A road line identification mobile measurement method and system

Abstract

The present invention discloses a mobile measurement method and system for road line identification, and relates to the technical field of road line measurement. The method comprises: a measuring vehicle is equipped with measuring equipment, and road data is collected along a predetermined measurement route, and the collected data is transmitted to a cloud platform; the cloud platform merges the data collected by different measuring equipment to obtain a comprehensive road data set; the cloud platform reads the image data of the road line at each collection time point from the comprehensive road data set, and uses a continuity detection function to identify the road section where the road line is missing from the image data of the road line; the cloud platform completes the road line based on the road line prediction model and the comprehensive data of the road line missing section, as well as the known road line image data before and after the road line missing section; the cloud platform exports a complete road line file of the predetermined measurement route according to the selected output format. The present invention can adapt to the measurement and identification of road lines in complex and changeable road environments.

Description

Technical Field

[0001] The present invention relates to the technical field of road line measurement, and in particular to a road line identification mobile measurement method and system. Background Art

[0002] As one of the key technologies in the fields of autonomous driving and intelligent transportation, road line recognition technology has made significant progress in recent years. These methods are mainly based on the fusion of multiple sensor technologies such as machine vision, laser radar (LiDAR), inertial navigation system (INS) and global positioning system (GPS). Although the mobile measurement method of road line recognition has made significant progress, there are still some defects and challenges: in complex and changing road environments, such as curves, ramps, intersections, etc., the characteristics of road lines may become blurred or unclear, resulting in inaccurate road line recognition. In addition, the wear and pollution of road markings may also affect the recognition effect. In order to solve the above problems, researchers in this field urgently need to develop a method that can still measure and identify road lines with high precision under complex or obstructed road conditions. Summary of the invention

[0003] The present invention provides a road line identification mobile measurement method, comprising:

[0004] Step 1: The measuring vehicle is equipped with measuring equipment to collect road data along the predetermined measuring route;

[0005] Step 2: Merge the data collected by different measurement devices to obtain a comprehensive road data set;

[0006] Step 3, read the image data of the road line at each acquisition time point from the comprehensive road data set, and use the continuity detection function to identify the road section where the road line is missing from the image data of the road line;

[0007] Step 4, completing the road line based on the comprehensive data of the road line prediction model and the missing road line section, as well as the known road line image data before and after the missing road line section;

[0008] Step 5. Export the complete road line file of the planned measurement route according to the selected output format.

[0009] The mobile measurement method for road line identification as described above, wherein the data collected by different measurement devices are merged, specifically comprises the following sub-steps:

[0010] Serialize the collected data according to the collection time;

[0011] Unify the coordinate system of each collected data at each collection time point;

[0012] Remove the redundant parts of each collected data, merge the collected data at the same collection time point, and organize them into a comprehensive road data set.

[0013] The method for mobile measurement of road line identification as described above, wherein a continuity detection function is used to identify a road section where a road line is missing from image data of the road line, specifically comprises the following sub-steps:

[0014] Extracting a set of contour points of a road line from the image data;

[0015] Use the continuity detection function to detect whether the collected road lines are continuous based on the extracted contour point set;

[0016] The starting coordinate value and the ending coordinate value of the missing section of the road line are determined based on the continuity detection result.

[0017] A road line recognition mobile measurement method as described above, wherein the road line is completed based on the comprehensive data of the road line prediction model and the missing road line section, and the known road line image data before and after the missing road line section, specifically comprising the following sub-steps:

[0018] Extracting comprehensive data of missing road segments from the comprehensive road data set, and extracting a sequence of coordinate points of a known route from known road line image data before and after the missing road segments;

[0019] The extracted comprehensive data and the coordinate point sequence of the known route are input into the road line prediction model, and the coordinate point sequence of the missing road line is output;

[0020] Use the output road line coordinate sequence to complete the road line.

[0021] The above-mentioned method for mobile measurement of road line recognition, wherein the output road line coordinate sequence is used to complete the road line, specifically comprises the following sub-steps:

[0022] Add a layer to the image of missing road segments in the comprehensive road dataset;

[0023] Add the output road line coordinate sequence as the inner contour line of the road line to a new layer;

[0024] Fill in the width of the known route to form a complete road line image.

