Paving width detection method and device, readable storage medium and paver

Abstract

The present invention proposes a paving width detection method and device, a readable storage medium, and a paver. The paving width detection method includes: acquiring image information of a paved road area through a calibrated vehicle-mounted image acquisition device; extracting an RGB image and a depth image of the paved road area from the image information of the paved road area; determining edge position information of a loose paved material area in the paved road area based on the RGB image; extracting point cloud data corresponding to the edge position information of the loose paved material area from the depth image; and determining the paving width based on the point cloud data. The paving width detection method provided by the present invention calculates the paving width value from the image of the paved road surface, which is a direct detection method. Compared with the indirect measurement of paving width in related technologies, such as using a distance sensor to measure the width of the ironing board, the detection result is more accurate, which can better help judge the paving quality and measure and predict the engineering quantity.

Description

Technical Field

[0001] The present invention relates to the technical field of paving machines, and in particular to a paving width detection method and device, a readable storage medium, and a paver. Background Art

[0002] Since paving width is an important indicator for measuring the quality of pavement paving construction and predicting paving engineering quantities, it is critical to obtain and record paving width in real time.

[0003] In the related art, an indirect method of measuring the paving width is generally adopted. For example, a distance sensor such as a laser sensor is installed on the ironing plate, and the width of the ironing plate is measured by the distance sensor to indirectly measure the paving width. The above indirect measurement method has the following main disadvantages: first, the measurement accuracy will be affected by many factors, such as the arch of the ironing plate, etc.; second, it is not flexible enough. When assembling the ironing plate, the installation position of the distance sensor may need to be adjusted. Summary of the Invention

[0004] The present invention aims to solve or improve at least one of the technical problems existing in the prior art or related art.

[0005] To this end, a first aspect of the present invention provides a paving width detection method based on a vehicle-mounted image acquisition device.

[0006] A second aspect of the present invention provides a paving width detection device based on a vehicle-mounted image acquisition device.

[0007] A third aspect of the present invention provides a paving width detection device based on a vehicle-mounted image acquisition device.

[0008] A fourth aspect of the present invention provides a readable storage medium.

[0009] A fifth aspect of the present invention provides a paving machine.

[0010] In view of this, according to the first aspect of the present invention, a paving width detection method based on a vehicle-mounted image acquisition device is proposed, including: acquiring image information of the paved road area through a calibrated vehicle-mounted image acquisition device; extracting an RGB image and a depth image of the paved road area from the image information of the paved road area; determining the edge position information of the loose paving material area in the paved road area based on the RGB image; extracting point cloud data corresponding to the edge position information of the loose paving material area from the depth image; and determining the paving width based on the point cloud data.

[0011] The paving width detection method provided by the present invention first obtains image information of the paved road area through a calibrated vehicle-mounted image acquisition device, and extracts the RGB image and depth image of the paved road area therefrom. Then, by pre-processing the RGB image of the paved road, the edge position information of the loose paving material area in the paved road area can be located. Point cloud data corresponding to the edge position information of the loose paving material area is extracted from the depth image. The paving width can be calculated based on the above point cloud data. The paving width is obtained by calculating the paving width from the image of the paved road area. It is a direct detection method, and is therefore applicable to both straight-plate pavers and telescopic pavers. Compared with the related art of using a distance sensor to measure the width of the ironing board and indirectly measuring the paving width, the detection result is more accurate, which can better help judge the paving quality and measure and predict the engineering quantity.

[0012] Among them, RGB image is an image in RGB color mode (Red Green Blue color mode). RGB color mode is a color standard in the industry. It obtains various colors by changing the three color channels of red, green, and blue and superimposing them on each other.

[0013] A depth image, also known as a range image, stores three-dimensional depth feature information. It is an image captured by an image collector using the distance (depth) values ​​of each point in a scene as pixel values. It directly reflects the geometry of the visible surface of the scene. Depth images can be converted into point cloud data through coordinate transformation.

[0014] Specifically, the vehicle-mounted image acquisition device may be a binocular vision camera or a TOF (Time of Fight Camera) camera.

[0015] The paving width detection method according to the present invention may also have the following technical features:

[0016] In the above technical solution, the step of determining the edge position information of the loose paved material area in the paved road area based on the RGB image specifically includes: determining the loose paved material area in the RGB image based on the image feature value; determining the corresponding edge position information based on the loose paved material area; wherein the image feature value includes any one of the following items or a combination thereof: color value, brightness value, grayscale value, texture value.

[0017] In this technical solution, when determining the edge position information of loose paving material areas within a paved road surface area based on an RGB image of the paved road surface, the image feature values ​​presented by the loose paving material areas in the image are significantly different from those presented by road surface areas without loose paving material or road boundary areas (such as boundary ground or wall surfaces). Therefore, the loose paving material areas can be determined based on the changes in the image feature values. Furthermore, the corresponding edge position information can be located based on the loose paving material areas.

[0018] The image feature value includes any one of a color value, a brightness value, a grayscale value, and a texture value, or a combination thereof, but is not limited thereto.

[0019] In any of the above technical solutions, the step of extracting the loose material area from the RGB image based on the image eigenvalues ​​specifically includes: taking any point in the loose material area as a seed point, taking the changing gradient of the image eigenvalues ​​as a growth condition, using a regional growing algorithm to determine the loose material area, and stopping the growth when the changing gradient of the image eigenvalues ​​is greater than a preset value.

[0020] In this technical solution, when determining the loose paving material area in the RGB image based on the image eigenvalues, a region growing algorithm can be used to extract the loose paving material area from the RGB image. Among them, the seed point can be selected as any point in the loose paving material area, and the change gradient of the image eigenvalues ​​is used as the growth condition. Pixels of similar properties are merged and generated in the image until the change gradient of the image eigenvalues ​​is greater than a preset value and the growth stops. Through the technical solution of the present invention, the paved pavement area on the paved road surface can be accurately determined, thereby obtaining more accurate edge position information, thereby calculating the paving width, and the detection result is more accurate.

[0021] Region growing is the process of growing groups of pixels or regions into larger regions. Starting from a set of seed points, regions are grown from these points by merging neighboring pixels with similar properties to each seed point, such as intensity, grayscale, texture color, etc.

