and the second image are provided with marking lines, and the marking lines divide the first image and the second image into a plurality of sub-areas which correspond to each other one by one;

identifying the marker lines of the first and second images, respectively;

matching the boundaries and pixel points of each corresponding subregion in the first image and the second image according to the identified marking line and a preset semi-global matching algorithm to obtain a plurality of matching subregions and matching parallax of each matching subregion;

and obtaining a three-dimensional reconstruction result of the road surface to be detected according to the matching parallax.

Optionally, the acquiring a first image and a second image captured at different angles on the same road surface to be detected, where the first image and the second image both have a marking line, and the step of dividing the first image and the second image into a plurality of sub-regions in one-to-one correspondence with the marking line includes:

controlling a laser generating device to project a laser beam on a road surface to be detected so that the laser beam can divide the road surface to be detected into a plurality of sections;

and controlling a binocular camera to shoot the road surface to be detected so as to obtain a first image and a second image, wherein the laser beam and the plurality of segments respectively form a marking line and a plurality of sub-regions on the first image and the second image correspondingly.

Optionally, before the step of controlling the binocular camera to shoot the road surface to be measured to obtain a first image and a second image, wherein the laser beam and the plurality of segments respectively form a marking line and a plurality of sub-regions on the first image and the second image, the method further includes:

controlling the work of the reinforcing light source to adjust the illumination of the binocular camera to be 150 LUX-300 LUX;

and calibrating the binocular camera according to a Zhang Yoghuang chessboard calibration method, and obtaining camera parameters of the binocular camera, wherein the camera parameters comprise internal parameters, external parameters and distortion parameters.

Optionally, after the step of calibrating the binocular camera according to the euhedral checkerboard calibration method and obtaining camera parameters of the binocular camera, where the camera parameters include an inside parameter, an outside parameter and a distortion parameter, and before the step of identifying the mark lines of the first image and the second image respectively, the method further includes:

correspondingly correcting the distortion of the first image and the second image according to the camera parameters;

according to the distribution conditions of the red, green and blue color components in the first image and the second image, correspondingly and roughly separating a target area and a background area in the first image and the second image;

and sequentially performing expansion, filling and corrosion morphological processing on the target area and the background area in the first image and the second image which are roughly separated preliminarily, correspondingly removing the background area, and obtaining the target area.

Optionally, the step of identifying the marking lines of the first and second images respectively comprises:

and respectively performing decorrelation stretching processing on target areas in the first image and the second image, and correspondingly identifying the mark lines of the first image and the second image according to the difference of red and green color components in the first image and the second image after the decorrelation stretching processing.

Optionally, the step of obtaining a three-dimensional reconstruction result of the road surface to be detected according to the matching parallax includes:

combining and superposing the matching parallaxes of the matching subregions to obtain a parallax matrix of a target region;

and obtaining the space coordinate of the target area according to the preset corresponding relation among the camera parameters, the parallax matrix and the space coordinate of the target area, and forming a three-dimensional reconstruction result of the road surface to be measured.

Optionally, the preset corresponding relationship is:

wherein, (X, Y, Z) is the space coordinate of the target area, D is the parallax matrix, B is the base line distance between the left camera and the right camera in the binocular camera, and (u)₀,v₀) As the center coordinates of the binocular camera, f_xAnd f_yEffective focal lengths of the binocular camera in the X direction and the Y direction are respectively; (u, v) are pixel coordinates of the target region in the first image.

In addition, the invention also provides a detection system of the three-dimensional texture topography of the road surface, which comprises a control device, wherein the control device comprises a memory, a processor and a detection program of the three-dimensional texture topography of the road surface, which is stored on the memory and can be operated on the processor, and the detection program of the three-dimensional texture topography of the road surface is configured to realize the steps of the detection method of the three-dimensional texture topography of the road surface.

Optionally, the detection system for the three-dimensional texture topography of the road surface further comprises a device main body, wherein the device main body comprises a base body, a laser generating device and a binocular camera, wherein:

the base body is provided with an installation area for loading a road surface to be tested;

the laser generating device comprises a plurality of laser generators which are arranged at intervals, the plurality of laser generators are movably arranged on the base body and project laser beams towards the mounting area;

the binocular camera is close to and keeps away from the direction of installing zone movable mounting in the pedestal.

In addition, the invention also provides a storage medium, wherein the storage medium is stored with a detection program of the three-dimensional texture topography of the road surface, and the detection program of the three-dimensional texture topography of the road surface realizes the steps of the detection method of the three-dimensional texture topography of the road surface when being executed by a processor.

