CN118279848A - Lane line vector information acquisition method, device, computer equipment and storage medium - Google Patents
Lane line vector information acquisition method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN118279848A CN118279848A CN202211743725.0A CN202211743725A CN118279848A CN 118279848 A CN118279848 A CN 118279848A CN 202211743725 A CN202211743725 A CN 202211743725A CN 118279848 A CN118279848 A CN 118279848A
- Authority
- CN
- China
- Prior art keywords
- tile
- lane line
- lane
- map
- splicing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000004590 computer program Methods 0.000 claims abstract description 41
- 238000000605 extraction Methods 0.000 claims abstract description 37
- 230000004927 fusion Effects 0.000 claims description 62
- 238000013507 mapping Methods 0.000 claims description 28
- 238000006243 chemical reaction Methods 0.000 claims description 18
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000004519 manufacturing process Methods 0.000 abstract description 10
- 238000010586 diagram Methods 0.000 description 31
- 238000004891 communication Methods 0.000 description 5
- 238000013136 deep learning model Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000007526 fusion splicing Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
The application relates to a lane line vector information acquisition method, a lane line vector information acquisition device, a lane line vector information acquisition computer device, a lane line vector information acquisition storage medium and a lane line vector information acquisition computer program product. The method comprises the following steps: taking two-dimensional maps of target areas under different resolutions; dividing the two-dimensional map to obtain a tile map set; splicing the tile graphs in the tile map set to obtain a tile splicing graph set; inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph; fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map; and converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area. By adopting the method, the lane line vector information can be obtained efficiently in the process of manufacturing the high-precision map.
Description
Technical Field
The present application relates to the field of high-precision map making, and in particular, to a lane line vector information acquisition method, apparatus, computer device, storage medium, and computer program product.
Background
The high-definition Map is also called a high-resolution Map (HD Map, high Definition Map), which is a Map specially used for the automatic driving service. For the traditional map, the object of the service is a person, as a driver, we can selectively use the map, while the high-precision map is different from the traditional map, the object of the service is an automatic driving system of a vehicle, and the service exists as a safety attribute, so that the safety of automatic driving is greatly influenced.
In high-definition maps, map element data is generally stored in the form of vector information, i.e., an ordered list of latitude and longitude coordinates. The drawing process of the high-precision map is the process of acquiring the road element data vector information. Currently, it is a viable solution to map high-resolution and high-positioning-accuracy maps. The traditional high-precision map making process requires map drawing staff to identify and manually draw vector information of elements such as lane lines, ground marks and the like on the collected two-dimensional road pictures, but the mode requires more consumed manpower, is low in efficiency, and has the problem that the lane line vector information is not obtained efficiently enough.
Disclosure of Invention
Based on this, it is necessary to provide an efficient lane line vector information acquisition method, apparatus, computer device, computer readable storage medium and computer program product in view of the above technical problems.
In a first aspect, the present application provides a lane line vector information acquisition method. The method comprises the following steps:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
Splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph;
Fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
In one embodiment, stitching tile maps in a tile map set to obtain a tile stitching set includes:
acquiring a splicing side length and a splicing step length, wherein the splicing step length is smaller than the splicing side length;
and according to the splicing side length and the splicing step length, splicing the tile graphs in the tile map set according to rows or columns to obtain a tile splicing graph set.
In one embodiment, based on a prediction result of relative coordinates of pixels of lane lines in a tile mosaic, fusing the lane lines in different tile mosaics to obtain absolute coordinates of pixels of the lane lines in a two-dimensional map, including:
Converting a pixel relative coordinate prediction result of a lane line in the tile mosaic into a pixel absolute coordinate;
And fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map so as to determine the fused lane lines in the two-dimensional map and determine the corresponding pixel absolute coordinates.
In one embodiment, fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map includes:
And aiming at any two lane lines, taking the any two lane lines as a first lane line and a second lane line respectively, and taking the lane line with longer length in the first lane line and the second lane line as a fusion result of the first lane line and the second lane line under the condition that the distance between the two end points of the first lane line and the distance between the second lane line are smaller than a fusion distance threshold value.
In one embodiment, the fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map further includes:
Determining a first mapping point on the second lane line closest to the first end point and a second mapping point on the first lane line closest to the second end point under the condition that the distance between the first end point of the first lane line and the second lane line is smaller than a fusion distance threshold and the distance between the second end point of the second lane line and the first lane line is smaller than the fusion distance threshold;
Dividing the first lane line into a first overlapping section and a first non-overlapping section based on the second mapping point, and dividing the second lane line into a second overlapping section and a second non-overlapping section based on the first mapping point;
Determining a target overlapping segment based on the first overlapping segment and the second overlapping segment;
and constructing a fusion result of the first lane line and the second lane line based on the target overlapping section, the first non-overlapping section and the second non-overlapping section.
In one embodiment, converting absolute coordinates of pixels of a lane line in a two-dimensional map to obtain lane line vector information in a target area includes:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the absolute abscissa and absolute ordinate of the lane line in the pixel coordinate system, respectively; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude is the latitude coordinate of the lane line; longitude is the longitude coordinate of the lane line.