[0022] In any of the above technical solutions, the step of determining the paving width based on the point cloud data specifically includes: the point cloud data includes first point cloud data and second point cloud data, and a boundary line is fitted according to the first point cloud data; the average distance from the second point cloud data to the boundary line is calculated, and the average distance is recorded as the paving width; wherein the first point cloud data and the second point cloud data respectively correspond to the first edge position information and the second edge position information of the loose paving material area in the direction perpendicular to the road extension direction.

[0023] In this technical solution, the point cloud data corresponding to the edge position information of the loose paving material area includes first point cloud data and second point cloud data. The first point cloud data and the second point cloud data correspond to the first edge position information and the second edge position information of the loose paving material area in a direction perpendicular to the extension of the road, respectively. When determining the paving width based on the above point cloud data, a boundary line is first fitted based on the first point cloud data, and then the average distance from the second point cloud data to the boundary line is calculated, and the average distance is recorded as the paving width. The calculation method of the present invention is simple and reliable, thereby obtaining a more accurate and reliable paving width, which is conducive to improving the paving quality.

[0024] Specifically, a first boundary line can be fitted based on the first point cloud data, and a second boundary line can be fitted based on the second point cloud data. The average distance between the first boundary line and the second boundary line can be calculated, and the average distance can be recorded as the paving width.

[0025] In any of the above technical solutions, before the step of obtaining image information of the paved road area through a calibrated vehicle-mounted image acquisition device, it also includes: obtaining image information of the road area through the vehicle-mounted image acquisition device; extracting a depth image of the road area from the image information of the road area; extracting third point cloud data corresponding to the road area other than the road boundary area and fourth point cloud data corresponding to the road boundary area from the depth image; calculating the point cloud normal vector based on the third point cloud data and the fourth point cloud data; calculating the external parameters of the vehicle-mounted image acquisition device based on the point cloud normal vector and the reference normal vector to calibrate the depth image collected by the vehicle-mounted image acquisition device; wherein the road area is a paved road area or a road area to be paved.

[0026] In this technical solution, before acquiring image information of a paved road surface area using a calibrated on-board image acquisition device, the on-board image acquisition device must undergo extrinsic calibration. Specifically, the on-board image acquisition device acquires image information of the road surface area and extracts a depth image of the road surface area. From this depth image, third point cloud data corresponding to the road surface area excluding the road boundary area and fourth point cloud data corresponding to the road boundary area are extracted. A point cloud normal vector can be calculated using these third and fourth point cloud data. By calculating this point cloud normal vector against a reference normal vector, the extrinsic parameters of the on-board image acquisition device can be obtained. Based on these extrinsic parameters, the on-board image acquisition device can be calibrated. Only after calibration is complete can the coordinate system of the on-board image acquisition device be associated and unified with the world coordinate system, i.e., the same coordinate system as the paver and screed. This ensures that the image data acquired by the on-board image acquisition device is reliable and consistent with reality. When the image data is reliable, the calculated paving width is more realistic and reliable.

[0027] In addition, it should be noted that, during the above calibration process, the depth image used is not limited to the depth image of the paved road area, but the depth image of the road area to be paved is also applicable.

[0028] In any of the above technical solutions, it also includes: obtaining the location information of the paved road surface; drawing a paving width map based on the location information and paving width of the paved road surface; and sending the paving width map to a display device for display.

[0029] In this technical solution, by obtaining the location information of the paved road surface and combining it with the paving width, a paving width map of the sampled paved road surface can be obtained, and the paving width map is sent to the display device for display. This allows a more intuitive view of the paving width of each section of road, which can be used to help judge the paving quality and the measurement and prediction of engineering quantities.

[0030] According to the second aspect of the present invention, a paving width detection device based on a vehicle-mounted image acquisition device is proposed, comprising: a memory storing programs or instructions; a processor, which implements the steps of the paving width detection method based on a vehicle-mounted image acquisition device as described in any of the above technical solutions when executing the programs or instructions.

[0031] The paving width detection device based on the vehicle-mounted image acquisition device provided by the present invention includes a memory and a processor, wherein the memory stores a program or instruction, and when the processor executes the above program or instruction, the steps of the paving width detection method as described in any of the above technical solutions are implemented. Therefore, the paving width detection device has all the beneficial effects of the paving width detection method, which will not be discussed one by one here.

[0032] According to a third aspect of the present invention, a paving width detection device based on a vehicle-mounted image acquisition device is proposed, comprising: an acquisition unit for acquiring image information of a paved road area through a calibrated vehicle-mounted image acquisition device; an extraction unit for extracting an RGB image and a depth image of the paved road area from the image information of the paved road area; a first processing unit for determining edge position information of a loose paved material area in the paved road area based on the RGB image; a second processing unit for extracting point cloud data corresponding to the edge position information from the depth image; and a calculation unit for determining the paving width based on the point cloud data.

[0033] The paving width detection device provided by the present invention includes an acquisition unit, an extraction unit, a first processing unit, a second processing unit and a calculation unit. First, the acquisition unit acquires image information of the paved road area through a calibrated vehicle-mounted image acquisition device, and the extraction unit extracts the RGB image and depth image of the paved road area from it. Then, the first processing unit can locate the edge position information of the loose paving material area in the paved road area by pre-processing the RGB image of the paved road. The second processing unit extracts point cloud data corresponding to the edge position information of the loose paving material area from the depth image. The calculation unit can calculate the paving width based on the above point cloud data. The paving width detection device provided by the present invention obtains the paving width by calculating the image of the paved road area. It is a direct detection method, so it is applicable to both straight-plate pavers and telescopic pavers. Compared with the related art of using a distance sensor to measure the width of the ironing board and indirectly measuring the paving width, the detection result is more accurate, which can better help judge the paving quality and measure and predict the engineering quantity.

[0034] Among them, RGB image is an image in RGB color mode (Red Green Blue color mode). RGB color mode is a color standard in the industry. It obtains various colors by changing the three color channels of red, green, and blue and superimposing them on each other.

