Ground feature contouring method, electronic device, and computer program product

Abstract

Embodiments of the present disclosure disclose a contour drawing method of a ground element, an electronic device and a computer program product. The method comprises: acquiring a circled range in a point cloud image; determining point cloud data surrounded in the circled range in the point cloud image; constructing a road plane based on the point cloud data in the circled range; constructing a gray scale image of the point cloud data in the circled range after projection on the road plane based on the reflection intensity of the point cloud data; determining an image block contour of foreground content in the gray scale image; and drawing an outer contour of the ground element in the point cloud image based on the image block contour. The technical solution draws the outer contour of the ground element in the point cloud image by a semi-automatic drawing method through the combination of artificial and machine when making a high-precision map, which can save the labor cost, reduce the complexity of manual operation and improve the production efficiency of the high-precision map while ensuring high precision.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of location-based services, and particularly relates to a contour drawing method of a ground element, an electronic device and a computer program product. BACKGROUND

[0002] With the development of location-based services (LBS), more and more application software integrates service capabilities related to electronic maps. In addition, some artificial intelligence devices also rely on electronic maps, for example, autonomous driving devices rely on high-precision maps to realize the driving process. Compared with the electronic navigation map used by the driver in the manual driving mode, the high-precision map describes the road through more accurate and rich information, so that the autonomous driving device can understand the road conditions, and then plan actions based on the road conditions, and realize the driving process by executing the actions.

[0003] In the production process of the high-precision map, the outer contour of the nonlinear ground element needs to be drawn to meet the positioning needs of autonomous driving. Because the autonomous driving has high requirements for data accuracy, the fully automatic method based on deep learning cannot meet the needs, and in some scenarios, it cannot guarantee high-precision requirements, especially for the outer frame processing, which often needs manual correction and redrawing.

[0004] However, manual drawing is difficult in a three-dimensional scene, and ground wear will affect the efficiency of manual operation, so a simple, efficient and stable drawing scheme is needed to solve the technical problem that manual drawing of the outer contour of the nonlinear ground element is relatively complex. SUMMARY

[0005] The present disclosure provides a contour drawing method of a ground element, an electronic device and a computer program product.

[0006] In a first aspect, the present disclosure provides a contour drawing method of a ground element, which comprises:

[0007] Obtaining a circled range in a point cloud image;

[0008] Determining point cloud data in the point cloud image that is enclosed in the circled range;

[0009] Constructing a road plane based on the point cloud data in the circled range;

[0010] Based on the reflection intensity of the point cloud data, constructing a gray scale image of the point cloud data in the circled range after projection on the road plane;

[0011] Determining an image block contour of foreground content in the gray scale image; wherein the foreground content is a component of the ground element;

[0012] drawing an outer contour of the ground feature in the point cloud image based on the image block contour.

[0013] Further, based on the reflection intensity of the point cloud data, a grayscale image of the point cloud data in the circumscribed range projected on the road plane is constructed, including:

[0014] projecting the point cloud data in the circumscribed range to the road plane;

[0015] generating the grayscale image based on the reflection intensity of the projected point cloud data.

[0016] Further, drawing an outer contour of the ground feature in the point cloud image based on the image block contour, including:

[0017] When the image block contour includes multiple, determining the vector data of the outer contour of the ground feature based on the image block contour with an area greater than or equal to a preset area threshold;

[0018] Rendering the outer contour of the ground feature on the point cloud image based on the vector data.

[0019] Further, the circumscribed range includes a point cloud frame; determining the point cloud data in the circumscribed range in the point cloud image, including:

[0020] Performing point cloud collision detection on the octree spatial structure corresponding to the point cloud data on the point cloud image and the point cloud frame to obtain rough point cloud data having collision with the point cloud frame;

[0021] Down-sampling the rough point cloud data;

[0022] Performing accurate point cloud collision detection based on the spatial coordinates of the down-sampled rough point cloud data and the point cloud frame;

[0023] Based on the detection result, eliminating the rough point cloud data outside the point cloud frame, and determining the point cloud data in the circumscribed range based on the retained rough point cloud data.

[0024] Further, determining the image block contour of the foreground content in the grayscale image, including:

[0025] Determining the image block contour of the foreground content in the grayscale image based on the reflection intensity difference between the point cloud data corresponding to the road surface and the point cloud data corresponding to the ground feature in the grayscale image.

[0026] Further, the method further includes:

[0027] Displaying a point cloud image;

[0028] receive a line drawing range and an outward expansion direction on the point cloud image;

[0029] determine the bounding range surrounding the ground element to be drawn contour based on the line drawing range and the outward expansion direction, the bounding range including a preset shape frame obtained after expanding a certain distance in the line drawing range to the outward expansion direction.

[0030] In a second aspect, the embodiments of the present disclosure provide a location-based service providing method, which comprises: drawing an outer contour of a ground element on a map by using the method of the first aspect, and providing a location-based service for a served object by drawing a map with the outer contour of the ground element, the location-based service comprising one or more of navigation, map rendering, and route planning.

[0031] In a third aspect, the embodiments of the present disclosure provide a contour drawing device for a ground element, which comprises:

[0032] The first acquisition module is configured to acquire a bounding range in a point cloud image.

[0033] The first determination module is configured to determine point cloud data in the point cloud image that is surrounded in the bounding range.

[0034] The first construction module is configured to construct a road plane based on the point cloud data in the bounding range.

[0035] The second construction module is configured to construct a grayscale image of the point cloud data in the bounding range projected on the road plane based on the reflection intensity of the point cloud data.

[0036] The second determination module is configured to determine an image block contour of foreground content in the grayscale image, wherein the foreground content is a component of the ground element.

[0037] The drawing module is configured to draw an outer contour of the ground element in the point cloud image based on the image block contour.

[0038] The functions can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0039] In one possible design, the structure of the above device includes a memory and a processor, the memory is used to store one or more computer instructions supporting the above device to execute the above corresponding method, and the processor is configured to execute the computer instructions stored in the memory. The above device can also include a communication interface for communication between the above device and other devices or communication networks.

[0040] In a fourth aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method of any of the above aspects.

[0041] In a fifth aspect, a computer readable storage medium is provided, which stores computer instructions for the above device, and the computer instructions are executed by a processor to implement the method of any of the above aspects.

[0042] In a sixth aspect, a computer program product is provided, which includes computer instructions, and the computer instructions are executed by a processor to implement the method of any of the above aspects.