According to the technical scheme provided by the invention, the first image and the second image are used for shooting different angles of the same road surface to be detected, corresponding sub-areas which are highly overlapped in the first image and the second image are obtained through the constraint action of the marking lines, and then the space coordinate of each matching sub-area is obtained through determining the matching parallax of each matching sub-area, so that the reconstruction of the integral three-dimensional texture morphology of the road surface to be detected is facilitated, and the accuracy of the detection result is improved.

Drawings

Fig. 1 is a schematic structural diagram of a control device of a hardware operating environment according to an embodiment of the present invention;

fig. 2 is a schematic flow chart of the method for detecting three-dimensional texture features provided by the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The invention provides a system for detecting a three-dimensional texture topography, which at least comprises a control device, and please refer to fig. 1, wherein fig. 1 is a schematic structural diagram of the control device of a hardware operating environment according to an embodiment of the present invention.

As shown in fig. 1, the control device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.

Those skilled in the art will appreciate that the configuration of the control device shown in fig. 1 does not constitute a limitation of the control device and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components.

As shown in fig. 1, the memory 1005, which is a kind of computer storage medium, may include an operating system, a network communication module, a user interface module, and a detection program of the three-dimensional texture topography of the road surface.

In the control apparatus shown in fig. 1, the processor 1001 and the memory 1005 may be provided in a three-dimensional texture feature detection system, and the three-dimensional texture feature detection system calls a three-dimensional texture feature detection program stored in the memory 1005 through the processor 1001 and performs the following operations:

acquiring a first image and a second image which are shot at different angles on the same road surface to be detected, wherein the first image and the second image are provided with marking lines, and the marking lines divide the first image and the second image into a plurality of sub-areas which correspond to each other one by one;

identifying the marker lines of the first and second images, respectively;

matching the boundaries and pixel points of each corresponding subregion in the first image and the second image according to the identified marking line and a preset semi-global matching algorithm to obtain a plurality of matching subregions and matching parallax of each matching subregion;

and obtaining a three-dimensional reconstruction result of the road surface to be detected according to the matching parallax. .

Further, the processor 1001 may call the detection program of the three-dimensional texture topography stored in the memory 1005, and further perform the following operations:

the method comprises the steps of obtaining a first image and a second image which are shot at different angles on the same road surface to be detected, wherein the first image and the second image are provided with marking lines, and the marking lines divide the first image and the second image into a plurality of sub-areas which correspond to each other one by one, and the steps comprise:

controlling a laser generating device to project a laser beam on a road surface to be detected so that the laser beam can divide the road surface to be detected into a plurality of sections;

and controlling a binocular camera to shoot the road surface to be detected so as to obtain a first image and a second image, wherein the laser beam and the plurality of segments respectively form a marking line and a plurality of sub-regions on the first image and the second image correspondingly.

Further, the processor 1001 may call the detection program of the three-dimensional texture topography stored in the memory 1005, and further perform the following operations:

before the step of controlling the binocular camera to shoot the road surface to be measured so as to obtain a first image and a second image, wherein the laser beam and the plurality of segments respectively and correspondingly form a marking line and a plurality of sub-regions on the first image and the second image, the method further comprises the following steps of:

controlling the work of the reinforcing light source to adjust the illumination of the binocular camera to be 150 LUX-300 LUX;

and calibrating the binocular camera according to a Zhang Yoghuang chessboard calibration method, and obtaining camera parameters of the binocular camera, wherein the camera parameters comprise internal parameters, external parameters and distortion parameters.

Further, the processor 1001 may call the detection program of the three-dimensional texture topography stored in the memory 1005, and further perform the following operations:

calibrating the binocular camera according to a Zhang-York calibration method, and obtaining camera parameters of the binocular camera, wherein the camera parameters include an inside parameter, an outside parameter and a distortion parameter, and further includes, after the step of identifying the respective marking lines of the first image and the second image, respectively:

correspondingly correcting the distortion of the first image and the second image according to the camera parameters;

according to the distribution conditions of the red, green and blue color components in the first image and the second image, correspondingly and roughly separating a target area and a background area in the first image and the second image;

and sequentially performing expansion, filling and corrosion morphological processing on the target area and the background area in the first image and the second image which are roughly separated preliminarily, correspondingly removing the background area, and obtaining the target area.