In a second aspect, the application further provides a lane line vector information acquisition device. The device comprises:
the map acquisition module is used for acquiring two-dimensional maps of the target area under different resolutions;
The map segmentation module is used for segmenting the two-dimensional map to obtain a tile map set;
The tile map splicing module is used for splicing the tile maps in the tile map set to obtain a tile splicing map set;
the lane line extraction module is used for inputting the tile mosaic image in the tile mosaic image set into the lane line extraction model and outputting a pixel relative coordinate prediction result of the lane line in the tile mosaic image;
The lane line fusion module is used for fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
the coordinate conversion module is used for converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor which when executing the computer program performs the steps of:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
Splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph;
Fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
Splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph;
Fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
Splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph;
Fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
The lane line vector information acquisition method, the lane line vector information acquisition device, the computer equipment, the storage medium and the computer program product acquire two-dimensional maps of target areas under different resolutions; dividing the two-dimensional map to obtain a tile map set; splicing the tile graphs in the tile map set to obtain a tile splicing graph set; inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph; fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map; and converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area. The method comprises the steps of dividing and splicing a two-dimensional map of a target area in the whole lane line vector information acquisition process, acquiring tile spliced images with proper resolution, inputting the tile spliced images into a lane line extraction model, acquiring pixel relative coordinate prediction results of lane lines in each tile spliced image, fusing the lane lines in different tile spliced images based on the pixel relative coordinate prediction results, acquiring pixel absolute coordinates of the lane lines in the two-dimensional map, converting the pixel absolute coordinates, finally acquiring vector information of the lane lines in the two-dimensional map, and realizing efficient acquisition of the lane line vector information in the high-precision map manufacturing process.
Drawings
FIG. 1 is an application environment diagram of a lane line vector information acquisition method in one embodiment;
FIG. 2 is a flow chart of a lane line vector information acquisition method according to an embodiment;
FIG. 3 is a flow chart of acquiring absolute coordinates of pixels of a lane line in one embodiment;
FIG. 4 is a schematic diagram of lane-line fusion splice in one embodiment;
FIG. 5 is a schematic diagram of lane-line fusion splice in another embodiment;
FIG. 6 is a flowchart of a lane line vector information acquiring method according to another embodiment;
FIG. 7 is a block diagram showing the construction of a lane line vector information acquiring apparatus in one embodiment;
Fig. 8 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The lane line vector information acquisition method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server. The user operates on the terminal 102 side, and the terminal 102 responds to the user operation, so that the lane line vector information is obtained efficiently.
Specifically, the terminal 102 acquires a two-dimensional map of the target area at different resolutions; dividing the two-dimensional map to obtain a tile map set; splicing the tile graphs in the tile map set to obtain a tile splicing graph set; inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph; fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map; and converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, portable wearable devices, and the internet of things devices may be intelligent vehicle devices and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.
In one embodiment, as shown in fig. 2, a lane line vector information obtaining method is provided, and the method is applied to the terminal 102 in fig. 1 for illustration, and includes the following steps:
S100: and acquiring two-dimensional maps of the target area under different resolutions.
The first step of high-precision map making is to collect map data, and to initially draw a two-dimensional map of a target area, collect images with high positioning precision of the target area through a map collecting vehicle to form two-dimensional maps with different grades. Vehicles used to collect map data are typically equipped with sensor packages including high-precision integrated navigation, high-beam lidar, high-pixel perception cameras, through which two-dimensional maps of the target area road at different resolutions can be obtained. In this embodiment, the obtained two-dimensional map includes the two-dimensional map of the target area road at the 23-level resolution.
S200: and dividing the two-dimensional map to obtain a tile map set.
The method for extracting the lane line vector information of the high-resolution road picture by using the deep learning technology is a method for improving the drawing efficiency of the high-precision map. But the deep learning model has certain requirements on the size of the processed image. The resolution of the image is too large, and the limitation of computing resources can be broken through; the resolution is too small and the map elements cannot be displayed completely. The 23 rd layer tile map with a square kilometer has a resolution of about 700000×70000, and an image with the size cannot be input into the deep learning model at a time to perform lane line pixel coordinate prediction. Downsampling the picture is not a reasonable solution, the shape of the lane lines on the map is slender, the map spans multiple areas, and all information of the lane lines can be directly lost by greatly downsampling. Therefore, the map is segmented, the lane line vector information is extracted by using the model, and finally, the lane line vector information of each part is fused, so that the method is an inevitable operation.
Based on the tile pyramid model, a two-dimensional map at different resolutions may be partitioned into many small map units, each of which is referred to as a map tile. Each map tile has a unique tile map level (level) and tile coordinate number (tileX, tileY), the tile typically being 256 x 256 pixels in size. From the top layer to the bottom layer of the tile pyramid, the length and width pixel values of the next layer are twice as large as those of the previous layer, the resolution of the map is higher and higher, but the geographical range represented is unchanged, namely, the higher the tile level is, the more tiles form the map, and the physical size and detail expression of the display map are correspondingly increased. In the tile pyramid model, if the level 0 two-dimensional map is represented by a 256×256 pixel picture, the level 23 two-dimensional map pixels are (256×2 23)×(256×223), and each pixel corresponds to a geographic area of about 1.4 square centimeters in low and medium latitude areas. The position and shape of the road lane line, the ground mark and other elements can be clearly displayed under the resolution.
In this embodiment, a two-dimensional map of the target area road at a resolution of 23 is divided into small map tiles, each tile having a resolution of 256×256, representing a geographic area of approximately 3.6x3.6 square meters, with each tile having a unique coordinate number (tileX, tileY). After the two-dimensional map of the target area road at the resolution of 23 is divided, a tile map set composed of map tiles with the resolution of 256×256 is obtained.
S300: and splicing the tile graphs in the tile map set to obtain a tile splicing graph set.
Since the resolution of the individual map tiles after segmentation is too small to display map elements completely, it is necessary to splice the map tiles after segmentation. The tile graphs at all positions are spliced with a certain step length, the splicing step length can be set smaller than the splicing side length, so that overlapping splicing is performed, for example, the splicing side length is 5 tiles, and the splicing step length can be set to 2 tiles. The splicing side length determines the resolution of each tile splice graph, and the splicing step length determines the upper left corner number of the next tile splice graph.