[0043] The technical solutions provided by the embodiments of the present disclosure can include the following beneficial effects:

[0044] In the process of drawing the outer contour of the ground feature, the embodiments of the present disclosure manually assist in drawing the approximate range of the ground feature to be drawn, automatically determine all point cloud data contained in the approximate range based on the approximate range and the point cloud image, convert all point cloud data contained in the approximate range into a grayscale image through the reflection intensity of the point cloud data, determine the image block contour with foreground content based on the grayscale image, and render the outer contour of the ground feature on the point cloud image based on the image block contour. In the process of making a high-precision map, the embodiments of the present disclosure draw the outer contour of the ground feature on the point cloud image through a semi-automatic drawing method by combining manual work with machine work, which can save labor costs, reduce the complexity of manual operation, and improve the efficiency of making a high-precision map while ensuring high precision.

[0045] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0046] Other features, objects, and advantages of the present disclosure will become more apparent from the following detailed description of the non-limiting embodiments, taken in conjunction with the accompanying drawings. In the drawings:

[0047] Figure 1 An implementation flowchart of a contour drawing method of a ground feature according to an embodiment of the present disclosure is shown;

[0048] Figure 2 A structural schematic diagram of an electronic device suitable for implementing the contour drawing method of a ground feature according to an embodiment of the present disclosure is shown;

[0049] Figure 3 A data flow schematic diagram of the contour drawing method of a ground feature according to an embodiment of the present disclosure in the execution scenario of a data collection vehicle is shown;

[0050] Figure 4 This diagram illustrates an interactive scenario when a method for drawing the outline of ground features according to an embodiment of the present disclosure is executed in a data production center.

[0051] Figures 5(a)-5(c) This illustration shows the effect of manually drawn lines, defined areas, and outer contours for road traffic signs in one embodiment of the present disclosure.

[0052] Figure 6 A schematic diagram of a scene in which the outer contour of a ground element is drawn manually according to an embodiment of the present disclosure is shown.

[0053] Figure 7 A structural block diagram of a ground feature contour drawing apparatus according to an embodiment of the present disclosure is shown. Detailed Implementation

[0054] In the following, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement them. Furthermore, for clarity, portions unrelated to the description of the exemplary embodiments have been omitted from the drawings.

[0055] In this disclosure, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of features, figures, steps, behaviors, components, parts or combinations thereof disclosed in this specification, and do not preclude the possibility of the presence or addition of one or more other features, figures, steps, behaviors, components, parts or combinations thereof.

[0056] It should also be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0057] The details of the embodiments of this disclosure are described in detail below through specific examples.

[0058] Figure 1 A flowchart illustrating an implementation of a method for drawing the outline of ground features according to an embodiment of the present disclosure is shown.

[0059] like Figure 1 As shown, the method for drawing the outline of this ground feature includes the following steps S101-S105:

[0060] In step S101, the operator draws lines. In this step, the operator can draw a rough outline of the ground elements on the point cloud image (precision is not required for this drawn line). The point cloud data within this rough outline corresponds to the point cloud data of the area where the ground element is located. It should be noted that in this embodiment, the operator drawing the lines is not limited to a human operator; it can also be a machine-implemented operator.

[0061] In this embodiment, the ground element can be a nonlinear ground element on the road, that is, the ground element targeted by the embodiment of the present disclosure can be a planar element, which can include a planar traffic sign on the road surface except for lane lines, such as a direction arrow sign (see FIG. 5(a)), a road symbol sign, a guide rail, a zebra crossing, and the like. Figures 5(a)-5(c)

[0062] The point cloud image can be an image formed by point cloud data, and the point cloud data can be a set of three-dimensional point data of an actual road surface collected by a three-dimensional laser scanning instrument.

[0063] In a conventional ground element drawing method, the shape points of the ground element need to be manually drawn, but this is difficult to operate in a three-dimensional scene and has low production efficiency.

[0064] Therefore, in the embodiment of the present disclosure, an operator pre-defines the approximate range of the ground element to be drawn on the point cloud image, which is referred to as the defined range. The defined range provided by the operator can be seen in the white box shown in FIG. 5(b). It should be noted that the ground element to be drawn is a nonlinear ground element. After the operator provides the defined range of the ground element to be drawn on the point cloud image, the outer contour of the ground element can be accurately generated based on the point cloud data on the point cloud image, the defined range, and the reflection intensity of the point cloud data. This method improves the contour drawing efficiency of the ground element, reduces the difficulty of manual operation while ensuring the drawing accuracy, and saves labor costs.

[0065] In this embodiment, the defined range can be a range of all point cloud data corresponding to the ground element to be drawn on the point cloud image, for example, the defined range can be a point cloud frame that can frame the ground element on the point cloud image, for example, the point cloud frame can be a rectangle.

[0066] In some embodiments, the point cloud image can be displayed to the operator through a browser or other application interface. The operator can define a range around the ground element to be drawn based on the editing function provided by the browser or other application interface. The defined range can include all point cloud data of the ground element. In other embodiments, a machine can also be used as an operator to define a range on the point cloud image.

[0067] In step S102, a point cloud frame is generated based on the line drawn by the operator, and the point cloud data in the point cloud frame is calculated. That is, in this step, based on the defined range given by the operator through the line drawing, the point cloud data in the point cloud image enclosed in the defined range is determined.

[0068] ​The following describes a method for manually drawing lines on a point cloud image and determining the point cloud data enclosed within a defined area based on these manually drawn lines:

[0069] 2.1 Draw a line manually along the ground element to determine both the orientation of the ground element and the approximate range of the point cloud frame.

[0070] 2.2 Generate a bounding box for the point cloud data based on line drawing and expansion.

[0071] 2.3 Point cloud collision detection is performed using the octree spatial structure of the point cloud frame and point cloud data to roughly determine the point cloud data within the point cloud frame.

[0072] 2.4 The point cloud frame and the roughly determined point cloud data are sent as parameters to the WebAssembly browser in order to generate accurate outlines of ground elements.

[0073] It should be noted that the above is only one possible implementation method. The embodiments disclosed herein are not limited to the above implementation method, and other implementation methods can also be used. For example, a simple machine learning model can be trained to draw lines on the point cloud image to generate point cloud bounding boxes that surround the point cloud data containing the ground elements. In addition, the point cloud bounding boxes and the coarsely determined point cloud data are not limited to being input into the WebAssembly browser for precise contour generation, but can also be input into other similar software, or the precise contour generation process can be implemented through programming.

[0074] In step 103, the road plane is calculated based on the point cloud data within the delineated area.

[0075] After determining all point cloud data within the demarcated area, since point cloud data is three-dimensional data while ground features on the road surface are planar graphics, i.e., two-dimensional graphics, in order to accurately determine the outer contour of the ground features, a road plane is first fitted based on all point cloud data within the demarcated area. In some embodiments, the least squares method can be used to fit the point cloud plane, which is the road plane.