[0025] The present invention also provides a road line identification mobile measurement system, comprising: a road measurement data acquisition module, a road measurement data fusion module, a road line missing section identification module, a road line missing section completion module, and a road line file export module;

[0026] A road measurement data collection module is used to collect road data along a predetermined measurement route using a measurement vehicle;

[0027] The road measurement data fusion module is used to merge the data collected by different measurement devices to obtain a comprehensive road data set;

[0028] A road line missing section identification module is used to read the image data of the road line at each acquisition time point from the comprehensive road data set, and use a continuity detection function to identify the road section with missing road lines from the image data of the road line;

[0029] A road line missing section completion module is used to complete the road line using the road line prediction model and the comprehensive data of the road line missing section, as well as the known road line image data before and after the road line missing section;

[0030] The road line file export module is used to export the complete road line file of the predetermined measurement route according to the selected output format.

[0031] The beneficial effects achieved by the present invention are as follows: it can adapt to the measurement and identification of road lines in complex and changeable road environments, provide rapid identification technology, and realize a working mode of measuring and identifying at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0033] Figure 1 This is a flow chart of a road line identification mobile measurement method provided in Example 1 of the present application;

[0034] Figure 2 This is a schematic diagram of a road line recognition mobile measurement system provided in Example 2 of the present application. DETAILED DESCRIPTION

[0035] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0036] Embodiment 1

[0037] like Figure 1As shown, the first embodiment of the present application provides a road line recognition mobile measurement method, including:

[0038] Step S10: A plurality of measuring devices are mounted on the measuring vehicle to collect road data along a predetermined measuring route, and the measuring vehicle transmits the road data collected by different measuring devices at different collection time points to the cloud platform;

[0039] The measuring equipment on board includes: GPS receiver, lidar, camera, level and inertial navigation system. The GPS receiver is used to obtain its current world coordinates, the lidar is used to collect three-dimensional point cloud data of the road, the camera is used to collect images of the road, the level is used to collect elevation data of the road, and the inertial navigation system is used in conjunction with the GPS receiver to obtain more accurate positioning data.

[0040] After calibrating the measuring equipment, install it on the measuring vehicle. Collect road data on the predetermined measuring route as the vehicle travels. The collection frequency needs to match the driving speed of the measuring vehicle to ensure the accuracy of road data collection. The conversion formula between the collection frequency and the driving speed of the measuring vehicle is expressed as:

[0041] Where T represents the acquisition frequency of the measuring device, Indicates the current speed of the measuring vehicle. represents the conversion ratio between the driving speed unit and the measuring range unit of the i-th measuring device, is an adjustable parameter, is the measurement range difference of the i-th measuring device, is the measurement speed of the i-th measurement device, i ranges from 1 to n, and n is the total number of measurement devices on board.

[0042] The measuring vehicle collects road data along a predetermined measuring route, and transmits the collected road data to a cloud platform. In the embodiment of the present application, the measuring vehicle performs mobile measurement of road lines, and a cloud platform with strong data processing capabilities performs road line identification.

[0043] Step S20: The cloud platform merges the road data collected by different measuring devices at different collection time points to obtain a comprehensive road data set;

[0044] Since the road data collected by different measurement equipment are relatively independent, in order to comprehensively analyze and use these data, they must be synchronized in time and space and merged into a comprehensive road data set. Specifically:

[0045] Step S21: serialize the collected data according to the collection time;

[0046] Data is collected synchronously in time to ensure that all collected data at the same collection time point originate from the same measurement point, that is, all collected data at each collection time point only describes the characteristics of a section of the road line.

[0047] Step S22: unify the coordinate system of each collected data at each collection time point;

[0048] Different measuring devices use different coordinate systems. Here, the coordinate system of the GPS receiver needs to be used as the standard. The coordinate systems of other measuring devices should be translated, rotated, scaled, converted, etc. into the coordinate system of the GPS receiver for subsequent spatial analysis and comparison.

[0049] Step S23: removing redundant parts of various collected data, merging various collected data at the same collection time point, and arranging them into a comprehensive road data set;

[0050] In the actual data collection process, in order to ensure the integrity of the collected data, the collection frequency will be set higher, which will result in redundant data collected each time. After removing these redundant parts, what remains is the complete road line data on the predetermined measurement route. The coordinate system of each collected data has been unified in the previous step. In this step, it is only necessary to remove the collected data with the same coordinates between every two adjacent collection time points.