[0035] A depth image, also known as a range image, stores three-dimensional depth feature information. It is an image captured by an image collector using the distance (depth) values ​​of each point in a scene as pixel values. It directly reflects the geometry of the visible surface of the scene. Depth images can be converted into point cloud data through coordinate transformation.

[0036] Specifically, the vehicle-mounted image acquisition device may be a binocular vision camera or a TOF camera.

[0037] In the above technical solution, the first processing unit is specifically used to determine the loose material area in the RGB image based on the image feature value; determine the corresponding edge position information based on the loose material area; wherein the image feature value includes any one of the following items or a combination thereof: color value, brightness value, grayscale value, texture value.

[0038] In this technical solution, when the first processing unit determines the edge position information of the loose paving material area within the paved road area based on the RGB image of the paved road surface, the image feature values ​​presented by the loose paving material area in the image are significantly different from the image feature values ​​presented by the unpaved road area or the road boundary area (such as the boundary ground or wall surface). Therefore, the loose paving material area can be determined based on the change in the image feature values. The corresponding edge position information can then be located based on the loose paving material area.

[0039] The image feature value includes any one of a color value, a brightness value, a grayscale value, and a texture value, or a combination thereof, but is not limited thereto.

[0040] In any of the above technical solutions, the first processing unit is specifically used to use any point in the loose material area as a seed point, the changing gradient of the image eigenvalue as a growth condition, and adopt a regional growing algorithm to determine the loose material area, and stop growing when the changing gradient of the image eigenvalue is greater than a preset value.

[0041] In this technical solution, when the first processing unit determines the loose paving material area in the RGB image based on the image eigenvalues, the region growing algorithm can be used to extract the loose paving material area from the RGB image. Among them, the seed point can be selected as any point in the loose paving material area, and the gradient of the image eigenvalue is used as the growth condition. The pixels of similar properties in the image are merged and generated until the gradient of the image eigenvalue is greater than the preset value. The growth stops. Through the technical solution of the present invention, the paved pavement area on the paved road surface can be accurately determined, so that more accurate edge position information can be obtained, and the paving width can be calculated based on this, and the detection result is more accurate.

[0042] Region growing is the process of growing groups of pixels or regions into larger regions. Starting from a set of seed points, regions are grown from these points by merging neighboring pixels with similar properties to each seed point, such as intensity, grayscale, texture color, etc.

[0043] In any of the above technical solutions, the calculation unit is specifically used to: the point cloud data includes first point cloud data and second point cloud data, and the boundary line is fitted according to the first point cloud data; the average distance from the second point cloud data to the boundary line is calculated, and the average distance is recorded as the paving width; wherein the first point cloud data and the second point cloud data respectively correspond to the first edge position information and the second edge position information of the loose paving material area in the direction perpendicular to the road extension direction.

[0044] In this technical solution, the point cloud data corresponding to the edge position information of the loose paving material area includes first point cloud data and second point cloud data. The first point cloud data and the second point cloud data correspond to the first edge position information and the second edge position information of the loose paving material area in a direction perpendicular to the extension of the road, respectively. When the calculation unit determines the paving width based on the above point cloud data, a boundary line is first fitted based on the first point cloud data, and then the average distance from the second point cloud data to the boundary line is calculated, and the average distance is recorded as the paving width. The calculation method of the present invention is simple and reliable, thereby obtaining a more accurate and reliable paving width, which is conducive to improving the paving quality.

[0045] Specifically, a first boundary line may be fitted based on the first point cloud data, and a second boundary line may be fitted based on the second point cloud data. The average distance between the first boundary line and the second boundary line may be calculated, and the average distance may be recorded as the paving width.

[0046] In any of the above technical solutions, the paving width detection device also includes a calibration unit, which is used to: extract third point cloud data corresponding to the road surface area and fourth point cloud data corresponding to the road boundary area from the depth image before the step of extracting point cloud data corresponding to the edge position information of the loose paving material area from the depth image; calculate the point cloud normal vector based on the third point cloud data and the fourth point cloud data; calculate the external parameters of the vehicle-mounted image acquisition device based on the point cloud normal vector and the reference normal vector to calibrate the depth image captured by the vehicle-mounted image acquisition device.

[0047] In this technical solution, before processing the depth image to extract the point cloud data corresponding to the paving edge position information, the vehicle-mounted image acquisition device needs to be calibrated for external parameters. Specifically, the above-mentioned depth image is obtained by a calibration unit, and the third point cloud data corresponding to the road surface area and the fourth point cloud data corresponding to the road boundary area are extracted therefrom. The point cloud normal vector can be calculated through the third point cloud data and the fourth point cloud data. By calculating the point cloud normal vector and the reference normal vector, the external parameters of the vehicle-mounted image acquisition device can be obtained. Based on this external parameter, the vehicle-mounted image acquisition device can be calibrated. After the calibration is completed, the coordinate system of the vehicle-mounted image acquisition device can be associated and unified with the world coordinate system, that is, in the same coordinate system as the paver and the screed. In this way, the image data obtained by the vehicle-mounted image acquisition device is reliable and consistent with reality. When the image data is reliable, the calculated paving width is more realistic and reliable.

[0048] In addition, it should be noted that, during the above calibration process, the depth image used is not limited to the depth image of the paved road area, but the depth image of the road area to be paved is also applicable.

[0049] In any of the above technical solutions, the paving width detection device also includes a display unit, which is used to: obtain location information of the paved road surface; draw a paving width map based on the location information and paving width of the paved road surface; and send the paving width map to the display device for display.

[0050] In this technical solution, by obtaining the location information of the paved road surface and combining it with the paving width, a paving width map of the sampled paved road surface can be obtained, and the paving width map is sent to the display device for display. This allows a more intuitive view of the paving width of each section of road, which can be used to help judge the paving quality and the measurement and prediction of engineering quantities.

[0051] According to a fourth aspect of the present invention, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the paving width detection method as described in any of the above technical solutions are implemented.

[0052] The readable storage medium provided by the present invention can implement the steps of the paving width detection method of any of the above-mentioned technical solutions when the stored program or instruction is executed. Therefore, it has all the beneficial effects of the above-mentioned paving width detection method, which will not be discussed one by one here.