[0076] In some embodiments, a road plane can be fitted based on the point cloud data selected by the point cloud frame; the following describes an implementation process of fitting a road plane using a WebAssembly browser.

[0077] 3.1 Since the incoming point cloud data is relatively dense, it can be downsampled first to reduce the amount of computation.

[0078] 3.2 For downsampled point cloud data, accurate collision detection between the point cloud and the buffer frame can be performed based on the buffer frame, and only the point cloud data within the buffer frame can be retained.

[0079] 3.3 Use the least square method to fit the plane of the point cloud, i.e. the road plane.

[0080] In step 104, the outer contour of the ground element is calculated.

[0081] The point cloud data can be projected onto the road plane, and a gray value can be obtained based on the reflection intensity of each point in the point cloud data. Based on the gray value and the projection point coordinates of the point cloud data on the road plane, the point cloud data within the delineated range can be transformed into a gray image. That is, each pixel point in the gray image corresponds to a point in the point cloud data projected onto the road plane, and the pixel value corresponds to the reflection intensity of the point. The higher the reflection intensity, the higher the pixel value. The lower the reflection intensity, the lower the pixel value.

[0082] Since the ground element on the road surface is usually formed using a material with a lighter color compared to the asphalt on the road surface, the reflection intensity of the ground element is higher than that of the road surface. Therefore, in the gray image obtained in this way, the gray value of the area where the ground element is located is larger than that of the road surface area. Therefore, the image block contour of the foreground content in the gray image can be calculated based on this difference in gray value.

[0083] It can be understood that the foreground content in the gray image can be a component of the ground element, and the background area outside the foreground content corresponds to the road surface area.

[0084] Referring to FIG. 5(b), the ground element "right arrow" on the road surface is white, and the point cloud image in the white box is the manually delineated range. After being mapped into a gray image, the color of the road surface part is darker, and the reflection intensity is lower. Therefore, after being mapped into a gray image, the gray value is smaller. The color of the white arrow part is lighter, and the reflection intensity is higher. Therefore, after being mapped into a gray image, the gray value is larger. Therefore, based on the gray value in the gray image, the foreground content corresponding to the white arrow part and the background area corresponding to the road surface can be distinguished. Considering the case of discoloration due to wear and tear on the white arrow, as shown in FIG. 5(b), the straight line segment part on the white arrow, some areas in the middle are no longer white due to discoloration, but have a darker color. Therefore, after being mapped into a gray image, these discolored areas will also be identified as road surface areas. Therefore, the finally identified foreground content is not a complete white arrow-shaped area, but multiple white image blocks separated by the road surface. In the embodiments of the present disclosure, the contours of the multiple image blocks corresponding to the identified foreground content need to be further processed to obtain the complete contour of the ground element.

[0085] In some embodiments, the outer contour of the multiple image blocks can be determined, and the outer contour is determined as the complete outer contour of the ground element.

[0086] In some embodiments, after the image block outline corresponding to the foreground content is determined in the grayscale image, the outer contour of the ground element can be rendered in the point cloud data based on the vector data of the image block outline, as shown in the white rectangular frame in FIG. 5(c) which is the outer contour of the white arrow on the road, and the four vertex coordinates on the white rectangular frame are the vector data corresponding to the outer contour.

[0087] In some embodiments, the image block outline and the outer contour of the ground element can both be a circumscribed frame, and the vector data of the image block outline can include the vertex coordinates of the circumscribed frame. Therefore, the vertex coordinates and other vector data of the outer contour of the ground element can be ultimately determined based on the vertex coordinates and other data of the circumscribed frame of each image block outline.

[0088] Therefore, in one optional implementation of the present embodiment, the step S104 of calculating the outer contour of the ground element further includes the following steps:

[0089] constructing a road plane based on the point cloud data within the circumscribed range;

[0090] constructing a grayscale image of the point cloud data within the circumscribed range projected on the road plane based on the reflection intensity of the point cloud data;

[0091] determining the image block outline of the foreground content in the grayscale image;

[0092] drawing the outer contour of the ground element in the point cloud image based on the image block outline.

[0093] The following describes an implementation of calculating the outer contour of the ground element. It can be understood that the calculation of the outer contour of the ground element in the present embodiment is not limited to the following implementation, as long as it is based on the point cloud data within the circumscribed range to construct a road plane, based on the reflection intensity of the point cloud data to construct a grayscale image of the point cloud data within the circumscribed range projected on the plane, and determine the image block outline including the foreground content in the grayscale image, and based on these image block outlines to calculate the outer contour of the ground element, all of which are within the scope disclosed in the present embodiment.

[0094] 4.1 Project all the point clouds in the buffer frame onto the road plane;

[0095] 4.2 Affine the point clouds in the road plane to a two-dimensional image, and map the reflection intensity of the point clouds to the interval of 0-255 as the color value of the corresponding pixel on the two-dimensional image to form a two-dimensional image;

[0096] 4.3 Since individual point cloud reflection intensity affects as noise, Gaussian filtering can be performed on the two-dimensional image to remove noise points;

[0097] 4.4 Calculate the content outline in the 2D image. Because the reflectivity of the ground element itself and the reflectivity of the ground point cloud are usually different, the outline of the ground element can be accurately calculated. In addition, considering the actual wear of the ground element on the road surface, the size of the content outline can be judged. Outlines with a larger area can be superimposed to calculate the complete outline of the ground element. The calculated complete outline is the vector data on the 2D image, that is, the coordinates of the boundary points of the complete outline. For example, when the complete outline is represented by a rectangular border, the vector data can be the four vertices of the rectangular border. As shown in Figure 5(c), the white line box is the complete outline of the ground element "right turn arrow". Its four vertices are the vector data obtained by superimposing the outlines of multiple image blocks corresponding to the foreground content.

[0098] In step 105, the outer contour of the ground element is rendered.

[0099] The following describes one way to render the outer contour of ground elements.

[0100] 5.1 Rendering vector data based on the calculated outer contour of ground elements. Considering that the calculated vector data is data from a two-dimensional image, the vector data can be transformed into a point cloud image through an inverse affine transformation, and then rendered on the point cloud image based on the transformed vector data.

[0101] 5.2 For manual or other external objects, it is only necessary to verify whether the outer contour meets the requirements.

[0102] Figure 2 A schematic diagram of the structure of an electronic device suitable for implementing the contour drawing method of ground features according to embodiments of the present disclosure.