Further, the processor 1001 may call the detection program of the three-dimensional texture topography stored in the memory 1005, and further perform the following operations:

the step of identifying the marking lines of each of the first and second images, respectively, comprises:

and respectively performing decorrelation stretching processing on target areas in the first image and the second image, and correspondingly identifying the mark lines of the first image and the second image according to the difference of red and green color components in the first image and the second image after the decorrelation stretching processing.

Further, the processor 1001 may call the detection program of the three-dimensional texture topography stored in the memory 1005, and further perform the following operations:

the step of obtaining the three-dimensional reconstruction result of the road surface to be detected according to the matching parallax comprises the following steps:

combining and superposing the matching parallaxes of the matching subregions to obtain a parallax matrix of a target region;

and obtaining the space coordinate of the target area according to the preset corresponding relation among the camera parameters, the parallax matrix and the space coordinate of the target area, and forming a three-dimensional reconstruction result of the road surface to be measured.

Further, the processor 1001 may call the detection program of the three-dimensional texture topography stored in the memory 1005, and further perform the following operations:

the preset corresponding relation is as follows:

wherein, (X, Y, Z) is the space coordinate of the target area, D is the parallax matrix, B is the base line distance between the left camera and the right camera in the binocular camera, and (u)₀,v₀) As the center coordinates of the binocular camera, f_xAnd f_yEffective focal lengths of the binocular camera in the X direction and the Y direction are respectively; (u, v) are pixel coordinates of the target region in the first image.

According to the technical scheme provided by the invention, the first image and the second image shoot different angles of the same road surface to be detected, the matching sub-regions which are highly overlapped in the first image and the second image are obtained through the constraint action of the marking lines, and then the space coordinate of each matching sub-region is obtained through determining the matching parallax of each matching sub-region, so that the reconstruction of the integral three-dimensional texture morphology of the road surface to be detected is facilitated, and the accuracy of the detection result is improved.

In addition, the detection system of the three-dimensional texture morphology can further comprise a device main body, wherein the device main body comprises a base body, a laser generating device and a binocular camera, and the base body is provided with an installation area used for loading a road surface to be detected; the laser generating device comprises a plurality of laser generators which are arranged at intervals, the plurality of laser generators are movably arranged on the base body and project laser beams towards the mounting area; the binocular camera is close to and keeps away from the direction of installing zone movable mounting in the pedestal.

Specifically, the specific expression form of the seat body is not limited, and the seat body may be a box structure or a rack structure as required. The mounting area is for example a load board, the road surface of awaiting measuring of spacing installation on the load board, the pedestal is including erectting the first crossbeam of load board top, first spout has been seted up on the first crossbeam, two mesh cameras pass through a slider sliding connection first spout, two cameras all face about two mesh cameras load the board setting. Wherein, the binocular camera can be selected but not limited to: the outer diameter of the lens is 15mm, the center distance is 6cm, and the resolution and the frame rate are 2560 multiplied by [email protected] fps. It should be noted that the binocular camera can be used for directly purchasing existing products, and two cameras with the same specification can be assembled on the sliding block at intervals through the circuit board so as to realize simultaneous snapshot on the same road surface to be detected.

The base body further comprises a second cross beam arranged at intervals of the first cross beam, similarly, a second sliding groove is formed in the second cross beam, and the laser generating device is slidably mounted in the second sliding groove through another sliding block. The laser generating device comprises a plurality of laser generators, and the laser generators respectively project a plurality of lasers towards the bearing plate so as to divide the road surface to be measured into a plurality of sections. The plurality of lasers collectively form a laser beam. The number of the laser generators is not limited, but if the number of the laser generators is large, namely the number of the generated laser beams is large, the laser beams are distributed on the road surface to be detected more densely, so that the laser beams are easy to interfere with each other, and the subsequent laser beam extraction and three-dimensional reconstruction work are not facilitated; on the contrary, if the set number of the laser generators is small, that is, the number of the generated laser beams is small, the distribution on the road surface to be detected is sparse, and the formed segmentation area is large, so that the matching of subsequent pixel points is not facilitated. Therefore, the number of the laser generators is set to be within a suitable range, and preferably, 6.

Further, the sliding blocks on the first sliding groove and the second sliding groove are preferably made of iron materials, the sliding blocks are provided with mounting holes, and the mounting holes are detachably mounted on the first sliding groove or the second sliding groove through fastening screws. Each laser generator respectively through the strong magnetism support adsorbablely install in of a stereoplasm hose on the slider, the stereoplasm hose is through adjusting laser generator's transmitting surface orientation, the incident position, the angle and the direction of laser are conveniently adjusted.