[0051] After removing the redundant parts, the collected data at the same collection time point are merged into a sub-dataset, and then these sub-datasets are stored in the comprehensive road dataset in sequence. The comprehensive road dataset P is expressed as: ,in represents the coordinate value obtained by the GPS receiver at the jth acquisition time point, It represents the sub-dataset formed by merging the collected data at the j-th collection time point, and the sub-dataset includes the image data of the road line. j ranges from 1 to m, and m is the total number of collection time points in the entire predetermined measurement route.

[0052] Step S30: the cloud platform reads the image data of the road line at each acquisition time point from the comprehensive road data set, and uses a continuity detection function to identify the missing road section from the image data of the road line;

[0053] Under complex and changeable measurement road conditions, the road line images collected by the camera may be blocked, blurred, worn, etc., resulting in the loss of road line recognition results. Therefore, in this step, it is necessary to identify these road sections with missing road lines. Specifically:

[0054] Step S31: extracting edge features, color features, and texture features of road lines from the image data and arranging them into an image feature set;

[0055] Use edge detection algorithms (such as Canny edge detector) to extract edge features of road lines in the image, use color space conversion methods to extract color features of road lines, use gray-level co-occurrence matrix to extract texture features of road lines, and organize these features to form an image feature set D.

[0056] Step S32: using a continuity detection function to detect whether the collected road lines are continuous according to the extracted image feature set;

[0057] Each time, the image feature set (i.e., the set of edge features, color features, and texture features) of two adjacent acquisition time points is input into the continuity detection function. The return value is used to determine whether the road lines collected at these two acquisition time points are continuous. The judgment result is the detection result of the road line continuity. The continuity detection function is expressed as:

[0058] in , Represents the image feature set of the road line collected at two adjacent collection time points (j and j+1 time points), , , Respectively represent the support of road line edge, color, and texture for continuity, Represents the image feature set The edge feature point set within Represents the image feature set The edge feature point set within Represents the distance between edge feature point a and edge feature point b, It represents the distance between edge feature point b and edge feature point a, sup and inf are supremum and infimum respectively. is the color feature difference calculation function, , R, G, and B are color features, representing the values ​​of the road line image in the red channel, green channel, and blue channel, respectively. , , is the image feature set The color characteristics in , , is the image feature set The color characteristics in represents the feature weight on the e-th row and u-th column in the texture matrix, Represented in the image feature set The eigenvalue of the texture matrix in row e and column u is Represented in the image feature set The eigenvalue of the e-th row and u-th column of the texture matrix, e ranges from 1 to E, E is the total number of rows in the texture matrix, u ranges from 1 to U, U is the total number of columns in the texture matrix, and the texture matrix refers to the texture features of the road line extracted using the gray-level co-occurrence matrix.

[0059] judge Is it greater than , is the detection threshold. If it is greater than, the detection result is 0, indicating that the road lines collected at the current two adjacent collection time points (j and j+1) are not continuous. Otherwise, the detection result is 1, indicating that the road lines collected at the current two adjacent collection time points are continuous.

[0060] Step S33: determining the starting coordinate value and the ending coordinate value of the missing section of the road line based on the detection result of the continuity detection function;

[0061] The coordinate values ​​obtained by the GPS receiver at two adjacent acquisition time points with a detection result of 0, that is, and As the starting coordinate value and ending coordinate value of the missing section of the road line, respectively. If the detection results are continuously 0, the coordinate values ​​obtained by the GPS receiver at the first and last collection time points in the continuous sequence are taken as the starting coordinate value and ending coordinate value of the missing section of the road line, respectively. For example: , then take is the starting coordinate value, as the ending coordinate value.

[0062] Step S40: The cloud platform completes the road line based on the road line prediction model and the comprehensive data of the road line missing section, as well as the known road line image data before and after the road line missing section, to obtain complete road line data within the predetermined measurement route.

[0063] The road line prediction model can predict the missing road line coordinates based on the known route coordinates before and after the missing road section, as well as the elevation data and radar data of the missing road section. Specifically:

[0064] Step S41: extracting comprehensive data of the road line missing section in the comprehensive road data set, and extracting a coordinate point sequence of a known route from known road line image data before and after the road line missing section;

[0065] According to the starting coordinate value and ending coordinate value of the missing section of the road line, the corresponding comprehensive data is extracted from the comprehensive road data set, and then the coordinate data of the known route in the previous cycle of the starting coordinate value and the next cycle of the ending coordinate value are extracted. Here, the coordinate point sequence on the innermost contour line of the relevant road line in the image captured by the camera device is taken. The extraction period of the known route is twice the prediction period. The prediction period refers to the number of collection time points experienced in the missing section of the road line.