[0103] like Figure 2 As shown, the electronic device 200 includes a processing unit 201, which can be implemented as a CPU, GPU, FPGA, NPU, or other processing unit. The processing unit 201 can execute various processes according to any of the methods described above in this disclosure, based on a program stored in the read-only memory (ROM) 202 or a program loaded from the storage portion 208 into the random access memory (RAM) 203. The RAM 203 also stores various programs and data required for the operation of the electronic device 200. The processing unit 201, ROM 202, and RAM 203 are interconnected via a bus 204. An input / output (I / O) interface 205 is also connected to the bus 204.

[0104] The following components are connected to the I / O interface 205: an input section 206 including a keyboard, a mouse, etc.; an output section 207 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 208 including a hard disk, etc.; and a communication section 209 including a network interface card such as a LAN card, a modem, etc. The communication section 209 performs communication processing via a network such as the Internet. A drive 210 is also connected to the I / O interface 205 as necessary. A removable medium 211 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 210 as necessary, so that a computer program read out therefrom is installed in the storage section 208 as necessary.

[0105] The electronic device described above can be a device for implementing any of the methods described above in the embodiments of the present disclosure. The device can be a client device or a cloud device.

[0106] Figure 3 A data flow diagram of an execution scenario of the contour drawing method of a ground element according to an embodiment of the present disclosure is shown. As shown in Figure 3 , a data collection vehicle can be provided with a client device and a laser radar. The client device can be implemented as an electronic device as shown in Figure 2 . The laser radar can collect point cloud data of a road and provide the point cloud data to the client device.

[0107] The client device displays a point cloud image corresponding to the point cloud data on a client interface for a human operator to draw a line. The human operator circumscribes a range of a ground element to be drawn on the point cloud image by a drawing circle, and the client executes the method in the above embodiment based on the acquired circumscribed range to automatically determine an accurate contour of the ground element on the point cloud image. Based on the accurate contour, drawing data of the ground element, such as point cloud data within the accurate contour and other related data, can be determined. The drawing data can be sent to a map server, which can store the drawing data as part of map data. When using the map data to draw a map, the drawing data can be used to draw the ground element at a corresponding position on the map.

[0108] Figure 4 A diagram of an interaction scenario when the contour drawing method of a ground element according to an embodiment of the present disclosure is executed in a data production center is shown. As shown in Figure 4 , a cloud device can be a device provided in a data production center. The cloud device can be implemented by an electronic device as shown in Figure 2 . A data collection device, such as a data collection vehicle equipped with a laser radar, acquires point cloud data of a road and provides the point cloud data to the cloud device.

[0109] The cloud device sends a point cloud image corresponding to the point cloud data to a client device used by a human operator. The point cloud image is displayed on the client device.

[0110] The human operator circulates a range of the ground element to be drawn on the point cloud image. The client device determines the accurate contour of the ground element on the point cloud image based on the circled range, and determines the drawing data of the ground element based on the accurate contour. The drawing data is sent to a map server, and the map server stores the drawing data as part of the map data. When the map data is used to draw a map, the drawing data is used to draw the ground element at a corresponding position of the map.

[0111] In the process of drawing the outer contour of the ground element, the embodiments of the present disclosure assist in drawing the approximate range of the ground element to be drawn by an external object such as a human or a machine model running machine, etc. Based on the approximate range and the point cloud image, all point cloud data contained in the approximate range is automatically determined, all point cloud data contained in the approximate range is converted into a grayscale image through the reflection intensity of the point cloud data, the image block contour with foreground content is determined based on the grayscale image, and the outer contour of the ground element is rendered on the point cloud image based on the image block contour. In the process of making a high-precision map, the embodiments of the present disclosure draw the outer contour of the ground element on the point cloud image by a semi-automatic drawing method through the combination of the external object and the machine. The outer contour of the ground element can be drawn while ensuring high precision, saving labor cost, reducing the complexity of manual operation, and improving the efficiency of making a high-precision map.

[0112] In an optional implementation of the present embodiment, the step of drawing the outer contour of the ground element in the point cloud image based on the image block contour in step S104 further includes the following steps:

[0113] When the image block contour includes multiple image block contours, the vector data of the outer contour of the ground element is determined based on the image block contour with an area greater than or equal to a preset area threshold;

[0114] The outer contour of the ground element is rendered on the point cloud image based on the vector data.

[0115] In the optional implementation, due to the fact that the graphics of the ground element on the actual road may be worn, such as unclear boundary, and a certain block of the road surface is displayed due to wear. Therefore, after the point cloud data in the delineated range is converted into a grayscale image, the image block contour determined from the grayscale image is not necessarily the complete contour of the ground element, but may be multiple image block contours after segmentation in the worn area. Therefore, in this case, the complete outer contour of the ground element can be obtained by superimposing the multiple image block contours.

[0116] In addition, considering that there may be some stains around the ground element, or other reasons cause the image block contour to be detected in the grayscale image, therefore, the image block contour with an area less than a preset threshold can also be removed by the area size, and only the image block contour with an area greater than or equal to the preset threshold is retained to obtain the complete outer contour of the ground element by superimposition.

[0117] In the grayscale image, the image block contour can be represented by vector data, for example, in the case of a rectangular image block contour, the coordinates of the four points of the rectangle can be used to represent it. Superimposing multiple image block contours is actually superimposing the coordinates of the four points corresponding to the multiple image block contours. For example, the vector data of the superimposed complete outer contour can be obtained by taking the maximum value, that is, the coordinates of the four points of the complete outer contour are taken such that the complete outer contour is the maximum circumscribed rectangle of the multiple image block contours.

[0118] In an optional implementation of the embodiment, the delineated range includes a point cloud frame; the step of determining the point cloud data in the point cloud image enclosed in the delineated range further includes the following steps:

[0119] Perform point cloud collision detection on the octree spatial structure corresponding to the point cloud data on the point cloud image and the point cloud frame to obtain rough point cloud data having collision with the point cloud frame;

[0120] Downsample the rough point cloud data;

[0121] Perform accurate point cloud collision detection based on the spatial coordinates of the downsampled rough point cloud data and the point cloud frame;

[0122] Remove the rough point cloud data outside the point cloud frame based on the detection result, and determine the point cloud data in the delineated range based on the retained rough point cloud data.

[0123] In the optional implementation, the point cloud data is usually stored in the form of an octree spatial structure, each node of the octree spatial structure storing a block of point cloud data, and a block of point cloud data can include multiple points in the point cloud data. In order to reduce the amount of calculation, the point cloud collision detection is first performed on the point cloud block corresponding to the point cloud frame in the octree spatial structure and the circumscribed range, and the point cloud block that has collision with the point cloud frame is determined as the rough point cloud data.