[0053] In a fifth aspect, the present invention provides a paving machine, comprising: a paving width detection device based on a vehicle-mounted image acquisition device as described in any of the above technical solutions; and / or a readable storage medium as described in the above technical solutions.

[0054] The paver provided by this technical solution includes a paving width detection device based on a vehicle-mounted image acquisition device as in any of the above technical solutions, and / or a readable storage medium as in the above technical solution. Therefore, the paver has all the beneficial effects of the above paving width detection device and / or the above readable storage medium, which will not be repeated.

[0055] Additional aspects and advantages of the invention will become apparent from the description which follows, or may be learned by practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments with reference to the following drawings, in which:

[0057] Figure 1 This is one of the flow charts of a method for detecting paving width based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0058] Figure 2 This is a second flow chart of a method for detecting paving width based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0059] Figure 3 This is a third flow chart of a method for detecting paving width based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0060] Figure 4 This is a fourth flow chart of a method for detecting paving width based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0061] Figure 5 1 is a schematic diagram of a calibration method for detecting paving width based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0062] Figure 6 This is one of the schematic block diagrams of a paving width detection device based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0063] Figure 7 This is a second schematic block diagram of a paving width detection device based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0064] Figure 8a 2 is a schematic structural diagram of a paving width detection device based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0065] Figure 8b is an RGB image of a paved road surface area in a paving width detection method based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0066] Figure 9 This is a fifth flow chart of a method for detecting paving width based on a vehicle-mounted image acquisition device according to an embodiment of the present invention;

[0067] Figure 10 It is a logical schematic diagram of a paving width detection method based on a vehicle-mounted image acquisition device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features therein can be combined with each other without conflict.

[0069] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0070] Refer to the following Figures 1 to 10The present invention describes a paving width detection method (hereinafter referred to as the paving width detection method) and a device, a readable storage medium, and a paving machine according to some embodiments of the present invention.

[0071] Example 1

[0072] Figure 1 FIG1 is a flow chart of a method for detecting paving width according to an embodiment of the present invention. The method for detecting paving width includes:

[0073] Step 102: Acquire image information of the paved road area using a calibrated vehicle-mounted image acquisition device;

[0074] Step 104: extracting an RGB image and a depth image of the paved road area from the image information of the paved road area;

[0075] Step 106, determining edge position information of the loose paving material area in the paved road area based on the RGB image;

[0076] Step 108: extracting point cloud data corresponding to edge position information of the loose paving material area from the depth image;

[0077] Step 110: Determine the paving width based on the point cloud data.

[0078] The paving width detection method provided in this embodiment first obtains image information of the paved road area through a vehicle-mounted image acquisition device, and extracts the RGB image and depth image of the paved road area therefrom. Then, by pre-processing the RGB image of the paved road, the edge position information of the loose paving material area in the paved road area can be located. Point cloud data corresponding to the edge position information of the loose paving material area is extracted from the depth image. The paving width can be calculated based on the above point cloud data. The paving width detection method provided in the embodiment of the present invention obtains the paving width by calculating the image of the paved road area. It is a direct detection method, and is therefore applicable to both straight-plate pavers and telescopic pavers. Compared with the related art of using a distance sensor to measure the width of the ironing board and indirectly measuring the paving width, the detection result is more accurate, which can better help judge the paving quality and measure and predict the engineering quantity.

[0079] Specifically, the vehicle-mounted image acquisition device may be a binocular vision camera, a TOF camera, or a lidar module.

[0080] Example 2

[0081] Figure 2 This is a second flow chart of a method for detecting paving width according to an embodiment of the present invention. The method for detecting paving width includes:

[0082] Step 202: Acquire image information of the paved road area using a calibrated vehicle-mounted image acquisition device;

[0083] Step 204 , extracting an RGB image and a depth image of the paved road area from the image information of the paved road area;

[0084] Step 206, determining the loose material area in the RGB image based on the image feature value;

[0085] Step 208, determining corresponding edge position information according to the loose material area;

[0086] Step 210: extracting point cloud data corresponding to edge position information of the loose paving material area from the depth image;

[0087] Step 212: Determine the paving width based on the point cloud data.

[0088] The image feature value includes any one of the following items or a combination thereof: color value, brightness value, grayscale value, and texture value.

[0089] In this embodiment, when determining edge position information of a loose paving material region within a paved road surface area based on an RGB image of the paved road surface, the image feature values ​​presented by the loose paving material region in the image are significantly different from those presented by a road surface region without loose paving material or a road boundary region (such as a boundary ground or wall). Therefore, the loose paving material region can be determined based on the change in the image feature values. Furthermore, the corresponding edge position information can be located based on the loose paving material region.

[0090] Example 3

[0091] In the above-mentioned embodiment 2, the step of determining the loose material area in the RGB image based on the image eigenvalues ​​specifically includes: taking any point in the loose material area as a seed point, taking the changing gradient of the image eigenvalues ​​as a growth condition, adopting a regional growing algorithm to determine the loose material area, and stopping the growth when the changing gradient of the image eigenvalues ​​is greater than a preset value.

[0092] In this embodiment, when determining the loose paving material area in the RGB image based on the image eigenvalue, the loose paving material area can be extracted from the RGB image using a region growing algorithm. Wherein, the seed point can be selected as any point in the loose paving material area, and the gradient of the image eigenvalue is used as the growth condition. Pixels of similar properties are merged and generated in the image until the gradient of the image eigenvalue is greater than a preset value and the growth stops. Through the technical solution of the present invention, the paved pavement area on the paved road surface can be accurately determined, thereby obtaining more accurate edge position information, thereby calculating the paving width, and the detection result is more accurate.

[0093] Example 4

[0094] Figure 3 This is a third flow chart of a method for detecting paving width according to an embodiment of the present invention. The method for detecting paving width includes:

[0095] Step 302: Acquire image information of the paved road area using a calibrated vehicle-mounted image acquisition device;

[0096] Step 304 , extracting an RGB image and a depth image of the paved road area from the image information of the paved road area;

[0097] Step 306, determining the loose material area in the RGB image based on the image feature value;

[0098] Step 308, determining corresponding edge position information according to the loose material area;

[0099] Step 310: extracting point cloud data corresponding to edge position information of the loose paving material area from the depth image, where the point cloud data includes first point cloud data and second point cloud data;

[0100] Step 312, fitting a boundary line according to the first point cloud data;

[0101] Step 314 : Calculate the average distance between the second point cloud data and the boundary line, and record the average distance as the paving width.