Taking a tile splicing diagram with the upper left corner tile diagram number (tx, ty) as an example of a first tile splicing diagram, the length and width directions of the tile splicing diagram are all 5 tile splicing lengths, and the tile diagram required by splicing the tile splicing diagram and the relative positions among the tile diagrams are shown in the following table 1:
table 1 tile map numbering for each tile map in tile splice map
(tx,ty) | (tx+1,ty) | (tx+2,ty) | (tx+3,ty) | (tx+4,ty) |
(tx,ty+1) | (tx+1,ty+1) | (tx+2,ty+1) | (tx+3,ty+1) | (tx+4,ty+1) |
(tx,ty+2) | (tx+1,ty+2) | (tx+2,ty+2) | (tx+3,ty+2) | (tx+4,ty+2) |
(tx,ty+3) | (tx+1,ty+3) | (tx+2,ty+3) | (tx+3,ty+3) | (tx+4,ty+3) |
(tx,ty+4) | (tx+1,ty+4) | (tx+2,ty+4) | (tx+3,ty+4) | (tx+4,ty+4) |
When the tiles are spliced, because the splicing step length is set to be 2 tiles, the upper left corner tile coordinate number of the first tile splicing diagram is (tx, ty), the upper left corner tile coordinate number of the next tile splicing diagram is (tx+2, ty), the upper left corner tile coordinate number of the next tile splicing diagram is (tx+4, ty), and so on. After the tile map of the row is spliced, the tile map of the next row is spliced, and the coordinate numbers of the upper left corner tiles of the tile map of the next row are as follows: (tx, ty+2), (tx+2, ty+2), (tx+4, ty+2), …. Because the tile splicing graphs are overlapped, the situation that the lane line is too short in a tile graph to be detected can be avoided to a great extent, so that the lane line cannot be spliced. After the overall stitching is completed, a tile stitching atlas is obtained, wherein each tile stitching atlas has a resolution of 1280 x 1280, representing a geographic area of approximately 18 x 18 meters, suitable for model input.
S400: and inputting the tile mosaic in the tile mosaic set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of the lane line in the tile mosaic.
And inputting all the tile spliced graphs in the spliced tile spliced graph set into a trained lane line extraction model, and selecting an ultrafast lane line detection algorithm (Ultra Fast Lane Detection) to obtain a pixel relative coordinate prediction result of the lane line of each tile spliced graph.
S500: and fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map.
After the lane line extraction model outputs a pixel relative coordinate prediction result (refX, refY) of the lane line, firstly, converting the pixel relative coordinate of the lane line in each tile mosaic into a pixel absolute coordinate (absX, absY) of the lane line in the whole two-dimensional map so as to splice the extraction results of the lane lines in different tile mosaics. The formula for converting the pixel relative coordinates into pixel absolute coordinates is: absX =256×tx+ refX, absY =256×ty+ refY. And then checking whether lane lines in different areas can be fused and spliced according to rows or columns, fusing and splicing lane lines which can be fused and spliced according to a preset rule, and keeping all lane lines which cannot be fused and spliced. Taking checking whether lane lines in different areas can be fused and spliced according to a row as an example, fusing and splicing a first tile splicing diagram and a second tile splicing diagram of each row to obtain a first fused area of the row, fusing and splicing lane lines which can be fused and spliced on the first fused area of the row according to a preset rule, and reserving all lane lines which cannot be fused and spliced; and then carrying out fusion splicing on the first fusion area of each column and the third tile splicing chart of the column to obtain a second fusion area of the column, carrying out fusion splicing on lane lines which can be fused and spliced on the second fusion area of the column according to a preset rule, and keeping all lane lines which cannot be fused and spliced, and the like. And after the fusion splicing result of each column is obtained, splicing the columns in the horizontal direction, so that the absolute coordinates of the pixels of the lane lines after fusion in the two-dimensional map are obtained.
S600: and converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
After the absolute pixel coordinates of the lane lines after fusion in the two-dimensional map are obtained, the absolute coordinates of the lane lines in the pixel coordinate system are required to be converted into longitude and latitude coordinates in the geographic coordinate system, and vector information of the lane lines in the target area high-precision map is obtained. The conversion formula is:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the abscissa and ordinate, respectively, of the lane line in the pixel absolute coordinate system; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
The lane line vector information acquisition method acquires two-dimensional maps of the target area under different resolutions; dividing the two-dimensional map to obtain a tile map set; splicing the tile graphs in the tile map set to obtain a tile splicing graph set; inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph; fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map; and converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area. The method comprises the steps of dividing and splicing a two-dimensional map of a target area in the whole lane line vector information acquisition process, acquiring tile spliced images with proper resolution, inputting the tile spliced images into a lane line extraction model, acquiring pixel relative coordinate prediction results of lane lines in each tile spliced image, fusing the lane lines in different tile spliced images based on the pixel relative coordinate prediction results, acquiring pixel absolute coordinates of the lane lines in the two-dimensional map, converting the pixel absolute coordinates, finally acquiring vector information of the lane lines in the two-dimensional map, and realizing efficient acquisition of the lane line vector information in the high-precision map manufacturing process.
In one embodiment, stitching tile maps in a tile map set to obtain a tile stitching map set, comprising:
acquiring a splicing side length and a splicing step length, wherein the splicing step length is smaller than the splicing side length;
and according to the splicing side length and the splicing step length, splicing the tile graphs in the tile map set according to rows or columns to obtain a tile splicing graph set.
The vector information of the lane lines in the high-resolution picture is extracted by using the deep learning model, so that the drawing efficiency of the high-precision map can be effectively improved, but the deep learning model has certain requirements on the resolution of the processed image, and the limitation of calculation resources can be broken through when the resolution is too large; too small a resolution may not be able to completely extract the lane lines in the target area. In order to fit the resolution of the tile map input to the lane line extraction model, it is necessary to splice the tile map after acquiring a tile map set having a resolution of 256×256.