[0124] It can be understood that the point cloud block includes multiple points in the point cloud data, and if the point cloud block has collision with the point cloud frame, it means that the point cloud block is located near the edge frame and can be a block of point cloud data across the edge frame. If it is a point cloud block across the edge frame, part of the point cloud data in the point cloud block is not in the circumscribed range. Therefore, the point cloud data in the circumscribed range cannot be directly determined based on the rough point cloud data, and the point cloud data in the point cloud block across the edge frame and located outside the point cloud frame also needs to be removed.

[0125] Therefore, the disclosed embodiments also perform a precise point cloud collision detection. In the precise point cloud collision detection process of the point cloud, the point cloud collision is performed on all points in the point cloud block that has collision with the point cloud frame based on the spatial coordinates of the points and the point cloud frame, and a detection result is determined. The detection result can include precise point cloud data that has collision with the point cloud frame, and based on the precise point cloud data, point cloud data located at the boundary of the point cloud frame and point cloud data located inside the point cloud frame can be determined. The point cloud data located at the boundary of the point cloud frame and the point cloud data located inside the point cloud frame can be determined as the point cloud data in the circumscribed range.

[0126] In an optional implementation of the present embodiment, the step of determining the image block contour of the foreground content in the grayscale image further includes the following steps:

[0127] Based on the difference in reflection intensity between the point cloud data corresponding to the road surface and the point cloud data corresponding to the ground element in the grayscale image, the image block contour of the foreground content in the grayscale image is determined.

[0128] In the optional implementation, as described above, the pixel value of each pixel point in the grayscale image is determined based on the reflection intensity of the corresponding point cloud data. According to common sense, the color of the ground element in the actual road is lighter than the color of the road surface, and the reflection intensity of the ground element is greater than the reflection intensity of the road surface. Therefore, the reflection intensity of the point cloud data on the ground element and the road surface in the point cloud data is different, and after the grayscale image is generated through the affine transformation, the difference in reflection intensity is reflected in the grayscale value. The part with greater reflection intensity forms the foreground content in the grayscale image, and the part with smaller reflection intensity forms the background in the grayscale image. Therefore, the image block contour where the foreground content is located can be determined from the grayscale image based on the grayscale image and the difference in pixel value corresponding to the difference in reflection intensity.

[0129] In an optional implementation of the embodiment, the method further comprises the following steps:

[0130] displaying the point cloud image;

[0131] receiving a line drawing range and an outward expansion direction on the point cloud image;

[0132] determining the circumscribed range surrounding the ground element to be drawn based on the line drawing range and the outward expansion direction, the circumscribed range comprising a preset shape of a frame obtained by expanding a certain distance in the outward expansion direction within the line drawing range.

[0133] In the optional implementation, the point cloud image of the road can be displayed on the display interface, and the drawing personnel can draw a line on the ground element to be drawn contour using the editing function provided on the display interface. The direction of the line drawing can be along the length direction of the ground element, and the length of the line drawing can be greater than or equal to the total length of the ground element. Based on the line drawing operation of the drawing personnel, a line drawing range on the ground element can be obtained, which includes the starting point, the ending point and the line length of the line drawing. In addition, the drawing personnel can also specify the outward expansion direction after completing the line drawing, for example, the outward expansion direction can be perpendicular to the direction of the line drawing. After detecting the line drawing and the outward expansion direction of the drawing personnel, a frame of a preset shape can be obtained by expanding a certain distance in the outward expansion direction within the line drawing range. In some embodiments, the outward expansion distance can be controlled by the drawing personnel, or a default outward expansion distance can be set based on the pre-set, which can be set based on the actual application, and details are not described herein. It can be understood that the point cloud image can also be provided to a machine model or other external object, and the external object can provide the line drawing range and the outward expansion direction based on the corresponding manner.

[0134] In some embodiments, the drawing personnel can specify an outward expansion direction, and the actual outward expansion occurs in the outward expansion direction and the opposite direction of the outward expansion direction, so that a frame completely surrounding the ground element can be obtained.

[0135] In some embodiments, after obtaining an initial range based on the line drawing range and the outward expansion direction of the drawing personnel, the drawing personnel can also use the editing function provided on the display interface to modify the initial range, so that the initial range can completely surround the ground element while having the smallest area.

[0136] Figures 5(a)-5(c)This diagram illustrates the effect of manually drawn lines, defined areas, and outer contours for road traffic signs according to an embodiment of this disclosure. As shown in Figure 5(a), in the 3D point cloud image, a vertical line is manually drawn on the right-turn road traffic sign line using a mouse. The outward expansion direction is specified as perpendicular to this vertical line by dragging the mouse, resulting in the defined area shown in Figure 5(b). Figure 5(b) shows that the manually drawn defined area surrounds the right-turn road traffic sign, but the accuracy of this defined area is poor; the border is far from the outer boundary of the right-turn road traffic sign line and is significantly larger than the outer frame of the right-turn road traffic sign line. After processing according to the embodiment of this disclosure, the outer contour shown in Figure 5(c) is obtained. Figure 5(c) shows that the accuracy of this outer contour is high, and it perfectly matches the outer frame of the left-turn road traffic sign line.

[0137] Figure 6 This diagram illustrates a scene where the outline of a ground element is drawn using human interaction according to an embodiment of this disclosure. Figure 6 As shown, a point cloud image is displayed in a browser. Users draw lines and specify outward expansion directions on the point cloud image through an interactive input interface. The browser can automatically generate a delineated area based on the drawn lines and outward expansion directions. If the automatically generated delineated area is correct, it can be submitted to the backend for processing. Operators can also manually adjust the delineated area. After receiving the delineated area and point cloud data, the backend performs 3D geometric calculations to determine the point cloud data within the delineated area. It then performs planar fitting on the point cloud data and projects it onto the fitted plane. The projected data is then used to obtain a 2D image. The 2D image is processed to obtain the outer contours of the ground elements. These contours are automatically drawn on the point cloud image and then displayed on the browser interface.

[0138] According to an embodiment of the present disclosure, a location-based service provision method is provided. This method uses the above-described method for drawing the outline of ground elements to draw the outer contour of ground elements on a map, and provides location-based services to the service recipient using a map with the outline of the ground elements drawn on it. The location-based services include one or more of navigation, map rendering, and route planning.

[0139] In this embodiment, the location-based service providing method can be executed on a terminal, which can be a mobile phone, an iPad, a computer, a smart watch, an autonomous vehicle, etc. When making a map, such as a high-definition map, the contour drawing method of the ground element described above can be used to draw the outer contour of the ground element. It should be noted that the high-definition map can be a map with accurate vehicle position information and rich road element data information. The high-definition map has detailed ground expression, and the road network data expressed in the high-definition map includes accurate lane shape, and also includes detailed data such as slope, curvature, heading, elevation, roll, etc. of each lane. Therefore, for the ground element in the high-definition map, the contour drawing method of the ground element described above can be used for drawing.