[0102] The image feature value includes any one of a color value, a brightness value, a grayscale value, a texture value, or a combination thereof, but is not limited thereto. The first point cloud data and the second point cloud data correspond to first edge position information and second edge position information of the loose paving material area in a direction perpendicular to the road extension direction, respectively.

[0103] In this embodiment, the point cloud data corresponding to the edge position information of the loose paving material area includes first point cloud data and second point cloud data. The first point cloud data and the second point cloud data correspond to the first edge position information and the second edge position information of the loose paving material area in a direction perpendicular to the extension of the road, respectively. When determining the paving width based on the above point cloud data, a boundary straight line is first fitted based on the first point cloud data, and then the average distance from the second point cloud data to the boundary straight line is calculated, and the average distance is recorded as the paving width. The calculation method of the embodiment of the present invention is simple and reliable, thereby obtaining a more accurate and reliable paving width, which is conducive to improving the paving quality.

[0104] Specifically, a first boundary line may be fitted based on the first point cloud data, and a second boundary line may be fitted based on the second point cloud data. The average distance between the first boundary line and the second boundary line may be calculated, and the average distance may be recorded as the paving width.

[0105] Example 5

[0106] Figure 4 This is a fourth flow chart of a method for detecting paving width according to an embodiment of the present invention. The method for detecting paving width includes:

[0107] Step 402: Acquire image information of the paved road area using a calibrated vehicle-mounted image acquisition device;

[0108] Step 404 , extracting an RGB image and a depth image of the paved road area from the image information of the paved road area;

[0109] Step 406, determining edge position information of the loose paving material area in the paved road area based on the RGB image;

[0110] Step 408 , extracting point cloud data corresponding to edge position information of the loose paving material area from the depth image;

[0111] Step 410, determining the paving width based on the point cloud data;

[0112] Step 412: Obtain location information of the paved road surface; draw a paving width map based on the location information and paving width of the paved road surface;

[0113] Step 414: Send the paving width map to a display device for display.

[0114] In this embodiment, by obtaining the location information of the paved road surface and combining it with the paving width, a paving width map of the sampled paved road surface can be obtained, and the paving width map is sent to a display device for display. This allows a more intuitive view of the paving width of each section of road, which can be used to help determine the paving quality and measure and predict the engineering quantity.

[0115] Example 6

[0116] Figure 5 FIG. 1 is a schematic diagram of a method for calibrating a depth image in a paving width detection method according to an embodiment of the present invention. The paving width detection method includes:

[0117] Step 502: Acquire image information of the road surface area through the vehicle-mounted image acquisition device;

[0118] Step 504: extracting a depth image of the road surface area from the image information of the road surface area;

[0119] Step 506 , extracting third point cloud data corresponding to the road surface area excluding the road boundary area and fourth point cloud data corresponding to the road boundary area from the depth image;

[0120] Step 508: Calculate the point cloud normal vector based on the third point cloud data and the fourth point cloud data;

[0121] Step 510 : Calculate the extrinsic parameters of the vehicle-mounted image acquisition device according to the point cloud normal vector and the reference normal vector to calibrate the depth image acquired by the vehicle-mounted image acquisition device.

[0122] The road surface area is a paved road surface area or a road surface area to be paved. In this embodiment, before obtaining image information of the paved road surface area through a calibrated on-board image acquisition device, the on-board image acquisition device needs to be calibrated for external parameters. Specifically, the image information of the road surface area is obtained through the on-board image acquisition device, and a depth image of the road surface area is extracted therefrom. The third point cloud data corresponding to the area of ​​the road surface area other than the road boundary area and the fourth point cloud data corresponding to the road boundary area are extracted from the depth image. The point cloud normal vector can be calculated through the above-mentioned third point cloud data and fourth point cloud data. By calculating the point cloud normal vector and the reference normal vector, the external parameters of the depth image can be obtained. The depth image can be calibrated based on this external parameter. After the calibration is completed, it can be associated and unified with the world coordinate system, that is, in the same coordinate system as the paver and the screed. The image data obtained in this way is reliable and consistent with reality. When the image data is reliable, the calculated paving width is more realistic and reliable.

[0123] Example 7

[0124] Figure 6 This is one of the schematic block diagrams of a paving width detection device 600 according to an embodiment of the present invention.

[0125] The paving width detection device 600 includes:

[0126] Memory 602 stores programs or instructions;

[0127] Processor 604, when executing a program or instruction, implements the steps of the paving width detection method as in any of the above embodiments.

[0128] The paving width detection device 600 provided in this embodiment includes a memory 602 and a processor 604, wherein the memory 602 stores a program or instruction, and when the processor 604 executes the above program or instruction, it implements the steps of the paving width detection method as in any of the above embodiments. Therefore, the paving width detection device 600 has all the beneficial effects of the paving width detection method, which will not be discussed one by one here.

[0129] Example 8

[0130] Figure 7 This is the second schematic block diagram of the paving width detection device 700 according to an embodiment of the present invention.

[0131] The paving width detection device 700 includes:

[0132] An acquisition unit 702 is configured to acquire image information of a paved road area through a vehicle-mounted image acquisition device;

[0133] An extraction unit 704 is configured to extract an RGB image and a depth image of the paved road area from the image information of the paved road area;

[0134] A first processing unit 706 is configured to determine edge position information of a loose paving material area in a paved road area based on the RGB image;

[0135] A second processing unit 708 is configured to extract point cloud data corresponding to edge position information from the depth image;

[0136] The calculation unit 710 is used to determine the paving width according to the point cloud data.