In addition, the detecting system of three-dimensional texture appearance still includes the reinforcement light source, the reinforcement light source is cavity annular light source, the reinforcement light source cover is established the periphery of binocular camera, the parameter adaptation of reinforcement light source in binocular camera setting, optional but not the restriction is as: the inner diameter is 16mm, the outer diameter is 34mm, the working voltage is 24V, the power is 3.5W, and the brightness can be manually adjusted through a knob.

Based on the hardware structure, the embodiment of the detection method for the three-dimensional texture morphology of the pavement is provided.

Referring to fig. 2, fig. 2 is a schematic flow chart of an embodiment of the method for detecting a three-dimensional texture topography of a road surface of the invention.

In this embodiment, the method for detecting the three-dimensional texture topography of the road surface includes the following steps:

s10: acquiring a first image and a second image which are shot at different angles on the same road surface to be detected, wherein the first image and the second image are provided with marking lines, and the marking lines divide the first image and the second image into a plurality of sub-areas which correspond to each other one by one;

in this embodiment, when the detection system for the three-dimensional texture topography of the road surface does not include a photographing device, the first image and the second image may be acquired by an external device; when the detection system for the three-dimensional texture topography of the road surface comprises the binocular cameras, the first image and the second image can be obtained by respectively shooting the left camera and the right camera of the binocular cameras; because the shot object is the same road surface to be detected, the image shapes and the basic features of the first image and the second image are consistent, the marking lines on the first image and the marking lines on the second image are arranged in a one-to-one correspondence mode, and the plurality of sub-areas on the first image and the plurality of sub-areas on the second image are arranged in a one-to-one correspondence mode. The specific expression form of the marking lines is not limited, and the marking lines can be embodied as transverse or longitudinal grid lines, or as transverse and longitudinal staggered grid lines.

In addition, the marking line may be directly marked on the first image and the second image, but in order to ensure the uniformity of both the first image and the second image, in the present embodiment, it is preferable to form a dividing line on the road surface to be measured by handwriting, a light beam, or other means in advance, and the dividing line is developed on the first image and the second image to form the marking line, and the following embodiments are all explained based on this setting.

S30: identifying the marker lines of the first and second images, respectively;

since the mark line formed after the first image and the second image are developed by the dividing line is easily blurred due to the influence of factors such as resolution, illumination, color, and the like, in this embodiment, it is preferable to perform a related stretching process on the mark line first, so as to quickly identify the mark line, so that the boundary of each sub-area on the first image and the second image is clear, and the boundary matching operation performed later is facilitated.

S40: matching the boundaries and pixel points of each corresponding subregion in the first image and the second image according to the identified marking line and a preset semi-global matching algorithm to obtain a plurality of matching subregions and matching parallax of each matching subregion;

in this embodiment, the first image and the second image are taken from different angles on the same road surface to be measured, according to the principle of parallax, a parallax exists between an imaging point formed on the first image and a corresponding imaging point formed on the second image for one base point on the road surface to be measured, and by solving the parallax, the three-dimensional space coordinate of the base point can be obtained. For the first image and the second image with larger imaging areas, the defects of complex operation and low accuracy exist in the process of directly matching the first image and the second image, and the first image and the second image need to be converted into matching between the corresponding sub-areas, and then matching between corresponding pixel points is carried out in the corresponding sub-areas. In view of the above, identifying the mark lines of the first image and the second image, that is, obtaining the boundary line of each sub-region, and by overlapping the corresponding mark lines, the corresponding sub-regions can be substantially determined; then, through the semi-global matching algorithm, the boundary of the sub-region can be forcibly matched, and the matching between the corresponding imaging points in the sub-region can be rapidly and efficiently obtained, so that the corresponding matching sub-region and the matching parallax thereof are obtained. The semi-global matching algorithm may refer to the prior art, and is not described in detail herein. Forced matching of the boundary, namely, strictly establishing matching between the corresponding pixel points at the boundary in the corresponding sub-region by multiplying the color brightness of the marking lines in the first image and the second image; in contrast, the non-boundary pixels can be selected to match with other pixels adjacent to the corresponding pixels.

There are various technical schemes for obtaining the matching sub-region, and one embodiment thereof is described as follows: firstly, identifying a marking line, and generating position coordinate information and sequencing information of the marking line in the first image and the second image respectively; then, setting a segmentation template according to the position coordinate information and the sequencing information, wherein the segmentation template is that under the sequencing of the specific marking lines, the unextracted area part on the left of the position coordinate of the marking line is set as 1, and the rest is set as 0; and then multiplying the first image and the second image by the segmentation template respectively to complete the segmentation and extraction of each matching subarea.