[0140] In the process of making a high-definition map, the outer contour of the ground element of any road can be drawn according to the embodiments of the present disclosure. The high-definition map can be provided to a served object to provide more accurate location services, such as navigation services, path planning services, and / or map rendering services, etc.

[0141] The served object can be a mobile phone, an iPad, a computer, a smart watch, an autonomous vehicle, a robot, etc. Before navigating, planning a path, or rendering a road on a map for the served object, the ground element on the high-definition map can be made based on the method described above, and then the service is provided based on the high-definition map when navigating, planning a path, or rendering a map. For specific details, please refer to the description of the attribute determination method of the target road element described above, which will not be described here.

[0142] The following is an apparatus embodiment of the present disclosure, which can be used to execute the method embodiments of the present disclosure.

[0143] Figure 7 A structural block diagram of a contour drawing device of a ground element according to an embodiment of the present disclosure is shown. The device can be realized as part or all of an electronic device through software, hardware, or a combination of the two. As shown, the contour drawing device of the ground element includes: Figure 7

[0144] The first acquisition module 701 is configured to acquire a circumscribed range in a point cloud image;

[0145] The first determination module 702 is configured to determine point cloud data in the point cloud image that is enclosed in the circumscribed range;

[0146] The first construction module 703 is configured to construct a road plane based on the point cloud data in the circumscribed range;

[0147] The second construction module 704 is configured to construct a grayscale image of the point cloud data in the circumscribed range after projection on the road plane based on the reflection intensity of the point cloud data.​

[0148] The second determination module 705 is configured to determine an image block contour of foreground content in the grayscale image; wherein the foreground content is a component of the ground element.

[0149] The drawing module 706 is configured to draw an outer contour of the ground element in the point cloud image based on the image block contour.

[0150] In this embodiment, the ground element can be a nonlinear ground element on the road, that is, the ground element to which the embodiment of the present disclosure is directed can be a planar element, and the planar element can include planar traffic signs on the road surface except lane lines, such as direction arrow signs, road symbol signs, guide strips, and zebra crossings.

[0151] The point cloud image can be an image formed by point cloud data, and the point cloud data can be a set of three-dimensional point data of an actual road surface collected by a three-dimensional laser scanning instrument.

[0152] In a conventional ground element drawing method, the shape points of the ground element need to be manually drawn, but this is more difficult to operate in a three-dimensional scene and has low production efficiency.

[0153] Therefore, the embodiment of the present disclosure adopts a man-machine combined manner, manually pre-defines the approximate range of the ground element to be drawn on the point cloud image, which is referred to as a defined range. It should be noted that the ground element to be drawn is a nonlinear ground element. After the man provides the defined range of the ground element to be drawn on the point cloud image, the machine can accurately generate the outer contour of the ground element based on the point cloud data on the point cloud image, the defined range, and the reflection intensity of the point cloud data. This method improves the contour drawing efficiency of the ground element, reduces the difficulty of manual operation while ensuring the drawing accuracy, and saves the labor cost.

[0154] In this embodiment, the defined range can be a range including all point cloud data corresponding to the ground element to be drawn on the point cloud image, for example, the defined range can be a frame such as a rectangular frame that can frame the ground element on the point cloud image.

[0155] The point cloud image can be displayed to the drawing personnel through a browser or other application interface, and the drawing personnel can define a range around the ground element to be drawn based on the editing function provided by the browser or other application interface. All point cloud data of the ground element can be included in the defined range.

[0156] Based on the circumscribed range, all point cloud data contained in the circumscribed range can be determined. In some embodiments, the point cloud data surrounded in the circumscribed range can be obtained through point cloud collision detection, and then the point cloud data in the circumscribed range can be converted into a grayscale image, and the grayscale value in the grayscale image can be determined based on the reflection intensity of the point cloud data. The stronger the reflection intensity is, the higher the corresponding grayscale value is, and the lower the reflection intensity is, the lower the corresponding grayscale value can be.

[0157] After determining all point cloud data located in the circumscribed range, since the point cloud data is three-dimensional data and the ground element on the road surface is a planar graph, in order to accurately determine the outer contour of the ground element, a road plane is first fitted based on all point cloud data in the circumscribed range. In some embodiments, the least square method can be used to fit the point cloud plane, which is the road plane.

[0158] Since the ground element on the road surface is usually formed by using a material with a lighter color compared to the asphalt on the road surface, the reflection intensity of the ground element is higher than that of the road surface. Therefore, in the grayscale image obtained in this way, the grayscale value of the area where the ground element is located is larger than that of the road surface area. Therefore, the image block contour of the foreground content in the grayscale image can be calculated based on the difference in grayscale value.

[0159] It can be understood that the foreground content in the grayscale image corresponds to the ground element, and the background area outside the foreground content corresponds to the road surface area.

[0160] After determining the image block contour corresponding to the foreground content in the grayscale image, the outer contour of the ground element can be rendered in the point cloud data based on the vector data of the image block contour. In some embodiments, the image block contour and the outer contour of the ground element can both be a circumscribed box, and the vector data of the image contour can include the vertex coordinates of the circumscribed box, etc.

[0161] In the process of drawing the outer contour of the ground element, the embodiments of the present disclosure manually assist in drawing the approximate range of the ground element to be drawn, automatically determine all point cloud data contained in the approximate range based on the approximate range and the point cloud image, convert all point cloud data contained in the approximate range into a grayscale image through the reflection intensity of the point cloud data, determine the image block contour with foreground content based on the grayscale image, and render the outer contour of the ground element on the point cloud image based on the image block contour. The embodiments of the present disclosure combine manual work with machine work to draw the outer contour of the ground element on the point cloud image in a semi-automatic drawing manner when making a high-precision map, which can save labor costs, reduce the complexity of manual operation, and improve the efficiency of making a high-precision map while ensuring high precision.

[0162] In an optional implementation of the embodiment, the second construction module comprises:

[0163] a projection submodule configured to project the point cloud data within the circumscribed range to the road plane;

[0164] a generation submodule configured to generate the grayscale image based on the reflection intensity of the projected point cloud data.