[0137] The paving width detection device 700 provided in this embodiment includes an acquisition unit 702, an extraction unit 704, a first processing unit 706, a second processing unit 708 and a calculation unit 710. First, the acquisition unit 702 acquires image information of the paved road area through a calibrated vehicle-mounted image acquisition device, and the extraction unit 704 extracts the RGB image and depth image of the paved road area from it. Then, the first processing unit 706 can locate the edge position information of the loose paving material area in the paved road area by pre-processing the RGB image of the paved road. The second processing unit 708 extracts point cloud data corresponding to the edge position information of the loose paving material area from the depth image. The calculation unit 710 can calculate the paving width based on the above point cloud data. The paving width detection device 700 provided by the present invention calculates the paving width by using an image of the paved road area. It is a direct detection method and is therefore applicable to both straight-plate pavers and telescopic pavers. Compared with the related art method of using a distance sensor to measure the width of the ironing board and indirectly measuring the paving width, the detection result is more accurate, which can better help judge the paving quality and measure and predict the project volume.

[0138] Among them, RGB image is an image in RGB color mode (Red Green Blue color mode). RGB color mode is a color standard in the industry. It obtains various colors by changing the three color channels of red, green, and blue and superimposing them on each other.

[0139] A depth image, also known as a range image, stores three-dimensional depth feature information. It is an image captured by an image collector using the distance (depth) values ​​of each point in a scene as pixel values. It directly reflects the geometry of the visible surface of the scene. Depth images can be converted into point cloud data through coordinate transformation.

[0140] Specifically, the vehicle-mounted image acquisition device may be a binocular vision camera or a TOF camera.

[0141] Embodiment 9

[0142] In the above embodiment, the first processing unit 706 is specifically used to determine the loose material area in the RGB image based on the image feature value; determine the corresponding edge position information based on the loose material area; wherein the image feature value includes any one of the following items or a combination thereof: color value, brightness value, grayscale value, texture value.

[0143] In this embodiment, when the first processing unit 706 determines the edge position information of the loose paving material area in the paved road area based on the RGB image of the paved road surface, the image feature values ​​presented by the loose paving material area in the image are significantly different from the image feature values ​​presented by the unpaved road area or the road boundary area (such as the boundary ground or wall surface) in the image. Therefore, the loose paving material area can be determined based on the change in the image feature values. Then, the corresponding edge position information can be located based on the loose paving material area.

[0144] The image feature value includes any one of a color value, a brightness value, a grayscale value, and a texture value, or a combination thereof, but is not limited thereto.

[0145] Example 10

[0146] In any of the above embodiments, the first processing unit 706 is specifically used to use any point in the loose material area as a seed point, the changing gradient of the image eigenvalue as a growth condition, and adopt a regional growing algorithm to determine the loose material area, and stop growing when the changing gradient of the image eigenvalue is greater than a preset value.

[0147] In this embodiment, when the first processing unit 706 determines the loose paving material area in the RGB image according to the image eigenvalue, the region growing algorithm can be adopted to extract the loose paving material area from the RGB image. Wherein, the seed point can be selected as any point in the loose paving material area, with the change gradient of the image eigenvalue as the growth condition, and the pixels of similar properties are merged and generated in the image, until the change gradient of the image eigenvalue is greater than the preset value and stops growing. Through the embodiment of the present invention, the paving road area on the paved road can be accurately determined, thereby more accurate edge position information can be obtained, and the paving width can be calculated with this, and the detection result is more accurate.

[0148] Region growing is the process of growing groups of pixels or regions into larger regions. Starting from a set of seed points, regions are grown from these points by merging neighboring pixels with similar properties to each seed point, such as intensity, grayscale, texture color, etc.

[0149] Example 11

[0150] In any of the above embodiments, the calculation unit 710 is specifically used to: the point cloud data includes first point cloud data and second point cloud data, and the boundary line is fitted according to the first point cloud data; the average distance from the second point cloud data to the boundary line is calculated, and the average distance is recorded as the paving width; wherein the first point cloud data and the second point cloud data respectively correspond to the first edge position information and the second edge position information of the loose paving material area in the direction perpendicular to the road extension direction.

[0151] In this embodiment, the point cloud data corresponding to the edge position information of the loose paving material area includes first point cloud data and second point cloud data. The first point cloud data and the second point cloud data correspond to the first edge position information and the second edge position information of the loose paving material area in a direction perpendicular to the road extension direction, respectively. When the calculation unit 710 determines the paving width based on the above-mentioned point cloud data, a boundary line is first fitted based on the first point cloud data, and then the average distance from the second point cloud data to the boundary line is calculated, and the average distance is recorded as the paving width. The calculation method of the present invention is simple and reliable, thereby obtaining a more accurate and reliable paving width, which is conducive to improving the paving quality.

[0152] Specifically, a first boundary line may be fitted based on the first point cloud data, and a second boundary line may be fitted based on the second point cloud data. The average distance between the first boundary line and the second boundary line may be calculated, and the average distance may be recorded as the paving width.

[0153] Example 12

[0154] In any of the above embodiments, the paving width detection device 700 also includes a calibration unit 712, which is used to: before the step of obtaining image information of the paved road surface area through the calibrated vehicle-mounted image acquisition device, obtain image information of the road surface area through the vehicle-mounted image acquisition device; extract a depth image of the road surface area from the image information of the road surface area; extract third point cloud data corresponding to the road surface area other than the road boundary area and fourth point cloud data corresponding to the road boundary area from the depth image; calculate the point cloud normal vector based on the third point cloud data and the fourth point cloud data; calculate the external parameters of the vehicle-mounted image acquisition device based on the point cloud normal vector and the reference normal vector to calibrate the depth image collected by the vehicle-mounted image acquisition device; wherein the road surface area is a paved road surface area or a road surface area to be paved.

[0155] In this embodiment, before acquiring image information of a paved road surface area using a calibrated on-board image acquisition device, the on-board image acquisition device must undergo extrinsic calibration. Specifically, the on-board image acquisition device acquires image information of the road surface area and extracts a depth image of the road surface area. From this depth image, third point cloud data corresponding to the road surface area excluding the road boundary region and fourth point cloud data corresponding to the road boundary region are extracted. A point cloud normal vector is calculated from the third and fourth point cloud data. By calculating this point cloud normal vector against a reference normal vector, the extrinsic parameters of the on-board image acquisition device are obtained. Based on these extrinsic parameters, the on-board image acquisition device can be calibrated. Only after calibration is the coordinate system of the on-board image acquisition device unified with the world coordinate system, i.e., the same coordinate system as the paver and screed. This ensures that the image data acquired by the on-board image acquisition device is reliable and consistent with reality. When the image data is reliable, the calculated paving width is more realistic and reliable.