S50: and obtaining a three-dimensional reconstruction result of the road surface to be detected according to the matching parallax.

Based on the above, on the basis of obtaining the matching parallax between the corresponding pixel points in each matching sub-region, the matching parallax of each matching sub-region can be combined and superposed to obtain the parallax matrix of the road surface target region, the three-dimensional space coordinate of the road surface target region can be solved through the prior art, that is, the three-dimensional model of the matching sub-region where the pixel points are located can be reconstructed, and finally the purpose of reconstructing the whole three-dimensional model of the road surface to be detected is achieved, and the three-dimensional reconstruction result of the road surface to be detected is obtained, so that the follow-up visual inspection, query or measurement of the three-dimensional texture appearance condition of the road surface to be detected based on the three-dimensional reconstruction result is. In view of the above, when the detection system for the three-dimensional texture features of the road surface not only includes the control device, but also integrates the binocular camera and the laser generating device, the step S10 includes:

s11: controlling a laser generating device to project a laser beam on a road surface to be detected so that the laser beam can divide the road surface to be detected into a plurality of sections;

in this embodiment, the laser beam is selected as a grating formed by a plurality of parallel lasers arranged at equal intervals, and the grating is projected on the road surface to be measured and divides the road surface to be measured into a plurality of segments. Compare in operation such as the road surface of awaiting measuring is carved with the line strip, the projection of grating can not destroy the surperficial integrality on road surface of awaiting measuring also simplifies the work load of carrying out special measurement and mark through instruments such as chi, has convenient operation's advantage.

S14: and controlling a binocular camera to shoot the road surface to be detected so as to obtain a first image and a second image, wherein the laser beam and the plurality of segments respectively form a marking line and a plurality of sub-regions on the first image and the second image correspondingly.

In this embodiment, two cameras among the binocular camera can refer to prior art and adjust specific mounted position, then towards the road surface that awaits measuring shoots. The road surface to be detected is imaged on the first image and the second image, the laser line beams form the marking lines, the sub-areas are formed by the plurality of segments, the marking lines and the sub-areas in the first image and the second image are ensured to be respectively in one-to-one correspondence, and the subsequent matching process is facilitated to be simplified.

Further, in this embodiment, before the step S14, the method further includes:

s12: controlling the work of the reinforcing light source to adjust the illumination of the binocular camera to be 150 LUX-300 LUX;

s13: and calibrating the binocular camera according to a Zhang Yoghuang chessboard calibration method, and obtaining camera parameters of the binocular camera, wherein the camera parameters comprise internal parameters, external parameters and distortion parameters.

It should be noted that the purpose of dimming the binocular camera by the reinforcing light source is to stabilize the imaging quality, and the reinforcing light source can always ensure that the illuminance obtained by the binocular camera is balanced and stable no matter whether the ambient light environment is dark or bright; in addition, the reinforcing light source is arranged around the periphery of the binocular camera, so that the binocular camera can be adjusted in light in all directions, the influence of light and shadow on the road surface to be detected is eliminated, and the imaging quality is further optimized.

Among the above-mentioned camera parameters, the internal parameters are also parameters related to the characteristics of the binocular cameras themselves, such as focal lengths, pixel sizes, and the like of two cameras in the binocular cameras; the external parameters are parameters of the binocular camera in a world coordinate system, such as the position and the rotation direction of the binocular camera; the distortion parameters comprise a radial distortion coefficient and a tangential distortion coefficient, and the radial distortion occurs in the process of converting a camera coordinate system into a physical coordinate system; the reason for the tangential distortion is that the lens is not perfectly parallel to the image. In addition, the specific steps of the Zhang YOU positive chessboard marking method can refer to the prior art, and are not detailed here.

Since the binocular camera inevitably shoots other regions except the road surface to be detected during shooting, and names the regions on the first image and the second image, the imaging region of the road surface to be detected is a target region, and the other regions except the target region are background regions, and since the background regions have no research value and do not need to be three-dimensionally reconstructed, in this embodiment, after the step S13 and before the step S30, the method further includes:

s21: correspondingly correcting the distortion of the first image and the second image according to the camera parameters;

s22: according to the distribution conditions of the red, green and blue color components in the first image and the second image, correspondingly and roughly separating a target area and a background area in the first image and the second image;

s23: and sequentially performing expansion, filling and corrosion morphological processing on the target area and the background area in the first image and the second image respectively, correspondingly removing the background area, and obtaining the target area.