[0165] In the optional implementation,

[0166] The point cloud data can be projected to the road plane, and a grayscale value can be obtained based on the reflection intensity of each point in the point cloud data. Based on the grayscale value and the projection point coordinates of the point cloud data on the road plane, the point cloud data within the circumscribed range can be transformed into a grayscale image in an affine manner. That is, each pixel point in the grayscale image corresponds to a point in the point cloud data projected onto the road plane, and the pixel value corresponds to the reflection intensity of the point. The higher the reflection intensity, the higher the pixel value, and the lower the reflection intensity, the lower the pixel value.

[0167] In an optional implementation of the embodiment, the drawing module comprises:

[0168] a first determination submodule configured to, when the image block outlines comprise multiple image block outlines, determine the vector data of the outer contour of the ground element based on the image block outlines with an area greater than or equal to a preset area threshold;

[0169] a rendering submodule configured to render the outer contour of the ground element on the point cloud image based on the vector data.

[0170] In the optional implementation, the pattern of the ground element on the actual road may be worn, such as unclear boundaries, and some blocks in the middle may show the road surface due to wear. Therefore, after the point cloud data within the circumscribed range is converted into a grayscale image, the image block outlines determined from the grayscale image are not necessarily the complete contour of the ground element, but may be multiple image block outlines after segmentation in the worn area. Therefore, in this case, the complete outer contour of the ground element can be obtained by superimposing multiple image block outlines.

[0171] In addition, considering that there may be some stains around the ground element, or other reasons leading to the detection of image block outlines in the grayscale image, the image block outlines with an area less than a preset threshold can also be removed by the area size of the image block outlines, and only the image block outlines with an area greater than or equal to the preset threshold are used for superposition to obtain the complete outer contour of the ground element.

[0172] In the grayscale image, the image block contour can be represented by vector data, for example, in the case of a rectangular image block contour, the coordinates of the four points of the rectangle can be used for representation. Superimposing a plurality of image block contours is actually superimposing the coordinates of the four points corresponding to the plurality of image block contours. For example, the vector data of the complete outer contour after superposition can be obtained by taking the maximum value, that is, the coordinates of the four points of the complete outer contour are taken such that the complete outer contour is the maximum circumscribed rectangle of the plurality of image block contours.

[0173] In an optional implementation of the embodiment, the bounding range includes a point cloud box; the first determining module includes:

[0174] A first detection submodule is configured to perform point cloud collision detection on the octree spatial structure corresponding to the point cloud data on the point cloud image and the point cloud box to obtain rough point cloud data having collision with the point cloud box;

[0175] A downsampling submodule is configured to downsample the rough point cloud data;

[0176] A second detection submodule is configured to perform accurate point cloud collision detection based on the spatial coordinates of the rough point cloud data after downsampling and the point cloud box;

[0177] A second determining submodule is configured to exclude the rough point cloud data outside the point cloud box based on the detection result and determine the point cloud data within the bounding range based on the retained rough point cloud data.

[0178] In the optional implementation, the point cloud data is usually stored in the form of an octree spatial structure, each node of the octree spatial structure storing a block of point cloud data, and a block of point cloud data can include a plurality of points in the point cloud data. In order to reduce the amount of calculation, the octree spatial structure corresponding to the point cloud box of the bounding range is used for point cloud collision detection in the embodiment of the disclosure, and the point cloud block having collision with the point cloud box is determined as rough point cloud data.

[0179] It can be understood that a point cloud block includes a plurality of points in the point cloud data, and if the point cloud block has collision with the point cloud box, it means that the point cloud block is located near the edge frame and can be a block of point cloud data across the edge frame. If it is a point cloud block across the edge frame, part of the point cloud data in the point cloud block is not within the bounding range. Therefore, the rough point cloud data cannot be directly used to determine the point cloud data within the bounding range, and the point cloud data outside the point cloud box in the point cloud block across the edge frame also needs to be excluded.

[0180] To this end, the embodiments of the present disclosure also perform a precise point cloud collision detection. In the precise point cloud collision detection, for a point cloud block having a collision with the point cloud frame, point cloud collision is performed based on the spatial coordinates of all points in the point cloud block and the point cloud frame, and a detection result is determined. The detection result can include precise point cloud data having a collision with the point cloud frame, based on which point cloud data located at the boundary of the point cloud frame and point cloud data located inside the point cloud frame can be determined, and the point cloud data located at the boundary of the point cloud frame and the point cloud data located inside the point cloud frame can be determined as point cloud data in the circumscribed range.

[0181] In an optional implementation of the embodiment, the second determination module comprises:

[0182] The third determination submodule is configured to determine the image block outline of the foreground content in the grayscale image based on the reflection intensity difference between the point cloud data corresponding to the road surface and the point cloud data corresponding to the ground element in the grayscale image.

[0183] In the optional implementation, as described above, the pixel value of each pixel point in the grayscale image is determined based on the reflection intensity of the corresponding point cloud data. According to common sense, the color of the ground element in the actual road is lighter than the color of the road surface, and the reflection intensity of the ground element is greater than the reflection intensity of the road surface. Therefore, the reflection intensity of the point cloud data on the ground element and the road surface in the point cloud data is different, and after the grayscale image is generated through the affine transformation, the difference in the reflection intensity is reflected in the grayscale value. The part with greater reflection intensity forms the foreground content in the grayscale image, and the part with smaller reflection intensity forms the background in the grayscale image. Therefore, the image block outline where the foreground content is located can be determined from the grayscale image based on the grayscale image and the pixel value difference corresponding to the reflection intensity difference.

[0184] In an optional implementation of the embodiment, the device further comprises:

[0185] The display module is configured to display the point cloud image.

[0186] The receiving module is configured to receive a line drawing range and an outward expansion direction on the point cloud image.

[0187] The third determination module is configured to determine the circumscribed range of the ground element surrounding the to-be-drawn contour based on the line drawing range and the outward expansion direction, the circumscribed range comprising a preset shape of a frame obtained after expanding a certain distance in the line drawing range in the outward expansion direction.

[0188] In the optional implementation, the point cloud image of the road can be displayed on a display interface, and an editor provided on the display interface can be used by the drawing personnel to draw a line on the ground element to be drawn, the direction of the line can be along the length direction of the ground element, and the length of the line can be greater than or equal to the total length of the ground element. Based on the line drawing operation of the drawing personnel, a line drawing range on the ground element can be obtained, the line drawing range includes the starting point, the ending point and the line length of the line drawing. In addition, after the line drawing is completed, the drawing personnel can also specify an outward expansion direction of the line drawing, for example, the outward expansion direction can be perpendicular to the direction of the line drawing. After detecting the line drawing and the outward expansion direction of the drawing personnel, a frame of a preset shape can be obtained by expanding a certain distance in the outward expansion direction within the line drawing range. In some embodiments, the distance of the outward expansion can be controlled by the drawing personnel, or a default distance can be expanded based on a pre-set, which can be set based on the actual application, and details are not described herein.