Taking the two-dimensional map of the acquired target area road at the 23-level resolution as an example, the resolution of each tile map after segmentation is 256×256, representing a geographic area of about 3.6x3.6 square meters, and each tile map has a unique coordinate number (tileX, tileY). Before splicing the tile graphs, firstly determining the splicing side length and the splicing step length, and in order to avoid the situation that the lane lines are too short in a certain tile graph to be missed to be spliced, the tile splicing graphs can be overlapped, so that the splicing step length is smaller than the splicing side length, for example, the splicing side length is set to be 5 tiles, and the splicing step length can be set to be 2 tiles. The splicing side length determines the resolution of each tile splice graph, and the splicing step length determines the upper left corner number of the next tile splice graph.
The tile map with the upper left corner tile map number (tx, ty) is taken as the first tile map, and the tiles are spliced according to the splicing length of 5 tiles, so that the tile map required by the tile map and the relative positions among the tile maps are shown in table 1. According to the set splicing side length and the set splicing step length, the tile maps in the tile map set can be spliced according to rows or columns. When the tiles are spliced according to the rows, the upper left corner of the first tile splice diagram is numbered (tx, ty), the splicing step length is set to be 2 tiles, the upper left corner of the next tile splice diagram is numbered (tx+2, ty), the upper left corner of the next tile splice diagram is numbered (tx+4, ty), and the like. After one row of splicing is completed, next, splicing the next row of tile map, wherein the tile numbers at the upper left corner of the next row of tile splicing map are as follows in sequence: (tx, ty+2), (tx+2, ty+2), (tx+4, ty+2), …, thereby obtaining a tile splice atlas. Each tile mosaic in the tile mosaic set has a resolution of 1280 x 1280, representing a geographic area of approximately 18 x 18 meters, and is suitable as input to the lane line extraction model.
In the embodiment, the tile map with the resolution ratio of 256×256 in the tile mosaic map set is further stitched, and the stitching step length is set to be smaller than the stitching side length, so that overlapping stitching is realized, the situation that the lane lines cannot be stitched due to missed detection caused by too short lane lines in a tile map is effectively avoided, and accurate extraction of lane line vector information in the high-precision map manufacturing process is realized.
In one embodiment, as shown in fig. 3, based on the prediction result of the relative coordinates of the pixels of the lane lines in the tile mosaic, the fusion of the lane lines in different tile mosaics is performed to obtain the absolute coordinates of the pixels of the lane lines in the two-dimensional map, including:
s520: converting a pixel relative coordinate prediction result of a lane line in the tile mosaic into a pixel absolute coordinate;
S540: and fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map so as to determine the fused lane lines in the two-dimensional map and determine the corresponding pixel absolute coordinates.
In order to fuse and splice the lane line prediction results in the tile stitching graphs, the lane line extraction model outputs the pixel relative coordinates of the lane lines, and the pixel relative coordinates (refX, refY) of the lane lines in each tile stitching graph need to be converted into pixel absolute coordinates (absX, absY) of the lane lines in the whole two-dimensional map, and the conversion formula for converting the pixel relative coordinates into the pixel absolute coordinates is as follows: absX =256×tx+ refX, absY =256×ty+ refY. And according to the pixel absolute coordinates of the lane lines in the two-dimensional map, fusing and splicing the lane line prediction results in the different tile splicing diagrams, and finally obtaining the pixel absolute coordinates of the lane lines after fusing in the two-dimensional map.
In this embodiment, after the absolute coordinates of pixels of the lane lines in the two-dimensional map are obtained, the relative coordinates of pixels of the lane lines in the two-dimensional map are converted into the absolute coordinates according to the coordinate conversion formula, so that the lane lines in different tile splicing diagrams are fused and spliced, and efficient extraction of the lane line vector information in the high-precision map manufacturing process is realized.
In one embodiment, fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map includes:
And aiming at any two lane lines, taking the any two lane lines as a first lane line and a second lane line respectively, and taking the lane line with longer length in the first lane line and the second lane line as a fusion result of the first lane line and the second lane line under the condition that the distance between the two end points of the first lane line and the distance between the second lane line are smaller than a fusion distance threshold value.
After the absolute coordinates of pixels of lane lines after fusion in the two-dimensional map are obtained, a fusion distance threshold of the lane lines is set, and whether the lane lines in different areas can be fused and spliced is checked according to the set fusion distance threshold by rows or columns. Referring to fig. 4, for any two lane lines a and b in the area, if the distances from two end points of the line a to the other line b are smaller than the fusion distance threshold, it is indicated that the line a and the line b are completely in the same tile map and the distances are very similar, in which case, the line a and the line b are fused into one lane line (in this case, the line b) longer in the two.
In this embodiment, after obtaining the absolute coordinates of pixels of lane lines after fusion in the two-dimensional map, a fusion distance threshold of the lane lines is set, and whether the lane lines in different areas can be fused and spliced is checked according to the fusion distance threshold. Aiming at the situation that two lane lines in the area are completely arranged in the tile map and the distances of the two lane lines are very similar, the two lane lines are fused into a longer lane line in the two lane lines, so that the accurate extraction of the lane line vector information in the high-precision map manufacturing process is realized.
In an embodiment, the fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map further includes:
Determining a first mapping point on the second lane line closest to the first end point and a second mapping point on the first lane line closest to the second end point under the condition that the distance between the first end point of the first lane line and the second lane line is smaller than a fusion distance threshold and the distance between the second end point of the second lane line and the first lane line is smaller than the fusion distance threshold;
Dividing the first lane line into a first overlapping section and a first non-overlapping section based on the second mapping point, and dividing the second lane line into a second overlapping section and a second non-overlapping section based on the first mapping point;
Determining a target overlapping segment based on the first overlapping segment and the second overlapping segment;
and constructing a fusion result of the first lane line and the second lane line based on the target overlapping section, the first non-overlapping section and the second non-overlapping section.