[0156] In addition, it should be noted that, during the above calibration process, the depth image used is not limited to the depth image of the paved road area, but the depth image of the road area to be paved is also applicable.

[0157] Example 13

[0158] In any of the above embodiments, the paving width detection device 700 also includes a display unit 714, which is used to: obtain location information of the paved road surface; draw a paving width map based on the location information and paving width of the paved road surface; and send the paving width map to the display device for display.

[0159] In this embodiment, by obtaining the location information of the paved road surface and combining it with the paving width, a paving width map of the sampled paved road surface can be obtained, and the paving width map is sent to a display device for display. This allows a more intuitive view of the paving width of each section of road, which can be used to help determine the paving quality and measure and predict the engineering quantity.

[0160] Example 14

[0161] Since paving width is an important indicator for measuring the quality of pavement paving construction and predicting paving engineering quantities, it is critical to obtain and record paving width in real time.

[0162] In response to the need for detecting the width of a paved road surface during road construction, this embodiment provides a paving width detection method and device based on a vehicle-mounted image acquisition device.

[0163] Figure 8a 2 is a schematic structural diagram of a paving width detection device based on a vehicle-mounted image acquisition device according to this embodiment; Figure 8b is an RGB image of the paved road surface area in the paving width detection method based on the vehicle-mounted image acquisition device of this embodiment; Figure 9 It is a flow chart of the paving width detection method of this embodiment. Figure 10 This is a schematic diagram of the paving width detection method of this embodiment. Figure 8a 、 Figure 8b 、 Figure 9 、 Figure 10 The paving width detection method is explained in detail.

[0164] like Figure 10 As shown, the paving width detection method provided in this embodiment can, on the one hand, determine the paving area (loose material area) based on the RGB image of the paved road surface, and then determine the image boundary line of the paving area. On the other hand, it can also perform external parameter calibration based on the depth image of the paved road surface. Then, based on the calibrated depth image and image boundary line, the point cloud boundary scatter points are determined. The point cloud boundary scatter points include left boundary scatter points (first point cloud data) and right boundary scatter points (second point cloud data). The paving width can be calculated based on these left and right boundary scatter points.

[0165] like Figure 8a As shown, the above-mentioned paving width detection device includes an image acquisition module, a display module, a satellite differential positioning module, an industrial computer, etc.

[0166] Power the image acquisition module and industrial computer, connect the image acquisition module and the industrial computer via a network cable, and test them to ensure data connectivity.

[0167] like Figure 9 As shown, the above-mentioned paving width detection method includes:

[0168] Step 902: Collecting images of the paved road surface by installing a visual camera measurement module on the paver;

[0169] Step 904: While collecting the paved road surface image, the high-precision position information and corresponding satellite timing data obtained by the satellite differential positioning system are transmitted together with the image to the industrial computer on the detection platform mounted on the paver;

[0170] Step 906: pre-process the collected image using an algorithm to locate edge points on both sides of the road surface in the image;

[0171] Step 908: By calculating the size between the two edges and combining it with satellite positioning data, the width of the sampling area can be obtained and stored.

[0172] Step 910: By installing a display module on the paver, the width data combined with the satellite positioning information is transmitted to the display screen for display.

[0173] Among them, the image acquisition module can output RGB images and depth images at the same time, specifically, it can be a binocular camera or a TOF camera or a lidar sensor.

[0174] Specifically, the display module may be an LED display screen.

[0175] Specifically, in step 904, the satellite positioning module is parsed to extract the timestamp and longitude and latitude coordinate data in the GNSS data stream for repackaging with the paving width data.

[0176] Specifically, in step 906, while the paver is paving, the visual camera samples the paving surface behind it. A depth image of the visual camera is acquired and calibrated with external parameters. The data indicates that the loose paved surface is level. During calibration, the vehicle is driven onto a flat surface with the visual camera facing the road. Data on the road boundary and paving material area in the depth image are collected and the point cloud normal vector is calculated. Knowing that the vehicle body is facing the wall and the ground is level, the normal vector between the lower boundary of the screed coordinate system and the ground can be established. At this time, singular value decomposition can be used to obtain the external parameters RT from the visual camera to the screed coordinate system:

[0177]

[0178] V,U=svd(H);

[0179] R=V×U T .

[0180] Here, H is the matrix composed of the four normal vectors, V and U are the singular values ​​calculated using the SVD (Singular Value Decomposition) algorithm, and R is the rotation matrix of the extrinsic parameters. After calculating the rotation matrix R, the translation T is configured based on the installation position.

[0181] Furthermore, the loose material area is extracted from the image returned by the visual sensor. The boundary of the paving area is calculated using region growing, and the seed point is selected as the lower middle area of ​​the image (the loose material area close to the screed). Figure 8b In the example, the area indicated by 804 is the seed point selection area, and the seed area grows to the left, right and top. The growth conditions are color and image brightness. If the color and brightness gradient is too large, the process is terminated. Figure 8b In FIG, the area indicated by 802 is the road boundary area.

[0182] In step 908, the edge position is located according to the image, the edge dense points are extracted from the depth image, and the loose paving width is calculated based on the point clouds on both sides:

[0183] Specifically, the RANdom Sampling Consensus (RANSAC) algorithm is used to fit the left line. The point cloud is P. The fitted interior points are PL, and the line equation is y = kx + b. Thus, PL, k, b = RANSAC(P). The mean of the right point cloud is obtained, and the projected distance from these points to the left line equation is calculated to obtain the current paving width W. W is output to the display module and combined with the GPS location to draw a paving width map.

[0184] Example 15

[0185] This embodiment provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the paving width detection method of any of the above embodiments are implemented.

[0186] The readable storage medium provided in this embodiment can implement the steps of the paving width detection method of any of the above embodiments when the program or instructions stored therein are executed. Therefore, it has all the beneficial effects of the above-mentioned paving width detection method and will not be discussed one by one here.