[0189] In some embodiments, the drawing personnel can specify an outward expansion direction, and the actual outward expansion occurs in the outward expansion direction and the opposite direction of the outward expansion direction, so that a frame that completely surrounds the ground element can be finally obtained.

[0190] In some embodiments, after obtaining an initial range based on the line drawing range of the drawing personnel and the outward expansion direction, the drawing personnel can also use the editing function provided on the display interface to modify the initial range, so that the initial range can completely surround the ground element while having the smallest area.

[0191] According to an embodiment of the present disclosure, a location-based service providing device draws an outer contour of a ground element on a high-definition map by using the ground element contour drawing device, and provides a location-based service to a service object by using the high-definition map with the drawn outer contour of the ground element, the location-based service including one or more of navigation, map rendering, and route planning.

[0192] In the embodiment, the location-based service providing device can be executed on a terminal, and the terminal is a mobile phone, an iPad, a computer, a smart watch, an autonomous vehicle, etc. When the high-definition map is made, the contour drawing device of the ground element can be used to draw the outer contour of the ground element in a manual combination manner.

[0193] According to the embodiments of the present disclosure, the outer contour of the ground element can be drawn for any road to make a high-definition map, and the high-definition map can be provided to the service object to provide more accurate location services, such as navigation services, path planning services, and / or map rendering services, etc.

[0194] The served object can be a mobile phone, an iPad, a computer, a smart watch, an autonomous vehicle, a robot, etc. Before the served object is navigated, a path is planned, or a road on a map is rendered, the ground elements on the high-definition map can be made based on the above device, and then when the served object is navigated, a path is planned, or a road on a map is rendered, the service is provided based on the high-definition map. For details, please refer to the description of the attribute determination device of the target road element above, which will not be repeated here.

[0195] The flowcharts and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams can represent a module, a segment, or a portion of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in a different order than that shown in the figures. For example, two blocks noted in succession can actually be executed substantially concurrently, or they can sometimes be executed in reverse order, depending on the functionality involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by dedicated hardware-based systems that perform specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0196] The units or modules described in the embodiments of the present disclosure can be implemented by software or by hardware. The described units or modules can also be arranged in a processor, and the name of the unit or module does not constitute a limitation on the unit or module itself in some cases. As another aspect, the present disclosure also provides a computer readable storage medium, which can be a computer readable storage medium included in the device described in the above embodiments, or can exist separately from the device and not be assembled into the device. The computer readable storage medium stores one or more programs, which are executed by one or more processors to perform the methods described in the present disclosure.

[0197] The above description is merely preferred embodiments of the present disclosure and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or equivalent features without departing from the inventive concept. For example, the above features can be replaced with technical features disclosed in the present disclosure (but not limited to) having similar functions to form technical solutions.

Claims

1. A method of contouring a ground element, wherein, The ground feature is a non-linear planar ground feature on the road, and the method includes: Obtain the delineated area in the point cloud image, where the point cloud image refers to an image formed from point cloud data, and the delineated area is a point cloud frame generated based on the lines drawn by the operator; Determine the point cloud data within the defined area in the point cloud image; A road plane is fitted based on the point cloud data within the defined area; The point cloud data within the defined area is projected onto the road plane. Based on the reflection intensity of the projected point cloud data, a grayscale image of the point cloud data within the defined area projected onto the road plane is constructed through affine transformation. Determine the outline of image blocks containing foreground content in the grayscale image; wherein the foreground content is a component of the ground features; The outer contour of the ground feature in the point cloud image is drawn based on the image block contour.

2. The method of claim 1, wherein, Drawing the outer contour of the ground features in the point cloud image based on the image patch contour includes: When the image block contour includes multiple elements, the vector data of the outer contour of the ground feature is determined based on the image block contour whose area is greater than or equal to a preset area threshold. The outer contour of the ground feature is rendered on the point cloud image based on the vector data.

3. The method according to claim 1, wherein, The defined area includes the point cloud frame; Determining the point cloud data within the defined area in the point cloud image includes: The octree spatial structure corresponding to the point cloud data on the point cloud image is used to perform point cloud collision detection with the point cloud frame to obtain coarse point cloud data that collides with the point cloud frame. The coarse point cloud data is downsampled; Accurate point cloud collision detection is performed based on the spatial coordinates of the downsampled coarse point cloud data and the point cloud frame. Based on the detection results, the coarse point cloud data outside the point cloud frame is removed, and the point cloud data within the defined range is determined based on the retained coarse point cloud data.

4. The method according to claim 1, wherein, Determining the image block outline of foreground content in the grayscale image includes: Based on the difference in reflection intensity between the point cloud data corresponding to the road surface and the point cloud data corresponding to the ground features in the grayscale image, the image block outline of the foreground content in the grayscale image is determined.

5. The method according to claim 1, wherein, The method further includes: Display point cloud images; Receive the range and outward direction of the line drawn on the point cloud image; Based on the drawing line range and the outward expansion direction, the defined range of the ground elements surrounding the outline to be drawn is determined. The defined range includes a border of a preset shape obtained by extending a certain distance in the outward expansion direction within the drawing line range.

6. A location-based service provisioning method, wherein, include: Using the method described in any one of claims 1-5, the outer contours of ground elements on a map are drawn, and the map with the outer contours of the ground elements is used to provide location-based services to the served object, wherein the location-based services include one or more of navigation, map rendering, and route planning.

7. A device for drawing the outline of ground features, wherein, The ground feature is a non-linear planar ground feature on the road, and the device includes: The first acquisition module is configured to acquire a delineated area in a point cloud image, wherein the point cloud image refers to an image formed by point cloud data, and the delineated area is a point cloud frame generated based on lines drawn by the operator. The first determining module is configured to determine point cloud data in the point cloud image that surrounds the defined area; The first construction module is configured to fit a road plane based on point cloud data within the defined area; The second construction module is configured to project the point cloud data within the defined range onto the road plane, and based on the reflection intensity of the projected point cloud data, construct a grayscale image of the point cloud data within the defined range projected onto the road plane through affine transformation. The second determining module is configured to determine the outline of image blocks containing foreground content in the grayscale image; wherein the foreground content is a component of the ground features; The drawing module is configured to draw the outer contour of the ground features in the point cloud image based on the image block contour.

8. An electronic device, wherein, The method includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method of any one of claims 1-6.

9. A computer program product comprising computer instructions, wherein, When executed by a processor, the computer instructions implement the method described in any one of claims 1-6.

Images