Referring to fig. 5, for any two lane lines a and b in the area, if the distance from one end point of the line a to the other line b is smaller than the fusion distance threshold and the distance from one end point of the line b to the other line a is smaller than the fusion distance threshold, it is indicated that the line a and the line b may exist in different tile graphs, and some of the two lines are very close to each other. In this case, the lines a and b are divided into overlapping sections and non-overlapping sections, respectively, to obtain (a_o, a_s) and (b_o, b_s), and for the two overlapping sections a_o and b_o, they are fused into the longer one of the two (in this case, the b_o section) or the one with higher prediction confidence outputted by the lane line extraction model; for two non-overlapping segments a_s and b_s, they are all reserved.
In this embodiment, for the case that two lane lines exist in different tile graphs in the region and the distances between the line segments in the two lane lines are very similar, the two lane lines are respectively divided into an overlapping segment and a non-overlapping segment, and for the two overlapping segments, the two overlapping segments are fused into a longer segment or a segment with higher prediction confidence, and for the non-overlapping segment, all the lane line vector information is reserved, so that accurate extraction of the lane line vector information in the high-precision map manufacturing process is realized.
In one embodiment, converting absolute coordinates of pixels of a lane line in a two-dimensional map to obtain lane line vector information in a target area includes:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the abscissa and ordinate, respectively, of the lane line in the pixel absolute coordinate system; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
After the absolute pixel coordinates of the lane lines in the two-dimensional map are extracted, the pixel coordinate information on the tile mosaic is required to be converted into lane line vector information marked on the high-precision map. And finally obtaining lane line vector information in the whole target area according to a conversion formula between the absolute coordinates of the pixels and the geographic longitude and latitude coordinates.
In the embodiment, the absolute coordinates of the pixels of the lane lines in the two-dimensional map are converted to obtain the lane line vector information (longitude and latitude coordinates) in the target area, so that the internal manufacturing efficiency of the high-precision map is effectively improved, and the efficient extraction of the lane line vector information in the high-precision map manufacturing process is realized.
In order to describe the technical solution of the lane line vector information acquiring method of the present application in detail, a specific application example will be adopted in the following, and the whole processing procedure will be described in detail with reference to fig. 6, which specifically includes the following steps:
1. The two-dimensional map of the target area under different resolutions is obtained through the map collecting vehicle, the map under different resolutions comprises the two-dimensional map of the target area road under 23-level resolution, and the positions and the shapes of the road lane lines and other elements can be clearly displayed under the resolution.
2. And dividing the acquired two-dimensional map to acquire a tile map set. Each tile in the set of tile maps has a resolution of 256 x 256 representing a geographic area of approximately 3.6x3.6 square meters, with each tile having a unique coordinate number (tileX, tileY).
3. And splicing tiles in the tile map set to obtain a tile splicing map set. Setting the splicing step length smaller than the splicing side length, so as to splice overlapped tiles, taking a tile with the splicing side length of 5 and a tile with the splicing step length of 2 as an example, and assuming that the tile diagram number of the upper left corner of the first tile splicing diagram is (tx, ty), the relative positions of the tile diagram required by the tile splicing diagram and each tile diagram are shown in the following table:
table 1 tile map numbering for each tile map in tile splice map
(tx,ty) | (tx+1,ty) | (tx+2,ty) | (tx+3,ty) | (tx+4,ty) |
(tx,ty+1) | (tx+1,ty+1) | (tx+2,ty+1) | (tx+3,ty+1) | (tx+4,ty+1) |
(tx,ty+2) | (tx+1,ty+2) | (tx+2,ty+2) | (tx+3,ty+2) | (tx+4,ty+2) |
(tx,ty+3) | (tx+1,ty+3) | (tx+2,ty+3) | (tx+3,ty+3) | (tx+4,ty+3) |
(tx,ty+4) | (tx+1,ty+4) | (tx+2,ty+4) | (tx+3,ty+4) | (tx+4,ty+4) |
When the tiles are spliced, because the splicing step length is set to be 2 tiles, the upper left corner of the first tile splicing diagram is numbered (tx, ty), the upper left corner of the next tile splicing diagram is numbered (tx+2, ty), the upper left corner of the next tile splicing diagram is numbered (tx+4, ty), and so on. After the tile map of the row is spliced, the tile map of the next row is spliced, and the upper left corner tile numbers of the tile map of the next row are as follows: (tx, ty+2), (tx+2, ty+2), (tx+4, ty+2), …. Because the tile splicing graphs are overlapped, the situation that the lane line is too short in a tile graph to be detected can be avoided to a great extent, so that the lane line cannot be spliced. After the overall stitching is completed, a tile stitching atlas is obtained, wherein each tile stitching atlas has a resolution of 1280 x 1280, representing a geographic area of approximately 18 x 18 meters, suitable for model input.
4. And inputting the tile splice graphs in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of the lane line in each tile splice graph.
5. And fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map. The method comprises the following specific steps:
a) Through a coordinate conversion formula: absX =256×tx+ refX, absY =256×ty+ refY, and converting the pixel relative coordinate prediction result (refX, refY) of the lane line in each tile mosaic into pixel absolute coordinates (absX, absY) of the lane line in the whole two-dimensional map.
B) And setting a fusion distance threshold value of the lane lines.
C) Checking whether lane lines in different areas can be fused and spliced according to rows or columns, fusing and splicing lane lines which can be fused and spliced according to a preset rule, and reserving all lane lines which cannot be fused and spliced. For any two lane lines a and b in the area, please refer to fig. 4, if the distances from the two end points of the line a to the other line b are smaller than the fusion distance threshold, the line a and the line b are fused into the longer lane line (the line b in this example). Referring to fig. 5, if the distance from one end point of the line a to the other line b is smaller than the fusion distance threshold value, and the distance from one end point of the line b to the other line a is smaller than the fusion distance threshold value, dividing the line a and the line b into an overlapping section and a non-overlapping section respectively to obtain (a_o, a_s) and (b_o, b_s), and fusing the two overlapping sections a_o and b_o into a longer section (b_o section in this example) of the two overlapping sections or a section with higher prediction confidence outputted by the lane line extraction model; for two non-overlapping segments a_s and b_s, they are all reserved.