In this embodiment, first, according to the distortion parameter, the first image and the second image are corrected, lines which generate distortion are corrected, and the like, so as to reduce the influence on the accuracy of the final detection result; and finally, completely removing the background region through a series of existing image processing means, leaving the more accurate target region, and being beneficial to reducing the workload of subsequent pixel point matching and parallax calculation, thereby obtaining a three-dimensional reconstruction result more quickly and efficiently.

It should be noted that there are various solutions for identifying the mark line from the image, and in this embodiment, the step S30 includes:

s31: and respectively performing decorrelation stretching processing on the target areas in the first image and the second image, and correspondingly identifying the mark lines of the first image and the second image according to the difference of red and green color components in the first image and the second image after the decorrelation stretching processing.

The decorrelation stretching processing is the prior art, and the basic principle is to perform coordinate transformation on an original image, then selectively stretch the transformed image, and finally recover the image through coordinate inverse transformation. In this embodiment, by performing decorrelation stretching processing on the first image and the second image, clearer and more accurate mark lines in the first image and the second image can be correspondingly identified. The specific operation can be referred to the prior art and is not detailed here.

Further, in this embodiment, the step S50 includes:

s51: combining and superposing the matching parallaxes of the matching sub-regions to obtain a parallax matrix of the target region;

s52: and obtaining the space coordinate of the target area according to the preset corresponding relation among the camera parameters, the parallax matrix and the space coordinate of the target area, and forming a three-dimensional reconstruction result of the road surface to be measured.

By repeating the above steps, the matching parallax of each matching sub-region is sequentially calculated, and then, with reference to the prior art, the obtained multiple matching parallaxes are combined and superimposed to obtain the parallax matrix of the whole target region. According to the known camera parameters and the parallax matrix, the space coordinates of the target area can be calculated. Therefore, the detection resolution can be greatly improved, the detection resolution reaches the pixel level, namely the size of one pixel reaches 0.02mm, and the method can be used for constructing multiple attribute characterization indexes and more comprehensively evaluating and analyzing the road surface skid resistance.

Specifically, in this embodiment, the preset corresponding relationship is:

wherein, (X, Y, Z) is the space coordinate of the target area, D is the parallax matrix, B is the base line distance between the left camera and the right camera in the binocular camera, and (u)₀,v₀) As the center coordinates of the binocular camera, f_xAnd f_yEffective focal lengths of the binocular camera in the X direction and the Y direction are respectively; (u, v) are pixel coordinates of the target region in the first image.

By establishing the preset corresponding relation, the space coordinates of each pixel point in the target area can be obtained more quickly and accurately under the condition that the camera parameters and the parallax matrix are known, so that the calculation process is simplified. In this embodiment, the control device may include MATLAB software, and the running time of the above processing flow and calculation process in the MATLAB software is only 4.9680s, which greatly improves the detection efficiency, and the repeatability of each solution result can be ensured by the control device completing the above processing flow and calculation process.

In order to more intuitively express the high-precision effect of the detection method for the three-dimensional texture morphology of the pavement, a conventional detection method, namely a laser-beam-free traditional binocular reconstruction algorithm is used as a comparative example, 7 position points are marked on the surface of the pavement to be detected by adopting a point laser, and the position points are sequentially marked as 1^#、2^#、3^#、4^#、5^#、6^#And 7^#Then, the elevations of 7 position points were measured with 1 of them using a vernier caliper with an accuracy of 0.01mm^#And taking the point as a reference point, calculating the elevation difference of the remaining six position points, and respectively calculating the absolute deviation and the relative deviation.

The results of analysis and evaluation of the test effects of the present example and comparative example are shown in table 1 below.

Table 1 analysis and evaluation results of test effects of examples and comparative examples

As can be seen from Table 1, the average relative deviation of the comparative example calculation result of the conventional binocular reconstruction algorithm without the laser constraint line reaches 48.38%, and the maximum relative deviation is more than 142.80%, which indicates that the conventional binocular reconstruction algorithm has poor detection effect and poor detection precision on the three-dimensional texture morphology of the road surface. The average absolute deviation of the calculation result of the embodiment is only 0.0222mm, the average relative deviation is only 3.61%, and even the maximum relative deviation is only 43.01%, so that the detection effect is obviously improved, and the detection method provided by the invention is proved to be capable of accurately detecting the three-dimensional texture topography of the road surface.

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.