[0066] Step S42: input the extracted comprehensive data and the coordinate point sequence of the known route into the road line prediction model, and output the coordinate point sequence of the missing road line;

[0067] The road line prediction model is expressed as: ,in is the predicted value of the road line, is the characteristic factor of road curvature, is the characteristic factor of road slope, is the curvature measurement value of the road section where the prediction point is located, is the slope measurement value of the road section where the prediction point is located, is the i-th known road line coordinate value, i ranges from 1 to n, and n is the number of known route coordinate points involved in the prediction. It is an adjustable parameter used to adjust the predicted distance. q is the power of the predicted distance. The road line prediction model predicts the coordinates of the road line within a certain distance based on the changing trend of the road line and the physical characteristics of the road section itself. By continuously adjusting the predicted distance Get every adjustment The coordinate sequence of the missing road line corresponding to the time until the prediction cycle ends. It should be noted that in the road line prediction model, the coordinate value will be used as a vector to participate in the calculation; the prediction distance is automatically adjusted, and the adjustment step is the same as the distance between the known route coordinate sequence.

[0068] Step S43: using the output road line coordinate sequence to complete the road line;

[0069] A new layer is added to the image of the missing road section in the comprehensive road dataset, and the road line coordinate sequence output by the road line prediction model is added to the new layer as the contour lines inside the supplementary road lines. Then, the road lines are supplemented according to the width of the original road lines to form a complete road line image.

[0070] After obtaining the complete road line image, the complete road line file of the predetermined measurement line can be exported according to the selected output format;

[0071] The export formats include three-dimensional road lines and planar road line images. The method for exporting planar road images is: read the road line image data collected by the camera device from the comprehensive road data set; use image recognition technology to extract complete road line pixels in the road line image data; mark the coordinate values, curvature, slope and other data on the road line pixels as needed; and export the road line pixels.

[0072] The method for exporting three-dimensional road lines is: fusing the radar data and elevation data in the comprehensive road dataset to obtain a more complete three-dimensional image of the road and its surrounding environment; adding the road image to the three-dimensional image by using a texture; and exporting the three-dimensional image.

[0073] Embodiment 2

[0074] like Figure 2 As shown, the second embodiment of the present application provides a road line identification mobile measurement system, including: a measurement vehicle 1 and a cloud platform 2, the measurement vehicle 1 is used to collect road data along a predetermined measurement route, and transmit the road data collected by different measurement devices at different collection time points to the cloud platform; the cloud platform is provided with a road measurement data fusion module 21, a road line missing section identification module 22, and a road line missing section completion module 23;

[0075] The measuring equipment carried by the measuring vehicle 1 includes: a GPS receiver, a laser radar, a camera, a level and an inertial navigation system. The GPS receiver is used to obtain its own current world coordinates, the laser radar is used to collect three-dimensional point cloud data of the road, the camera is used to collect images of the road, the level is used to collect elevation data of the road, and the inertial navigation system is used to cooperate with the GPS receiver to obtain more accurate positioning data.

[0076] After calibrating the measuring equipment, install it on the measuring vehicle. Collect road data on the predetermined measuring route as the vehicle travels. The collection frequency needs to match the driving speed of the measuring vehicle to ensure the accuracy of road data collection. The conversion formula between the collection frequency and the driving speed of the measuring vehicle is expressed as:

[0077] Where T represents the acquisition frequency of the measuring device, Indicates the current speed of the measuring vehicle. represents the conversion ratio between the driving speed unit and the measuring range unit of the i-th measuring device, is an adjustable parameter, is the measurement range of the ith measuring device, is the measurement speed of the i-th measurement device, i ranges from 1 to n, and n is the total number of measurement devices on board.

[0078] The road measurement data fusion module 21 is used to merge the data collected by different measurement devices to obtain a comprehensive road data set; it includes: a data serialization submodule, a coordinate system submodule, and a data merging submodule;

[0079] The data serialization submodule is used to serialize the collected data according to the collection time;

[0080] Data is collected synchronously in time to ensure that all collected data at the same collection time point originate from the same measurement point, that is, all collected data at each collection time point only describes the characteristics of a section of the road line.