6. And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area. The conversion formula between the absolute coordinates and the longitude and latitude coordinates of the pixel is as follows:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the abscissa and ordinate, respectively, of the lane line in the pixel absolute coordinate system; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, as shown in fig. 7, the embodiment of the application further provides a lane line vector information acquisition device for implementing the lane line vector information acquisition method. The device comprises:
The map acquisition module 701 is configured to acquire two-dimensional maps of the target area under different resolutions;
The map segmentation module 702 is configured to segment a two-dimensional map to obtain a tile map set;
a tile map stitching module 703, configured to stitch tile maps in the tile map set to obtain a tile stitching map set;
The lane line extraction module 704 is configured to input a tile mosaic in the tile mosaic set to the lane line extraction model, and output a prediction result of relative coordinates of pixels of a lane line in the tile mosaic;
The lane line fusion module 705 is configured to fuse lane lines in different tile mosaic graphs based on a prediction result of relative coordinates of pixels of the lane lines in the tile mosaic graphs, so as to obtain absolute coordinates of pixels of the lane lines in the two-dimensional map;
The coordinate conversion module 706 is configured to convert absolute coordinates of pixels of a lane line in the two-dimensional map to obtain lane line vector information in the target area.
The lane line vector information acquisition device acquires two-dimensional maps of target areas under different resolutions; dividing the two-dimensional map to obtain a tile map set; splicing the tile graphs in the tile map set to obtain a tile splicing graph set; inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph; fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map; and converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area. The method comprises the steps of dividing and splicing a two-dimensional map of a target area in the whole lane line vector information acquisition process, acquiring tile spliced images with proper resolution, inputting the tile spliced images into a lane line extraction model, acquiring pixel relative coordinate prediction results of lane lines in each tile spliced image, fusing the lane lines in different tile spliced images based on the pixel relative coordinate prediction results, acquiring pixel absolute coordinates of the lane lines in the two-dimensional map, converting the pixel absolute coordinates, finally acquiring vector information of the lane lines in the two-dimensional map, and realizing efficient acquisition of the lane line vector information in the high-precision map manufacturing process.
In one embodiment, the tile graph stitching module 703 is further configured to obtain a stitching side length and a stitching step length, where the stitching step length is smaller than the stitching side length; and according to the splicing side length and the splicing step length, splicing the tile graphs in the tile map set according to rows or columns to obtain a tile splicing graph set.
In one embodiment, the lane line fusion module 705 is further configured to convert a prediction result of the relative coordinates of pixels of the lane line in the tile stitching graph into absolute coordinates of the pixels; and fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map so as to determine the fused lane lines in the two-dimensional map and determine the corresponding pixel absolute coordinates.
In one embodiment, the lane line fusion module 705 is further configured to, for any two lane lines, respectively use the any two lane lines as a first lane line and a second lane line, and use a lane line with a longer length of the first lane line and the second lane line as a fusion result of the first lane line and the second lane line when the distances between two end points of the first lane line and the second lane line are both smaller than a fusion distance threshold.
In one embodiment, the lane line fusion module 705 is further configured to determine a first mapping point on the second lane line closest to the first end point and a second mapping point on the first lane line closest to the second end point if the distance between the first end point of the first lane line and the second lane line is less than the fusion distance threshold and the distance between the second end point of the second lane line and the first lane line is less than the fusion distance threshold; dividing the first lane line into a first overlapping section and a first non-overlapping section based on the second mapping point, and dividing the second lane line into a second overlapping section and a second non-overlapping section based on the first mapping point; determining a target overlapping segment based on the first overlapping segment and the second overlapping segment; and constructing a fusion result of the first lane line and the second lane line based on the target overlapping section, the first non-overlapping section and the second non-overlapping section.
In one embodiment, the coordinate conversion module 706 is further configured to obtain lane line vector information in the target area according to a coordinate conversion formula, where the coordinate conversion formula is:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the abscissa and ordinate, respectively, of the lane line in the pixel absolute coordinate system; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
The above-described respective modules in the lane line vector information acquiring apparatus may be realized in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 8. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a lane line vector information acquisition method. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in FIG. 8 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
Splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph;
Fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
In one embodiment, the processor when executing the computer program further performs the steps of: acquiring a splicing side length and a splicing step length, wherein the splicing step length is smaller than the splicing side length; and according to the splicing side length and the splicing step length, splicing the tile graphs in the tile map set according to rows or columns to obtain a tile splicing graph set.
In one embodiment, the processor when executing the computer program further performs the steps of: converting a pixel relative coordinate prediction result of a lane line in the tile mosaic into a pixel absolute coordinate; and fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map so as to determine the fused lane lines in the two-dimensional map and determine the corresponding pixel absolute coordinates.
In one embodiment, the processor when executing the computer program further performs the steps of: and aiming at any two lane lines, taking the any two lane lines as a first lane line and a second lane line respectively, and taking the lane line with longer length in the first lane line and the second lane line as a fusion result of the first lane line and the second lane line under the condition that the distance between the two end points of the first lane line and the distance between the second lane line are smaller than a fusion distance threshold value.