[0187] Example 16

[0188] This embodiment provides a paving machine, comprising: a paving width detection device based on a vehicle-mounted image acquisition device as in any of the above embodiments; and / or a readable storage medium as in the above embodiments.

[0189] The paver provided in this embodiment includes a paving width detection device based on a vehicle-mounted image acquisition device as in any of the above embodiments, and / or a readable storage medium as in the above embodiments. Therefore, the paver has all the beneficial effects of the above-mentioned paving width detection device and / or the above-mentioned readable storage medium, which will not be repeated.

[0190] In this specification, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance, unless otherwise expressly specified or limited. Terms such as "connect," "install," and "fix" should be interpreted broadly. For example, "connect" can refer to a fixed connection, a detachable connection, or an integral connection; and can be directly connected or indirectly connected through an intermediary. Those skilled in the art will understand the specific meanings of these terms in the present invention based on specific circumstances.

[0191] Throughout this specification, terms such as "one embodiment," "some embodiments," and "specific embodiments" mean that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0192] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A paving width detection method based on a vehicle-mounted image acquisition device, characterized in that: include: Acquire image information of the paved road area through a calibrated vehicle-mounted image acquisition device; Extracting an RGB image and a depth image of the paved road area from the image information of the paved road area; determining edge position information of a loose paved material area in a paved road area based on the RGB image; Extracting point cloud data corresponding to edge position information of the loose paving material area from the depth image; determining a paving width according to the point cloud data; The step of determining the paving width according to the point cloud data specifically includes: The point cloud data includes first point cloud data and second point cloud data, wherein the first point cloud data and the second point cloud data respectively correspond to first edge position information and second edge position information of the loose paving material area in a direction perpendicular to the road extension direction; Fitting a boundary line according to the first point cloud data; Calculating an average distance from the second point cloud data to the boundary line, and recording the average distance as the paving width; Before acquiring image information of the paved road area through the calibrated vehicle-mounted image acquisition device, the vehicle-mounted image acquisition device needs to be calibrated with external parameters; Before the step of acquiring image information of the paved road area by using the calibrated vehicle-mounted image acquisition device, the method further includes: Acquiring image information of the road surface area through the vehicle-mounted image acquisition device; extracting a depth image of the road surface area from the image information of the road surface area; Extracting third point cloud data corresponding to an area of ​​the road surface excluding a road boundary area and fourth point cloud data corresponding to the road boundary area from the depth image; Calculating a point cloud normal vector according to the third point cloud data and the fourth point cloud data; Calculating the extrinsic parameters of the vehicle-mounted image acquisition device according to the point cloud normal vector and the reference normal vector to calibrate the vehicle-mounted image acquisition device; The road surface area is a paved road surface area or a road surface area to be paved.

2. The paving width detection method according to claim 1, characterized in that: The step of determining edge position information of a loose paved material area in a paved road area based on the RGB image specifically includes: Determining a loose material area in the RGB image according to the image feature value; Determining corresponding edge position information according to the loose paving material area; The image feature value includes any one of the following items or a combination thereof: color value, brightness value, grayscale value, and texture value.

3. The paving width detection method according to claim 2, characterized in that: The step of determining the loose material area in the RGB image according to the image feature value specifically includes: Taking any point in the loose material area as a seed point and the changing gradient of the image eigenvalue as a growth condition, a regional growing algorithm is used to determine the loose material area, and growth is stopped when the changing gradient of the image eigenvalue is greater than a preset value.

4. The paving width detection method according to any one of claims 1 to 3, characterized in that: Also includes: Obtaining location information of the paved road surface; Drawing a paving width map based on the location information of the paved road surface and the paving width; The paving width map is sent to a display device for display.

5. A paving width detection device based on a vehicle-mounted image acquisition device, characterized in that: include: Memory, storing programs or instructions; A processor, which implements the steps of the paving width detection method based on a vehicle-mounted image acquisition device as described in any one of claims 1 to 4 when executing the program or instruction.

6. A paving width detection device based on a vehicle-mounted image acquisition device, characterized in that: include: an acquisition unit, configured to acquire image information of the paved road area through a calibrated vehicle-mounted image acquisition device; an extraction unit, configured to extract an RGB image and a depth image of the paved road area from the image information of the paved road area; a first processing unit, configured to determine edge position information of a loose paving material area in a paved road surface area based on the RGB image; a second processing unit, configured to extract point cloud data corresponding to the edge position information from the depth image; A calculation unit for determining a paving width based on the point cloud data; The point cloud data includes first point cloud data and second point cloud data, wherein the first point cloud data and the second point cloud data respectively correspond to first edge position information and second edge position information of the loose paving material area in a direction perpendicular to the road extension direction; The computing unit is further configured to fit a boundary line according to the first point cloud data; Calculating an average distance from the second point cloud data to the boundary line, and recording the average distance as the paving width; Before acquiring image information of the paved road area through the calibrated vehicle-mounted image acquisition device, the vehicle-mounted image acquisition device needs to be calibrated with external parameters; Before the step of acquiring image information of the paved road area by using the calibrated vehicle-mounted image acquisition device, the method further includes: Acquiring image information of the road surface area through the vehicle-mounted image acquisition device; extracting a depth image of the road surface area from the image information of the road surface area; Extracting third point cloud data corresponding to an area of ​​the road surface excluding a road boundary area and fourth point cloud data corresponding to the road boundary area from the depth image; Calculating a point cloud normal vector according to the third point cloud data and the fourth point cloud data; Calculating the extrinsic parameters of the vehicle-mounted image acquisition device according to the point cloud normal vector and the reference normal vector to calibrate the vehicle-mounted image acquisition device; The road surface area is a paved road surface area or a road surface area to be paved.

7. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by the processor, the steps of the paving width detection method based on the vehicle-mounted image acquisition device as claimed in any one of claims 1 to 4 are implemented.

8. A paver, characterized in that: include: The paving width detection device based on the vehicle-mounted image acquisition device according to claim 5 or 6; and / or The readable storage medium according to claim 7.

Images