[0081] The coordinate system submodule is used to unify the coordinate system of each collected data at each collection time point;

[0082] Different measuring devices use different coordinate systems. Here, the coordinate system of the GPS receiver needs to be used as the standard. The coordinate systems of other measuring devices should be translated, rotated, scaled, converted, etc. into the coordinate system of the GPS receiver for subsequent spatial analysis and comparison.

[0083] The data merging submodule is used to remove the redundant parts of each collected data, merge the collected data at the same collection time point, and organize them into a comprehensive road data set;

[0084] In the actual data collection process, in order to ensure the integrity of the collected data, the collection frequency will be set higher, which will result in redundant data collected each time. After removing these redundant parts, what remains is the complete road line data on the predetermined measurement route. The coordinate system of each collected data has been unified in the previous step. In this step, it is only necessary to remove the collected data with the same coordinates between every two adjacent collection time points.

[0085] After removing the redundant parts, the collected data at the same collection time point are merged into a sub-dataset, and then these sub-datasets are stored in the comprehensive road dataset in sequence. The comprehensive road dataset P is expressed as: ,in represents the coordinate value obtained by the GPS receiver at the jth acquisition time point, It represents the sub-dataset formed by merging the collected data at the jth collection time point. j ranges from 1 to m, and m is the total number of collection time points in the entire predetermined measurement line.

[0086] The road line missing section identification module 22 is used to read the image data of the road line at each acquisition time point from the comprehensive road data set, and use the continuity detection function to identify the road section where the road line is missing from the image data of the road line; it includes: an image data reading submodule, an image feature set extraction submodule, a continuity detection submodule, and a missing section positioning submodule;

[0087] The image data reading submodule is used to read the image data of the road line at each acquisition time point from the comprehensive road data set.

[0088] The image feature set extraction submodule extracts the edge features, color features, and texture features of the road line from the image data and organizes them into an image feature set;

[0089] Use edge detection algorithms (such as Canny edge detector) to extract edge features of road lines in the image, use color space conversion methods to extract color features of road lines, use gray-level co-occurrence matrix to extract texture features of road lines, and organize these features to form an image feature set D.

[0090] The continuity detection submodule is used to detect whether the collected road lines are continuous based on the extracted image feature set;

[0091] Each time, the image feature sets of two adjacent acquisition time points are input into the continuity detection function. The return value is used to determine whether the road lines collected at these two acquisition time points are continuous. The judgment result is the detection result of the road line continuity. The continuity detection function is expressed as:

[0092] in , They represent the image feature sets of the road lines collected at two adjacent collection time points (j and j+1 time points), , , Respectively represent the support of road line edge, color, and texture for road line continuity. Represents the image feature set The edge feature point set within Represents the image feature set The edge feature point set within Represents the distance between edge feature point a and edge feature point b, It represents the distance between edge feature point b and edge feature point a, sup and inf are supremum and infimum respectively. is the color feature difference calculation function, , R, G, and B are color features, representing the values ​​of the road line image in the red channel, green channel, and blue channel, respectively. , , is the image feature set The color characteristics in , , is the image feature set The color characteristics in represents the feature weight on the e-th row and u-th column in the texture matrix, Represented in the image feature set The eigenvalue of the texture matrix in row e and column u is Represented in the image feature set The eigenvalue of the e-th row and u-th column of the texture matrix, e ranges from 1 to E, E is the total number of rows in the texture matrix, u ranges from 1 to U, U is the total number of columns in the texture matrix, and the texture matrix refers to the texture features of the road line extracted using the gray-level co-occurrence matrix.

[0093] judge Is it greater than , is the detection threshold. If it is greater than, the detection result is 0, indicating that the road lines collected at the current two adjacent collection time points (j and j+1) are not continuous. Otherwise, the detection result is 1, indicating that the road lines collected at the current two adjacent collection time points are continuous.

[0094] The missing section positioning submodule is used to determine the starting coordinate value and the ending coordinate value of the missing section of the road line according to the continuity detection result;

[0095] The coordinate values ​​obtained by the GPS receiver at two adjacent acquisition time points with a detection result of 0, that is, and As the starting coordinate value and ending coordinate value of the missing section of the road line, respectively. If the detection results are continuously 0, the coordinate values ​​obtained by the GPS receiver at the first and last collection time points in the continuous sequence are taken as the starting coordinate value and ending coordinate value of the missing section of the road line, respectively. For example: , then take is the starting coordinate value, as the ending coordinate value.