In one embodiment, the processor when executing the computer program further performs the steps of: determining a first mapping point on the second lane line closest to the first end point and a second mapping point on the first lane line closest to the second end point under the condition that the distance between the first end point of the first lane line and the second lane line is smaller than a fusion distance threshold and the distance between the second end point of the second lane line and the first lane line is smaller than the fusion distance threshold;
Dividing the first lane line into a first overlapping section and a first non-overlapping section based on the second mapping point, and dividing the second lane line into a second overlapping section and a second non-overlapping section based on the first mapping point;
Determining a target overlapping segment based on the first overlapping segment and the second overlapping segment;
and constructing a fusion result of the first lane line and the second lane line based on the target overlapping section, the first non-overlapping section and the second non-overlapping section.
In one embodiment, the processor when executing the computer program further performs the steps of: converting the absolute coordinates of pixels of lane lines in the two-dimensional map according to a coordinate conversion formula, wherein the coordinate conversion formula is as follows:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the abscissa and ordinate, respectively, of the lane line in the pixel absolute coordinate system; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
Splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph;
Fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a splicing side length and a splicing step length, wherein the splicing step length is smaller than the splicing side length; and according to the splicing side length and the splicing step length, splicing the tile graphs in the tile map set according to rows or columns to obtain a tile splicing graph set.
In one embodiment, the computer program when executed by the processor further performs the steps of: converting a pixel relative coordinate prediction result of a lane line in the tile mosaic into a pixel absolute coordinate; and fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map so as to determine the fused lane lines in the two-dimensional map and determine the corresponding pixel absolute coordinates.
In one embodiment, the computer program when executed by the processor further performs the steps of: and aiming at any two lane lines, taking the any two lane lines as a first lane line and a second lane line respectively, and taking the lane line with longer length in the first lane line and the second lane line as a fusion result of the first lane line and the second lane line under the condition that the distance between the two end points of the first lane line and the distance between the second lane line are smaller than a fusion distance threshold value.
In one embodiment, the computer program when executed by the processor further performs the steps of: determining a first mapping point on the second lane line closest to the first end point and a second mapping point on the first lane line closest to the second end point under the condition that the distance between the first end point of the first lane line and the second lane line is smaller than a fusion distance threshold and the distance between the second end point of the second lane line and the first lane line is smaller than the fusion distance threshold;
Dividing the first lane line into a first overlapping section and a first non-overlapping section based on the second mapping point, and dividing the second lane line into a second overlapping section and a second non-overlapping section based on the first mapping point;
Determining a target overlapping segment based on the first overlapping segment and the second overlapping segment;
and constructing a fusion result of the first lane line and the second lane line based on the target overlapping section, the first non-overlapping section and the second non-overlapping section.
In one embodiment, the computer program when executed by the processor further performs the steps of: converting the absolute coordinates of pixels of lane lines in the two-dimensional map according to a coordinate conversion formula, wherein the coordinate conversion formula is as follows:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the abscissa and ordinate, respectively, of the lane line in the pixel absolute coordinate system; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
Splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
inputting the tile splice graph in the tile splice graph set into a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile splice graph;
Fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring a splicing side length and a splicing step length, wherein the splicing step length is smaller than the splicing side length; and according to the splicing side length and the splicing step length, splicing the tile graphs in the tile map set according to rows or columns to obtain a tile splicing graph set.
In one embodiment, the computer program when executed by the processor further performs the steps of: converting a pixel relative coordinate prediction result of a lane line in the tile mosaic into a pixel absolute coordinate; and fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map so as to determine the fused lane lines in the two-dimensional map and determine the corresponding pixel absolute coordinates.
In one embodiment, the computer program when executed by the processor further performs the steps of: and aiming at any two lane lines, taking the any two lane lines as a first lane line and a second lane line respectively, and taking the lane line with longer length in the first lane line and the second lane line as a fusion result of the first lane line and the second lane line under the condition that the distance between the two end points of the first lane line and the distance between the second lane line are smaller than a fusion distance threshold value.
In one embodiment, the computer program when executed by the processor further performs the steps of: determining a first mapping point on the second lane line closest to the first end point and a second mapping point on the first lane line closest to the second end point under the condition that the distance between the first end point of the first lane line and the second lane line is smaller than a fusion distance threshold and the distance between the second end point of the second lane line and the first lane line is smaller than the fusion distance threshold;
Dividing the first lane line into a first overlapping section and a first non-overlapping section based on the second mapping point, and dividing the second lane line into a second overlapping section and a second non-overlapping section based on the first mapping point;
Determining a target overlapping segment based on the first overlapping segment and the second overlapping segment;
and constructing a fusion result of the first lane line and the second lane line based on the target overlapping section, the first non-overlapping section and the second non-overlapping section.
In one embodiment, the computer program when executed by the processor further performs the steps of: converting the absolute coordinates of pixels of lane lines in the two-dimensional map according to a coordinate conversion formula, wherein the coordinate conversion formula is as follows:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the abscissa and ordinate, respectively, of the lane line in the pixel absolute coordinate system; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase ChangeMemory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as Static Random access memory (Static Random access memory AccessMemory, SRAM) or dynamic Random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.
Claims (10)
1. A lane line vector information acquisition method, the method comprising:
Acquiring two-dimensional maps of target areas under different resolutions;
Dividing the two-dimensional map to obtain a tile map set;
splicing the tile graphs in the tile map set to obtain a tile splicing graph set;
Inputting the tile mosaic in the tile mosaic set to a lane line extraction model, and outputting a pixel relative coordinate prediction result of a lane line in the tile mosaic;
fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
And converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
2. The method of claim 1, wherein the stitching tile maps in the tile map set to obtain a tile stitching set comprises:
Acquiring a splicing side length and a splicing step length, wherein the splicing step length is smaller than the splicing side length;
And according to the splicing side length and the splicing step length, splicing the tile graphs in the tile map set according to rows or columns to obtain a tile splicing graph set.