[0096] A road line missing section completion module 23 is used to complete the road line using the road line prediction model and the comprehensive data of the road line missing section, as well as the known road line image data before and after the road line missing section; it includes: a comprehensive data extraction submodule, a coordinate sequence output submodule, and a road line completion submodule;

[0097] A comprehensive data extraction submodule is used to extract comprehensive data of missing sections of road lines from the comprehensive road data set;

[0098] According to the starting coordinate value and ending coordinate value of the missing section of the road line, the corresponding comprehensive data is extracted from the comprehensive road data set, and then the coordinate data of the known route in the previous cycle of the starting coordinate value and the next cycle of the ending coordinate value are extracted. Here, the point coordinate sequence on the innermost contour line of the relevant road line in the image captured by the camera device is taken. The extraction period of the known route is twice the prediction period. The prediction period refers to the number of collection time points experienced in the missing section of the road line.

[0099] A coordinate sequence output submodule is used to input the extracted comprehensive data into the road line prediction model and output the coordinate sequence of the missing road line;

[0100] The road line prediction model predicts the coordinates of the points where the road line falls within a certain distance based on the changing trend of the road line and the physical characteristics of the road section itself. The coordinate sequence of the missing road line can be obtained by continuously adjusting the prediction distance until the prediction cycle ends. The road line prediction model is expressed as: ,in is the predicted value of the road line, is the characteristic factor of road curvature, is the characteristic factor of road slope, is the curvature measurement value of the road section where the prediction point is located, is the slope measurement value of the road section where the prediction point is located, is the i-th known road line coordinate value, i ranges from 1 to n, and n is the number of known route coordinate points involved in the prediction. It is an adjustable parameter used to adjust the prediction distance, and q is the power of the prediction distance.

[0101] It should be noted that in the road line prediction model, the coordinate value will be used as a vector to participate in the calculation; the prediction distance is automatically adjusted, and the adjustment step is the same as the distance between the known route coordinate sequence.

[0102] A road line completion submodule, used to complete the road line using the output road line coordinate sequence;

[0103] A new layer is added to the image of the missing road section in the comprehensive road dataset, and the road line coordinate sequence output by the road line prediction model is added to the new layer as the contour lines inside the supplementary road lines. Then, the road lines are supplemented according to the width of the original road lines to form a complete road line image.

[0104] Finally, export the complete road line file of the predetermined measurement route according to the selected output format;

[0105] The output formats include three-dimensional road lines and planar road line images. The method for exporting planar road images is: read the road line image data collected by the camera device from the comprehensive road data set; use image recognition technology to extract complete road line pixels in the road line image data; mark the coordinate values, curvature, slope and other data on the road line pixels as needed; and export the road line pixels.

[0106] The method for exporting three-dimensional road lines is: fusing the radar data and elevation data in the comprehensive road dataset to obtain a more complete three-dimensional image of the road and its surrounding environment; adding the road image to the three-dimensional image by using a texture; and exporting the three-dimensional image.

[0107] Corresponding to the above embodiment, an embodiment of the present invention provides a computer storage medium, including: at least one memory and at least one processor;

[0108] The memory is used to store one or more program instructions;

[0109] The processor is used for running one or more program instructions to execute a road line recognition mobile measurement method.

[0110] Corresponding to the above embodiment, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer storage medium includes one or more program instructions, and the one or more program instructions are used for a processor to execute a road line recognition mobile measurement method.

[0111] The embodiment disclosed in the present invention provides a computer-readable storage medium, in which computer program instructions are stored. When the computer program instructions are executed on a computer, the computer executes the above-mentioned road line identification and mobile measurement method.

[0112] In the embodiment of the present invention, the processor may be an integrated circuit chip having the ability to process signals. The processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0113] The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor for execution, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The processor reads the information in the storage medium and completes the steps of the above method in combination with its hardware.

[0114] The storage medium may be a memory, which may be, for example, a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.

[0115] Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.

[0116] The volatile memory may be a random access memory (RAM) which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDRSDRAM), enhanced synchronous DRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus DRAM (DRRAM).

[0117] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0118] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the present invention can be implemented using a combination of hardware and software. When software is used, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any media that facilitates the transmission of computer programs from one place to another. Storage media can be any available media that can be accessed by general or special-purpose computers.

[0119] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A road line recognition mobile measurement method, It is characterized in that include: Step 1: Multiple measuring devices are loaded on the measuring vehicle to collect road data along the predetermined measuring route. The measuring vehicle transmits the road data collected by different measuring devices at different collection time points to the cloud platform; Step 2: The cloud platform merges the road data collected by different measurement devices at different collection time points to obtain a comprehensive road data set; Step 3: The cloud platform reads the image data of the road line at each acquisition time point from the comprehensive road data set, and uses the continuity detection function to identify the missing sections of the road line from the image data of the road line; Step 4: The cloud platform completes the road line based on the road line prediction model and the comprehensive data of the missing road line section, as well as the known road line image data before and after the missing road line section, to obtain the complete road line data within the predetermined measurement route; Merging the data collected by different measurement devices includes the following sub-steps: Serialize the collected data according to the collection time point; Unify the coordinate system of each collected data at each collection time point; Remove the redundant parts of each collected data, merge the collected data at the same collection time point, and organize them into a comprehensive road data set; Using a continuity detection function to identify road segments with missing road lines from the image data of road lines, specifically including the following sub-steps: Extract edge features, color features, and texture features of the road line from the image data of the road line, and organize them into an image feature set; Use the continuity detection function to detect whether the collected road lines are continuous based on the extracted image feature set; Determine the starting coordinate value and the ending coordinate value of the missing section of the road line based on the continuity detection result; The road line is completed based on the comprehensive data of the road line prediction model and the missing road line section, as well as the known road line image data before and after the missing road line section, specifically including the following sub-steps: Extracting comprehensive data of missing road segments from the comprehensive road data set, and extracting a sequence of coordinate points of a known route from known road line image data before and after the missing road segments; The extracted comprehensive data and the coordinate point sequence of the known route are input into the road line prediction model, and the coordinate point sequence of the missing road line is output; Use the output coordinate point sequence of the missing road lines to complete the road lines.

2. A road line recognition mobile measurement method according to claim 1, It is characterized in that The output missing road line coordinate sequence is used to complete the road line, which specifically includes the following sub-steps: Add a layer to the image of missing road segments in the comprehensive road dataset; Add the output missing road line coordinate sequence as the inner contour line of the road line to a new layer; Fill in the width of the known route to form a complete road line image.

3. A road line recognition mobile measurement system, It is characterized in that include: The measuring vehicle and the cloud platform are equipped with a plurality of measuring devices for collecting road data along a predetermined measuring route, and transmitting the road data collected by different measuring devices at different collecting time points to the cloud platform; the cloud platform is provided with a road measuring data fusion module, a road line missing section identification module and a road line missing section completion module; Road measurement data acquisition module, used to obtain road data uploaded by the measurement vehicle; The road measurement data fusion module is used to merge the road data uploaded by the measurement vehicle to obtain a comprehensive road data set; A road line missing section identification module is used to read the image data of the road line at each acquisition time point from the comprehensive road data set, and identify the road line missing section from the image data of the road line using a continuity detection function; A road line missing section completion module is used to complete the road line based on the comprehensive data of the road line prediction model and the road line missing section, as well as the known road line image data before and after the road line missing section, so as to obtain the complete road line data within the predetermined measurement route; The road measurement data fusion module includes: The data serialization submodule is used to serialize the collected data according to the collection time; The coordinate system submodule is used to unify the coordinate system of each collected data at each collection time point; The data merging submodule is used to remove the redundant parts of each collected data, merge the collected data at the same collection time point, and organize them into a comprehensive road data set; The road line missing section recognition module includes: The image data reading submodule is used to read the image data of the road line at each acquisition time point from the comprehensive road data set; The image feature set extraction submodule extracts edge features, color features, and texture features of road lines from the image data of road lines and organizes them into an image feature set; The continuity detection submodule is used to detect whether the collected road lines are continuous based on the extracted image feature set; The missing section positioning submodule is used to determine the starting coordinate value and the ending coordinate value of the missing section of the road line according to the continuity detection result; The missing road segment completion module includes: A comprehensive data extraction submodule is used to extract comprehensive data of missing sections of road lines from the comprehensive road data set; A coordinate sequence output submodule is used to input the extracted comprehensive data into the road line prediction model and output the coordinate sequence of the missing road line; The road line completion submodule is used to complete the road lines using the output coordinate sequence of the missing road lines.

4. A road line recognition mobile measurement system according to claim 3, It is characterized in that The road line completion submodule is specifically used to: add a new layer to the image of the missing road line section in the comprehensive road dataset; add the output missing road line coordinate sequence as the contour line inside the road line to the new layer; fill in according to the width of the known route to form a complete road line image.

Images