3. The method according to claim 1, wherein the fusing the lane lines in different tile stitching graphs based on the prediction result of the pixel relative coordinates of the lane lines in the tile stitching graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map includes:
Converting a pixel relative coordinate prediction result of a lane line in the tile mosaic into a pixel absolute coordinate;
and fusing the lane lines in the two-dimensional map according to the pixel absolute coordinates of the lane lines in the two-dimensional map so as to determine the fused lane lines in the two-dimensional map and determine the corresponding pixel absolute coordinates.
4. A method according to claim 3, wherein the fusing of the lane lines in the two-dimensional map according to their pixel absolute coordinates comprises:
and aiming at any two lane lines, taking the any two lane lines as a first lane line and a second lane line respectively, and taking the lane line with longer length in the first lane line and the second lane line as a fusion result of the first lane line and the second lane line under the condition that the distances between the two end points of the first lane line and the distance between the second lane line are smaller than a fusion distance threshold value.
5. The method according to claim 4, wherein the method further comprises:
Determining a first mapping point on the second lane line closest to the first end point and a second mapping point on the first lane line closest to the second end point when the distance between the first end point of the first lane line and the second lane line is less than the fusion distance threshold and the distance between the second end point of the second lane line and the first lane line is less than the fusion distance threshold;
dividing the first lane line into a first overlapping section and a first non-overlapping section based on the second mapping point, and dividing the second lane line into a second overlapping section and a second non-overlapping section based on the first mapping point;
determining a target overlap segment based on the first overlap segment and the second overlap segment;
and constructing a fusion result of the first lane line and the second lane line based on the target overlapping section, the first non-overlapping section and the second non-overlapping section.
6. The method of claim 1, wherein converting absolute coordinates of pixels of a lane line in the two-dimensional map to obtain lane line vector information in the target area comprises:
x=absX/(256×2level)-0.5;
y=0.5-absY/(256×2level);
latitude=90-360×arctan(e-2πy)/π;
longitude=360×x;
Wherein absX and absY are the absolute abscissa and absolute ordinate of the lane line in the pixel coordinate system, respectively; level is the level of the tile map; x and y are the abscissa and the ordinate of the lane line on the normalized plane respectively; latitude and longitude are latitude and longitude coordinates, respectively, of the lane line in the geographic coordinate system.
7. A lane line vector information acquisition apparatus, characterized by comprising:
the map acquisition module is used for acquiring two-dimensional maps of the target area under different resolutions;
the map segmentation module is used for segmenting the two-dimensional map to obtain a tile map set;
the tile map splicing module is used for splicing the tile maps in the tile map set to obtain a tile splicing map set;
The lane line extraction module is used for inputting the tile mosaic image in the tile mosaic image set into a lane line extraction model and outputting a pixel relative coordinate prediction result of the lane line in the tile mosaic image;
The lane line fusion module is used for fusing the lane lines in different tile spliced graphs based on the pixel relative coordinate prediction results of the lane lines in the tile spliced graphs to obtain the pixel absolute coordinates of the lane lines in the two-dimensional map;
and the coordinate conversion module is used for converting the pixel absolute coordinates of the lane lines in the two-dimensional map to obtain lane line vector information in the target area.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211743725.0A CN118279848A (en) | 2022-12-30 | 2022-12-30 | Lane line vector information acquisition method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211743725.0A CN118279848A (en) | 2022-12-30 | 2022-12-30 | Lane line vector information acquisition method, device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN118279848A true CN118279848A (en) | 2024-07-02 |
Family
ID=91643274
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211743725.0A Pending CN118279848A (en) | 2022-12-30 | 2022-12-30 | Lane line vector information acquisition method, device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN118279848A (en) |
-
2022
- 2022-12-30 CN CN202211743725.0A patent/CN118279848A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10621696B2 (en) | Deep convolutional image up-sampling | |
EP2954468B1 (en) | Providing indoor facility information on a digital map | |
CN112750203A (en) | Model reconstruction method, device, equipment and storage medium | |
US20210383544A1 (en) | Semantic segmentation ground truth correction with spatial transformer networks | |
CN112530009A (en) | Three-dimensional topographic map drawing method and system | |
US11361490B2 (en) | Attention guidance for ground control labeling in street view imagery | |
CN116186354B (en) | Method, apparatus, electronic device, and computer-readable medium for displaying regional image | |
CN103123263A (en) | Method and device for displaying navigation electronic map | |
CN116091431A (en) | Case Liang Binghai detection method, apparatus, computer device, and storage medium | |
US20240125613A1 (en) | Map generation method and map generation apparatus | |
US20220058825A1 (en) | Attention guidance for correspondence labeling in street view image pairs | |
Yoo et al. | True orthoimage generation by mutual recovery of occlusion areas | |
CN117470246A (en) | Path planning method and device, storage medium and electronic equipment | |
CN115830073A (en) | Map element reconstruction method, map element reconstruction device, computer equipment and storage medium | |
CN116823966A (en) | Internal reference calibration method and device for camera, computer equipment and storage medium | |
CN116758206A (en) | Vector data fusion rendering method and device, computer equipment and storage medium | |
CN118279848A (en) | Lane line vector information acquisition method, device, computer equipment and storage medium | |
CN115797438A (en) | Object positioning method, device, computer equipment, storage medium and program product | |
CN115482152A (en) | Grid map import design software method and device and computer equipment | |
CN114238541A (en) | Sensitive target information acquisition method and device and computer equipment | |
CN118297796A (en) | Training method and device for road element extraction model and computer equipment | |
CN113362458A (en) | Three-dimensional model interpretation method for simulating multi-view imaging, terminal and storage medium | |
CN117689832B (en) | Traffic sign generation method, device, equipment and storage medium | |
CN117455980A (en) | Coordinate correction method, coordinate correction device, computer equipment and storage medium | |
CN117893990B (en) | Road sign detection method, device and computer equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination |