CN114022857A

CN114022857A - Method, device, equipment and medium for extracting and classifying rod-shaped ground objects

Info

Publication number: CN114022857A
Application number: CN202111357916.9A
Authority: CN
Inventors: 马浩; 张攀科
Original assignee: Beijing Geo Vision Tech Co ltd
Current assignee: Beijing Geo Vision Tech Co ltd
Priority date: 2021-11-16
Filing date: 2021-11-16
Publication date: 2022-02-08

Abstract

The embodiment of the invention discloses a method, a device, equipment and a medium for extracting and classifying rod-shaped ground objects, wherein the method comprises the following steps: acquiring laser point cloud data corresponding to two sides of a vehicle driving road; detecting whether each laser point in the laser point cloud data is an edge point of a rod-shaped ground object, and extracting an edge point set corresponding to each rod-shaped ground object in two sides of a road; clustering and segmenting the edge point set, and determining an edge point subset corresponding to each rod-shaped ground object; performing space growth of laser points based on the edge point subset corresponding to each rod-shaped ground object, and determining the geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset; and determining the specific category corresponding to each rod-shaped ground object based on the geometric attribute information corresponding to each rod-shaped ground object. By the technical scheme of the embodiment of the invention, the rod-shaped ground objects in the laser point cloud can be extracted and subdivided, and the accuracy and efficiency of extraction and classification are improved.

Description

Method, device, equipment and medium for extracting and classifying rod-shaped ground objects

Technical Field

The embodiment of the invention relates to computer technology, in particular to a method, a device, equipment and a medium for extracting and classifying rod-shaped ground objects.

Background

Rod-shaped ground objects on two sides of a road are important infrastructures of a city, and information of the rod-shaped ground objects is accurately and quickly updated to provide efficient data support for smart cities, intelligent transportation, smart gardens and the like.

At present, the automatic extraction and classification methods of rod-shaped ground objects include: the characteristic that the rod-shaped ground object is cylindrical can be used for extraction; the trunk can also be automatically extracted by slicing point cloud data to detect an arc by utilizing the characteristic that the cross section of the trunk is close to a circle; the rod-shaped ground object can also be extracted by means of grid projection.

However, in the process of implementing the present invention, the inventor finds that at least the following problems exist in the prior art:

the rod-shaped ground objects measured on two sides of the road are various, and the laser points reflected by the rod-shaped ground objects come from the ground objects with the same or different geometric shapes, so that the distribution form of laser point cloud data in a three-dimensional space presents random discreteness, the laser point cloud data are more complex for a complex city containing a large number of rod-shaped ground objects, and the conditions of ground object shielding and point cloud missing commonly exist, so that the accuracy and the efficiency of the existing rod-shaped ground object extraction mode are lower, and the rod-shaped ground objects are not further subdivided, such as street lamps, telegraph poles, traffic sign poles and the like.

Disclosure of Invention

The embodiment of the invention provides a method, a device, equipment and a medium for extracting and classifying rod-shaped ground objects, which are used for improving the accuracy and efficiency of extracting and classifying rod-shaped ground objects in laser point cloud while extracting and subdividing the rod-shaped ground objects.

In a first aspect, an embodiment of the present invention provides a method for extracting and classifying rod-shaped ground objects, including:

acquiring laser point cloud data corresponding to two sides of a vehicle driving road;

detecting whether each laser point in the laser point cloud data is an edge point of a rod-shaped ground object, and extracting an edge point set corresponding to each rod-shaped ground object in two sides of a road;

clustering and segmenting the edge point set, and determining an edge point subset corresponding to each rod-shaped ground object;

performing space growth of laser points based on an edge point subset corresponding to each rod-shaped ground object, and determining geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset;

and determining the specific category corresponding to each rod-shaped ground object based on the geometric attribute information corresponding to each rod-shaped ground object.

In a second aspect, an embodiment of the present invention further provides an apparatus for extracting and classifying a rod-shaped ground object, including:

the laser point cloud data acquisition module is used for acquiring laser point cloud data corresponding to two sides of the acquired vehicle driving road;

the edge point detection module is used for detecting whether each laser point in the laser point cloud data is an edge point of a rod-shaped ground object or not and extracting an edge point set corresponding to each rod-shaped ground object in two sides of a road;

the clustering and partitioning module is used for clustering and partitioning the edge point set to determine an edge point subset corresponding to each rod-shaped ground object;

the geometric attribute information determining module is used for carrying out space growth of laser points based on the edge point subset corresponding to each rod-shaped ground object, and determining geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset;

and the category determining module is used for determining the specific category corresponding to each rod-shaped ground object based on the geometric attribute information corresponding to each rod-shaped ground object.

In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:

one or more processors;

a memory for storing one or more programs;

when executed by the one or more processors, cause the one or more processors to implement a method for extracting and classifying a rod-shaped feature as provided by any embodiment of the invention.

In a fourth aspect, the embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for extracting and classifying the rod-shaped feature according to any of the embodiments of the present invention.

The embodiment of the invention can rapidly and accurately extract the edge points of the rod-shaped ground objects in the laser point cloud data by detecting whether each laser point in the laser point cloud data is the edge point of the rod-shaped ground objects and extracting the edge point sets corresponding to the rod-shaped ground objects on two sides of a road by utilizing the edge characteristics, determines the edge point subset corresponding to each rod-shaped ground object by clustering and dividing the edge point sets, thereby realizing the materialization of the rod-shaped ground objects, performs the spatial growth of the laser point based on the edge point subset corresponding to each rod-shaped ground object, determines the geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset, and determines the specific category corresponding to each rod-shaped ground object based on the geometric attribute information corresponding to each rod-shaped ground object, thereby realizing the extraction and subdivision of the rod-shaped ground objects, and the accuracy and efficiency of extraction and classification are improved.

Drawings

Fig. 1 is a flowchart of a method for extracting and classifying rod-shaped ground objects according to an embodiment of the present invention;

FIG. 2 is a diagram illustrating an effect of an extracted trunk according to an embodiment of the present invention;

fig. 3 is an effect diagram of an extracted street lamp according to an embodiment of the present invention;

fig. 4 is a flowchart of a method for extracting and classifying rod-shaped ground objects according to a second embodiment of the present invention;

FIG. 5 is a diagram illustrating a pair of a left edge point and a right edge point according to a second embodiment of the present invention;

fig. 6 is a flowchart of a method for extracting and classifying rod-shaped ground objects according to a third embodiment of the present invention;

fig. 7 is a schematic structural diagram of a device for extracting and classifying rod-shaped ground objects according to a fourth embodiment of the present invention;

fig. 8 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.

Example one

Fig. 1 is a flowchart of a method for extracting and classifying rod-shaped features according to an embodiment of the present invention, which is applicable to extracting and classifying rod-shaped features on two sides of a road in laser point cloud data. The method can be executed by a device for extracting and classifying the rod-shaped ground object, which can be realized by software and/or hardware, and is integrated in electronic equipment, such as a vehicle. As shown in fig. 1, the method specifically includes the following steps:

and S110, acquiring the collected laser point cloud data corresponding to two sides of the vehicle driving road.

In particular, laser point cloud data of a vehicle driving road and both sides of the road may be collected using an on-board laser scanning system, such as LiDAR (Light detection and Ranging). The laser point cloud data can comprise various ground object point clouds on two sides of a road so as to extract rod-shaped ground objects.

S120, detecting whether each laser point in the laser point cloud data is an edge point of the rod-shaped ground object, and extracting an edge point set corresponding to each rod-shaped ground object in two sides of the road.

Wherein, the edge point of the rod-shaped ground object may refer to a laser point at a boundary position of the rod-shaped ground object. For example a laser spot located on the surface of a light pole. The set of edge points may be a set consisting of all edge points in the laser point cloud data, i.e. a set of edge points of all rod-shaped objects in both sides of the road.

Specifically, whether each laser point in the laser point cloud data is an edge point of the rod-shaped ground object or not is detected, the edge point can be effectively separated from ground points and building points on a road, and therefore an edge point set in the laser point cloud data can be extracted based on edge features of the rod-shaped ground object.

Exemplarily, S120 may include: detecting whether each laser point meets a preset edge point condition or not according to the forward distance and the backward distance corresponding to each laser point in the laser point cloud data, and taking the laser points meeting the preset edge point condition as edge points of the rod-shaped ground object; the forward distance refers to a three-dimensional distance between a current laser point and a previous laser point on the same scanning line; the backward distance refers to the three-dimensional distance between the current laser point and the next laser point on the same scan.

The preset edge point condition may be set in advance based on the edge position feature of the rod-shaped feature. For example, the rod-shaped feature is characterized by a tubular feature in the vertical direction, such that a laser point before the left edge point (i.e. the left boundary point) and a laser point after the right edge point (i.e. the right boundary point) of the rod-shaped feature both fall on the ground or other features, so that the forward distance and the backward distance of the left edge point of the rod-shaped feature are larger, the forward distance and the backward distance of the right edge point are smaller, and the preset edge point condition can be set as follows: the current forward distance corresponding to the current laser point is greater than a first preset distance, and the current backward distance is less than a second preset distance, or the current backward distance is greater than the first preset distance, and the current forward distance is less than the second preset distance. The first preset distance is larger than the second preset distance.

Specifically, the forward distance and the backward distance corresponding to each laser point may be determined based on the three-dimensional coordinate information of each laser point in the laser point cloud data. For example, the three-dimensional euclidean distance between the current laser point and the previous laser point on the same scanning line may be used as the current forward distance corresponding to the current laser point, and the three-dimensional euclidean distance between the current laser point and the next laser point on the same scanning line may be used as the current backward distance corresponding to the current laser point. Whether the laser point meets the preset edge point condition or not can be detected based on the forward distance and the backward distance corresponding to each laser point. For example, if it is detected that the current forward distance corresponding to the current laser point is greater than a first preset distance and the current backward distance is less than a second preset distance, or the current backward distance is greater than the first preset distance and the current forward distance is less than the second preset distance, it may be determined that the current laser point is an edge point of the rod-shaped feature, otherwise, it is determined that the current laser point is not an edge point of the rod-shaped feature, which may be a ground point on a road, a building point, or a middle point of the rod-shaped feature. By utilizing the obvious difference of the distance between the edge points on the two sides of the rod-shaped ground object and the adjacent points in the scanning line in the forward direction and the backward direction, the edge points and the non-edge points can be separated quickly and effectively, and the edge point set is extracted.

And S130, clustering and segmenting the edge point set, and determining an edge point subset corresponding to each rod-shaped ground object.

Wherein the subset of edge points may be a set consisting of all edge points of a single rod-shaped feature. Each rod-shaped feature corresponds to a subset of edge points.

Specifically, the edge point set may be segmented by a clustering method to obtain an edge point subset corresponding to each rod-shaped feature, for example, the edge point set P is segmented into n edge point subsets S1, S2, … …, and Sn, and each edge point subset corresponds to one rod-shaped feature entity, so as to implement materialization of the rod-shaped feature.

S140, performing spatial growth of the laser points based on the edge point subset corresponding to each rod-shaped ground object, and determining the geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset.

Specifically, the rod-shaped ground object can be used for presenting a tubular shape in the vertical direction, and the laser points are spatially grown in the vertical direction on the basis of the edge point subset corresponding to each rod-shaped ground object to obtain a grown laser point subset, so that the edge points of the rod-shaped ground object can be refined, and the shape of the rod-shaped ground object is enriched. By performing statistical analysis on the edge point subset and the grown laser point subset corresponding to each rod-shaped ground object, the geometric attribute information corresponding to each rod-shaped ground object can be determined.

Exemplarily, S140 may include: growing the edge point subset corresponding to each rod-shaped ground object upwards in space; based on the grown subset of laser points and the subset of edge points, a first length of each rod-shaped feature in a direction perpendicular to a vehicle travel direction, a second length in the vehicle travel direction, and a third length in the vertical direction are determined.

Specifically, a set of edge points of the objected rod-shaped feature is grown spatially upward in the vertical direction, and a first length of each rod-shaped feature in the direction perpendicular to the vehicle traveling direction, a second length in the vehicle traveling direction, and a third length in the vertical direction are determined based on three-dimensional coordinate information of each point in the grown subset of laser points and the subset of edge points. For example, each entity of the rod-shaped ground object corresponds to a target subset consisting of a subset of edge points and a subset of growing laser points, and the attribute structure of each target subset Point _ Block may be defined as follows:

from this attribute structure it can be seen that: each entity corresponds to an encoded ID, a first Length x _ Length along the x-axis (i.e., perpendicular to the direction of vehicle travel), a second Length y _ Length along the y-axis (i.e., direction of vehicle travel), and a third Length z _ Length along the z-axis (i.e., vertical) in the laser scanning coordinate system. By forming a three-dimensional shape based on the three-dimensional coordinate information of each point in each target subset, a first Length x _ Length, a second Length y _ Length, and a third Length z _ Length corresponding to the corresponding rod-shaped feature can be determined, and the three lengths can reflect the shape characteristics of the corresponding rod-shaped feature.

And S150, determining the specific category corresponding to each rod-shaped ground object based on the geometric attribute information corresponding to each rod-shaped ground object.

Specifically, the geometric attribute information corresponding to each rod-shaped ground object can be judged by using the shape characteristics of each rod-shaped ground object, and the specific category to which each rod-shaped ground object belongs is determined, so that the fine classification of the rod-shaped ground objects is realized. Through the extraction classification mode of this embodiment, can effectively separate out shaft-like ground thing to the rate of accuracy and the integrality that the shaft-like ground thing drawed are all higher, can regard as automatic batch extraction function to use, effectively improve the operating efficiency.

Exemplarily, S150 may include: if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the trunk, determining that the concrete category of the rod-shaped ground object is the trunk; if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the street lamp, determining that the specific type of the rod-shaped ground object is the street lamp; if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the telegraph pole, determining that the specific type of the rod-shaped ground object is the telegraph pole; and if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the traffic sign post, determining that the specific type of the rod-shaped ground object is the traffic sign post.

Specifically, the external shape of each rod-shaped ground object is different, for example, the length of the trunk is generally lower than that of a street lamp and a telegraph pole, and the crowns connected with the trunk are distributed in a scattered manner in space. Street lamps usually have a lamp head extending from the top end of the lamp head in a direction perpendicular to the road, except for a vertical pole, and a wire extending from the top end of the lamp pole is connected to a vertical surface, i.e., a traffic sign, of the traffic sign. By utilizing the characteristics of different forms of the trunk, the street lamp, the telegraph pole and the traffic sign pole, the geometric attribute information corresponding to each rod-shaped ground object can be detected to accord with the shape characteristics of which rod-shaped ground object, so that which specific rod-shaped ground object is can be determined, the fine classification of the rod-shaped ground objects is realized, and the accuracy and the efficiency of extracting and classifying are improved. For example, fig. 2 shows an effect diagram of an extracted trunk. Fig. 3 shows an extracted effect diagram of the street lamp.

Illustratively, for a trunk, the salient feature of the trunk is the crown connected to it, which is a collection of points that are spatially more scattered, so that the trunk has a larger x _ Length, y _ Length, and z _ Length. For street lamps, according to statistical analysis, street lamps are relatively tall and substantially uniform rod-shaped objects, generally about 10 meters above the ground, i.e. z _ Length is usually about 10 meters, and street lamps have differences in height from other rod-shaped objects. Besides the remarkable features in height, the street lamp has a lamp cap with an extended top end, which has the obvious features: y _ Length is small, while x _ Length is typically around 2 meters. For utility poles, the top of the pole is typically connected to a wire that extends linearly over a large span in the x-or y-direction, and therefore the x _ Length or y _ Length of the pole is large, typically over 5 meters. For a traffic sign pole, the Length of the traffic sign pole is relatively fixed, being about 5 meters from the road surface, i.e., z _ Length is typically about 5 meters. The traffic sign is generally vertical to the road, and thus its y _ Length is small. In addition, the traffic sign has a unique feature: the reflection intensity of the laser spot is extremely high. By based on the above-mentioned geometric features, it is possible to detect which kind of geometric features of the rod-shaped feature the first Length x _ Length, the second Length y _ Length, and the third Length z _ Length of each rod-shaped feature conforms to, so that the rod-shaped feature can be classified into 4 specific categories of a trunk, a street lamp, a utility pole, and a traffic sign pole.

In the technical scheme of this embodiment, whether each laser point in the laser point cloud data is an edge point of a rod-shaped feature is detected, and an edge point set corresponding to each rod-shaped feature in two sides of a road is extracted, so that the edge points of the rod-shaped features in the laser point cloud data can be extracted quickly and accurately by using edge features, and an edge point subset corresponding to each rod-shaped feature is determined by clustering and dividing the edge point set, so that the rod-shaped features are materialized, spatial growth of the laser point is performed based on the edge point subset corresponding to each rod-shaped feature, geometric attribute information corresponding to each rod-shaped feature is determined based on the grown laser point subset and the edge point subset, and a specific category corresponding to each rod-shaped feature is determined based on the geometric attribute information corresponding to each rod-shaped feature, so that extraction and subdivision of the rod-shaped features are realized, and the accuracy and efficiency of extraction and classification are improved.

Example two

Fig. 4 is a flowchart of a method for extracting and classifying rod-shaped ground objects according to a second embodiment of the present invention, and this embodiment further optimizes a detection method of edge points in laser point cloud data based on the above embodiments. Wherein explanations of the same or corresponding terms as those of the above embodiments are omitted.

Referring to fig. 4, the method for extracting and classifying a rod-shaped feature provided in this embodiment specifically includes the following steps:

and S410, acquiring the collected laser point cloud data corresponding to two sides of the vehicle driving road.

And S420, taking the first laser point scanned by the laser in the laser point cloud data as the current laser point.

Specifically, it may be detected cyclically one by one whether each laser point in the laser point cloud data is an edge point of the rod-shaped ground object. The first laser spot of the laser scan can be taken as the current laser spot for the first cycle to detect the first laser spot for the first time.

S430, detecting whether the current laser point meets the preset left edge point condition or not according to the current forward distance and the current backward distance corresponding to the current laser point; if yes, the process proceeds to step S440, otherwise, the process proceeds to step S470.

The preset left edge point condition may be set in advance based on the position feature of the left edge point of the rod-shaped feature. Specifically, whether the current laser point is likely to be the left edge point of the rod-shaped ground object or not can be determined by detecting whether the current forward distance and the current backward distance corresponding to the current laser point satisfy the preset left edge point condition or not.

Exemplarily, S430 may include: and if the current forward distance corresponding to the current laser point is greater than the first preset distance and the current backward distance corresponding to the current laser point is less than the second preset distance, determining that the current laser point meets the preset left edge point condition.

Specifically, the previous laser point of the left edge point may fall on the ground or other ground object, so that the forward distance of the left edge point of the rod-shaped ground object is larger and the backward distance is smaller, and according to this characteristic, the preset left edge point condition may be set as: the current forward distance corresponding to the current laser point is greater than a first preset distance, and the current backward distance is less than a second preset distance, wherein the first preset distance is greater than the second preset distance. If the current forward distance corresponding to the current laser point is smaller than or equal to the first preset distance, or the current backward distance corresponding to the current laser point is larger than or equal to the second preset distance, it can be determined that the current laser point does not meet the preset left edge point condition, and at this time, the current laser point cannot be the left edge point.

S440, determining the spatial distance between the current laser point and the next laser point on the same scanning line, and detecting whether the spatial distance meets a preset distance condition; if so, the process proceeds to step S450, and if not, the process proceeds to step S480.

Fig. 5 shows a schematic diagram of the left edge point and the right edge point appearing in pairs. As shown in fig. 5, the left edge point and the right edge point on the same scan line are paired, and according to the characteristics, a preset distance condition may be set to assist in detecting the right edge point corresponding to the left edge point.

Specifically, when the current laser point meets the preset left edge point condition, that is, the current laser point may be a left edge point, a subsequent laser point located on the same scanning line as the current laser point may be obtained, and a three-dimensional euclidean distance between the current laser point and the subsequent laser point may be determined based on the three-dimensional coordinate information of the current laser point and the three-dimensional coordinate information of the subsequent laser point, as a spatial distance, and by detecting whether the spatial distance meets the preset distance condition, it may be determined whether the subsequent laser point may be a right edge point corresponding to the current edge point.

Exemplarily, S440 may include: if the spatial distance is smaller than or equal to a third preset distance, determining that the spatial distance meets a preset distance condition; wherein the third preset distance is set based on a maximum diameter of the rod-shaped ground object.

Wherein the third preset distance may be a maximum spatial distance between the left and right edge points that appear in pairs that are preset. The third preset distance may be determined based on the maximum diameter of the shaft-shaped ground object and the angle between the scanning line and the horizontal line. As shown in fig. 5, the angle between the scanning line and the horizontal line is 45 °, and the third preset distance may be set to be

Wherein L is the maximum diameter of the rod-shaped ground object.

Specifically, whether the spatial distance between the current laser point and the next laser point is less than or equal to a third preset distance is detected, if so, it may be determined that a preset distance condition is satisfied, and at this time, the next laser point may be a right edge point corresponding to the current edge point, and further detection and confirmation are required. If the spatial distance between the current laser point and the next laser point is greater than the third preset distance, it can be determined that the spatial distance does not satisfy the preset distance condition.

S450, detecting whether the next laser point meets the preset right edge point condition or not according to the current forward distance and the current backward distance corresponding to the next laser point; if yes, the process proceeds to step S460, otherwise, the process proceeds to step S490.

The preset right edge point condition may be set in advance based on the position characteristics of the right edge point of the rod-shaped feature. Specifically, when the spatial distance satisfies the preset distance condition, it indicates that the subsequent laser point may be the right edge point corresponding to the current edge point, and at this time, it may be continuously detected whether the current forward distance and the current backward distance corresponding to the subsequent laser point satisfy the preset right edge point condition, so as to determine whether the subsequent laser point is the right edge point corresponding to the current edge point.

Illustratively, S450 may include: and if the current backward distance corresponding to the next laser point is greater than the first preset distance and the current forward distance corresponding to the next laser point is less than the second preset distance, determining that the next laser point meets the preset right edge point condition.

Specifically, the rear laser point of the right edge point may fall on the ground or other ground object, so that the forward distance of the right edge point of the rod-shaped ground object is smaller and the backward distance is larger, and according to the characteristics, the preset right edge point condition may be set as: the current backward distance corresponding to the current laser point is larger than a first preset distance, and the current forward distance is smaller than a second preset distance, wherein the first preset distance is larger than the second preset distance. If the current backward distance corresponding to the latter laser point is less than or equal to the first preset distance, or the current forward distance corresponding to the latter laser point is greater than or equal to the second preset distance, it may be determined that the latter laser point does not satisfy the preset right-side edge point condition, that is, the latter laser point may not be a right-side laser point, which may be a middle point between the left-side edge point and the right-side edge point.

S460, taking the current laser point and the next laser point as edge points of the rod-shaped ground object, and executing the step S480.

Specifically, when the latter laser point meets the preset right edge point condition, it is indicated that the latter laser point is a right edge point corresponding to the current edge point, and correspondingly, the current laser point is also a left laser point, and at this time, both the current laser point and the latter laser point can be used as edge points of the rod-shaped ground object.

And S470, based on the laser scanning sequence, taking the laser point behind the current laser point as the current laser point, and returning to execute the operation of S430 until the current laser point is the last laser point.

Specifically, when the current laser point does not satisfy the preset left edge point condition, it indicates that the current laser point cannot be the left edge point, and the current laser point needs to be updated to continue to detect the next laser point, and at this time, it may be detected whether the current laser point is the last laser point, that is, whether all the laser points are detected completely, so as to determine whether the laser point needs to be detected continuously. When the current laser point is not the last laser point, a laser point subsequent to the current laser point may be used as the current laser point based on the laser scanning order, and the operation of S430 is performed in return to continue detecting whether the next laser point of the current laser point is an edge point. When the current laser point is the last laser point, indicating that all the laser points are detected, the operation of step S491 may be executed to obtain an edge point set composed of all the detected edge points.

And S480, based on the laser scanning sequence, taking the next laser point of the next laser point as the current laser point, and returning to execute the operation of S430 until the next laser point is the last laser point.

Specifically, when the spatial distance does not satisfy the preset distance condition, it indicates that the subsequent laser point is not a right edge point corresponding to the current edge point, and accordingly, the current laser point is not a left laser point, and at this time, it may be determined that the current edge point and the subsequent laser point are not edge points of the rod-shaped ground object, and at this time, it may be detected first whether the subsequent laser point is the last laser point, so as to determine whether the detection of the laser point needs to be continued. When the next laser point is not the last laser point, the next laser point of the next laser point may be taken as the current laser point based on the laser scanning order, and the operation of S430 is performed in return to continue detecting whether the next laser point of the next laser point is an edge point. When the current laser point is the last laser point, indicating that all the laser points are detected, the operation of step S491 may be executed to obtain an edge point set composed of all the detected edge points.

Specifically, after the current laser point and the subsequent laser point are both used as edge points of the rod-shaped ground object, it may be detected whether the subsequent laser point is the last laser point, so as to determine whether the detection of the laser point needs to be continued. When the next laser point is not the last laser point, the next laser point of the next laser point may be taken as the current laser point based on the laser scanning order, and the operation of S430 is performed in return to continue detecting whether the next laser point of the next laser point is an edge point. When the current laser point is the last laser point, indicating that all the laser points are detected, the operation of step S491 may be executed to obtain an edge point set composed of all the detected edge points.

S490, based on the laser scanning sequence, the next laser point of the next laser point is used as the next laser point, and the operation of S440 is executed again until the next laser point is the last laser point.

Specifically, when the latter laser point does not satisfy the preset right edge point condition, it indicates that the latter laser point may be a middle point between the left edge point and the right edge point, and is not the right laser point, and at this time, it may be detected first whether the latter laser point is the last laser point, so as to determine whether to continue the detection of the laser point. When the latter laser point is not the last laser point, the next laser point of the latter laser point may be used as the latter laser point based on the laser scanning sequence, and the operation of S440 is executed in return, so as to detect whether the spatial distance between the current laser point and the next laser point of the latter laser point satisfies the preset distance condition, and continue to detect whether the next laser point of the latter laser point satisfies the preset right-side edge point condition under the condition that the preset distance condition is satisfied, thereby detecting whether the latter laser point is the right-side laser point corresponding to the current laser point one by one.

S491, acquiring edge point sets corresponding to the rod-shaped ground objects on both sides of the road after the detection is finished.

S492, clustering and segmenting the edge point set, and determining the edge point subset corresponding to each rod-shaped ground object.

And S493, performing spatial growth of the laser points based on the edge point subset corresponding to each rod-shaped ground object, and determining the geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset.

And S494, determining a specific category corresponding to each rod-shaped ground object based on the geometric attribute information corresponding to each rod-shaped ground object.

According to the technical scheme of the embodiment, the edge points of the rod-shaped ground object in the laser point cloud data can be more accurately extracted by utilizing the obvious difference of the distance between the edge points on the two sides of the rod-shaped ground object and the adjacent points in the scanning line in the forward direction and the backward direction and the characteristic that the edge points on the two sides appear in pairs, and the accuracy of extraction and classification is further improved.

EXAMPLE III

Fig. 6 is a flowchart of a method for extracting and classifying rod-shaped ground objects according to a third embodiment of the present invention, and this embodiment further optimizes a detection method of edge points in laser point cloud data based on the above embodiments. Wherein explanations of the same or corresponding terms as those of the above embodiments are omitted.

Referring to fig. 6, the method for extracting and classifying rod-shaped ground objects provided in this embodiment specifically includes the following steps:

s610, acquiring laser point cloud data corresponding to two sides of the vehicle driving road.

S620, detecting whether each laser point in the laser point cloud data is an edge point of the rod-shaped ground object, and extracting an edge point set corresponding to each rod-shaped ground object in two sides of the road.

S630, an empty edge point subset is created, and the first edge point in the edge point subset is moved into the edge point subset.

Specifically, when performing cluster segmentation on the edge point set P, an empty edge point subset S1 is first created, and the first edge point in the edge point set P is moved into the edge point subset S1, where an edge point is reduced from the edge point set P.

And S640, taking the currently remaining first edge point in the edge point set as the current edge point.

Specifically, each edge point in the edge point set may be clustered cyclically, and moved into the corresponding edge point subset.

S650, determining a plane distance between the current edge point and each edge point in each edge point subset which is created currently, and determining a minimum plane distance corresponding to each edge point subset based on each plane distance.

Specifically, for each edge point subset S that has been created currently, a plane distance between the current edge point and each edge point in the edge point subset S may be determined based on the X-axis coordinate information and the Y-value coordinate information of the current edge point and the X-axis coordinate information and the Y-value coordinate information of each edge point in the edge point subset S, and the obtained plane distances are compared to determine a minimum plane distance.

And S660, based on the minimum plane distance and the fourth preset distance corresponding to each edge point subset, moving the current edge point into the corresponding edge point subset.

Wherein the fourth preset distance may be preset based on the maximum diameter of the rod-shaped ground object. Specifically, the edge subset S to which the current edge point should be clustered may be determined by detecting whether the minimum planar distance between the current edge point and the edge point in each edge point subset S exceeds a fourth preset distance.

Exemplarily, S660 may include: detecting whether a target edge point subset with the minimum plane distance smaller than or equal to a fourth preset distance exists or not; if yes, moving the current edge point into the target edge point subset; if not, an empty edge point subset is created, and the current edge point is moved into the created edge point subset.

Specifically, if it is detected that there is a target edge point subset with a minimum plane distance smaller than or equal to a fourth preset distance in each of the edge subsets S that have been created, the edge point subset into which the current edge point finally moves may be determined based on the number of the target edge point subsets. For example, if there is only one target edge point subset, the current edge point can be directly moved into the target edge point subset. If at least two target edge point subsets exist, the minimum plane distances corresponding to each target edge point subset can be compared, and the current edge point is moved into the target edge point subset with the shortest minimum plane distance. If it is detected that the minimum plane distance corresponding to each created edge subset is greater than the fourth preset distance, it indicates that the current edge point cannot be clustered into each created edge subset, at this time, an empty edge point subset may be created again, and the current edge point is moved into the created empty edge point subset.

S670, detecting whether the current edge point set is an empty set, if yes, going to S680, and if not, going to S640.

Specifically, whether the cluster segmentation operation of the edge point set is completed or not can be determined by detecting whether the current edge point set is an empty set or not. If the current edge point set is not an empty set, it indicates that the clustering is not completed, and at this time, the operation of S640 may be returned to perform so as to continue clustering on the next laser point in the edge point set. If the current edge point set is an empty set, the clustering is finished, at this time, the edge point set P is divided into n edge point subsets S1, S2, … … and Sn, each edge point subset corresponds to one rod-shaped ground object entity, and therefore the materialization of the rod-shaped ground objects is achieved.

And S680, performing spatial growth of the laser points based on the edge point subset corresponding to each rod-shaped ground object, and determining the geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset.

And S690, determining the specific category corresponding to each rod-shaped ground object based on the geometric attribute information corresponding to each rod-shaped ground object.

According to the technical scheme, the edge points in the edge point set are clustered one by one, so that each edge point subset can be obtained more accurately, materialization of the rod-shaped ground object is achieved, and the accuracy of extraction and classification is further improved.

On the basis of the above technical solution, before performing cluster segmentation on the edge point set and determining the edge point subset corresponding to each rod-shaped ground object, the method may further include: and filtering the interference points in the edge point set based on a plane projection mode of the two-dimensional point density to obtain a filtered edge point set.

Specifically, when the edge features are used for extracting the edge point set in the laser point cloud data, some interference points may also be extracted, such as points on a crown and points of a blocked area on buildings on two sides of a road, so that before clustering and segmentation are performed, the interference points in the edge point set can be filtered based on a planar projection mode of two-dimensional point density, laser points with larger two-dimensional point density are reserved in a neighborhood, most of the laser points fall on vertical ground objects, such as rod-shaped ground objects and building wall surfaces, most of building wall surface points are filtered by using the edge features, and thus the interference points in the edge point set can be further filtered by using the method. After the interference points are filtered, the filtered edge point sets can be subjected to clustering segmentation, and the edge point subset corresponding to each rod-shaped ground object is determined, so that the accuracy of extracting and classifying the rod-shaped ground objects is further improved.

Exemplarily, filtering the interference points in the edge point set based on a planar projection mode of two-dimensional point density to obtain a filtered edge point set, which may include: projecting each edge point in the edge point set to a horizontal plane, and determining the number of points contained in a preset range corresponding to each edge point; and filtering each edge point with the point number smaller than a preset two-dimensional point density threshold value from the edge point set to obtain a filtered edge point set.

The preset range corresponding to the edge point may be a circular area range with the edge point as a center and the preset length as a radius. The preset two-dimensional point density threshold may be determined based on the vehicle-mounted point cloud density and the point density on the rod-shaped ground object.

On the basis of the above technical solutions, after performing cluster segmentation on the edge point set and determining the edge point subset corresponding to each rod-shaped ground object, the method may further include: and screening each edge point subset based on a preset rod height threshold and a preset plane projection length threshold to obtain screened edge point subsets.

Specifically, after the edge point subset corresponding to each rod-shaped feature is determined, there may be an edge point subset corresponding to a rod-shaped feature whose edge subset is not identifiable, and at this time, a preset rod height threshold and a preset planar projection length threshold may be used to screen out an edge point subset within the threshold, and to eliminate an edge point subset not within the threshold, and to perform a fine classification of the rod-shaped feature based on the screened edge point subset, thereby further improving the accuracy of classification and identification.

The following is an embodiment of the device for extracting and classifying rod-shaped ground objects according to an embodiment of the present invention, which belongs to the same inventive concept as the method for extracting and classifying rod-shaped ground objects according to the above embodiments, and reference may be made to the above embodiment of the method for extracting and classifying rod-shaped ground objects for details that are not described in detail in the embodiment of the device for extracting and classifying rod-shaped ground objects.

Example four

Fig. 7 is a schematic structural diagram of an extracting and classifying device for rod-shaped land features according to a fourth embodiment of the present invention, which is applicable to extracting and classifying rod-shaped land features on two sides of a road in laser point cloud data. As shown in fig. 7, the apparatus includes: a laser point cloud data acquisition module 710, an edge point detection module 720, a cluster segmentation module 730, a geometric attribute information determination module 740, and a category determination module 750.

The laser point cloud data acquisition module 710 is used for acquiring laser point cloud data corresponding to two sides of a vehicle driving road; an edge point detection module 720, configured to detect whether each laser point in the laser point cloud data is an edge point of a rod-shaped feature, and extract an edge point set corresponding to each rod-shaped feature on both sides of the road; the clustering and partitioning module 730 is used for clustering and partitioning the edge point set to determine an edge point subset corresponding to each rod-shaped ground object; the geometric attribute information determining module 740 is configured to perform spatial growth of the laser points based on the edge point subset corresponding to each rod-shaped feature, and determine geometric attribute information corresponding to each rod-shaped feature based on the grown laser point subset and the edge point subset; and a category determining module 750, configured to determine a specific category corresponding to each rod-shaped feature based on the geometric attribute information corresponding to each rod-shaped feature.

Optionally, the edge point detecting module 720 is specifically configured to: detecting whether each laser point meets a preset edge point condition or not according to the forward distance and the backward distance corresponding to each laser point in the laser point cloud data, and taking the laser points meeting the preset edge point condition as edge points of the rod-shaped ground object;

the forward distance refers to a three-dimensional distance between a current laser point and a previous laser point on the same scanning line; the backward distance refers to the three-dimensional distance between the current laser point and the next laser point on the same scan.

Optionally, the edge point detecting module 720 specifically includes:

the current laser point determining unit is used for taking a first laser point scanned by laser in the laser point cloud data as a current laser point;

the left edge point detection unit is used for detecting whether the current laser point meets the preset left edge point condition or not according to the current forward distance and the current backward distance corresponding to the current laser point;

the spatial distance detection unit is used for determining the spatial distance between the current laser point and the next laser point on the same scanning line if the current laser point meets the preset left edge point condition, and detecting whether the spatial distance meets the preset distance condition;

the right edge point detection unit is used for detecting whether the next laser point meets the preset right edge point condition or not according to the current forward distance and the current backward distance corresponding to the next laser point if the spatial distance meets the preset distance condition;

and the first updating unit is used for taking the current laser point and the next laser point as the edge points of the rod-shaped ground object if the next laser point meets the preset right edge point condition, taking the next laser point of the next laser point as the current laser point based on the laser scanning sequence, returning to execute the operation of detecting whether the current laser point meets the preset left edge point condition or not according to the current forward distance and the current backward distance corresponding to the current laser point.

Optionally, the left edge point detecting unit is specifically configured to: if the current forward distance corresponding to the current laser point is greater than a first preset distance and the current backward distance corresponding to the current laser point is less than a second preset distance, determining that the current laser point meets the preset left edge point condition;

the right edge point detection unit is specifically configured to: if the current backward distance corresponding to the latter laser point is greater than the first preset distance and the current forward distance corresponding to the latter laser point is less than the second preset distance, determining that the latter laser point meets the preset right-side edge point condition;

the spatial distance detection unit is specifically configured to: if the spatial distance is smaller than or equal to a third preset distance, determining that the spatial distance meets a preset distance condition; wherein the third preset distance is set based on a maximum diameter of the rod-shaped ground object.

Optionally, the apparatus further comprises:

a second updating unit, configured to, if the current laser point does not satisfy the preset left edge point condition, based on the laser scanning sequence, take a laser point subsequent to the current laser point as the current laser point, and return to execute an operation of detecting whether the current laser point satisfies the preset left edge point condition according to the current forward distance and the current backward distance corresponding to the current laser point;

a third updating unit, configured to, if the spatial distance does not satisfy the preset distance condition, based on the laser scanning sequence, take a next laser point of a subsequent laser point as a current laser point, and return to execute an operation of detecting whether the current laser point satisfies the preset left edge point condition according to a current forward distance and a current backward distance corresponding to the current laser point;

and the fourth updating unit is used for taking the next laser point of the next laser point as the next laser point based on the laser scanning sequence if the next laser point does not meet the preset right-side edge point condition, returning to execute the operation of determining the spatial distance between the current laser point and the next laser point on the same scanning line, and detecting whether the spatial distance meets the preset distance condition.

Optionally, the cluster segmentation module 730 includes:

the edge point subset creating unit is used for creating an empty edge point subset and moving a first edge point in the edge point subset into the edge point subset;

a current edge point determining unit, configured to use a current remaining first edge point in the edge point set as a current edge point;

a minimum plane distance determining unit, configured to determine a plane distance between the current edge point and each edge point in each edge point subset that has been created currently, and determine a minimum plane distance corresponding to each edge point subset based on each plane distance;

and the current edge point moving-in unit is used for moving the current edge point into the corresponding edge point subset based on the minimum plane distance and the fourth preset distance corresponding to each edge point subset, and returning to execute the operation of taking the current remaining first edge point in the edge point set as the current edge point until the edge point set is an empty set.

Optionally, the current edge point moving-in unit is specifically configured to: detecting whether a target edge point subset with the minimum plane distance smaller than or equal to a fourth preset distance exists or not; if yes, moving the current edge point into the target edge point subset; if not, an empty edge point subset is created, and the current edge point is moved into the created edge point subset.

Optionally, the apparatus further comprises:

and the interference point filtering module is used for filtering the interference points in the edge point set based on a plane projection mode of two-dimensional point density before clustering and dividing the edge point set and determining the edge point subset corresponding to each rod-shaped ground object, so as to obtain the filtered edge point set.

Optionally, the interference point filtering module is specifically configured to: projecting each edge point in the edge point set to a horizontal plane, and determining the number of points contained in a preset range corresponding to each edge point; and filtering each edge point with the point number smaller than a preset two-dimensional point density threshold value from the edge point set to obtain a filtered edge point set.

Optionally, the apparatus further comprises:

and the edge point subset screening module is used for screening each edge point subset based on a preset rod height threshold and a preset plane projection length threshold after clustering and segmenting the edge point subset and determining the edge point subset corresponding to each rod-shaped ground object, so as to obtain the screened edge point subset.

Optionally, the geometric attribute information determining module 740 is specifically configured to: growing the edge point subset corresponding to each rod-shaped ground object upwards in space; based on the grown subset of laser points and the subset of edge points, a first length of each rod-shaped feature in a direction perpendicular to a vehicle travel direction, a second length in the vehicle travel direction, and a third length in the vertical direction are determined.

Optionally, the category determining module 750 is specifically configured to: if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the trunk, determining that the concrete category of the rod-shaped ground object is the trunk; if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the street lamp, determining that the specific type of the rod-shaped ground object is the street lamp; if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the telegraph pole, determining that the specific type of the rod-shaped ground object is the telegraph pole; and if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the traffic sign post, determining that the specific type of the rod-shaped ground object is the traffic sign post.

The device for extracting and classifying the rod-shaped ground object provided by the embodiment of the invention can execute the method for extracting and classifying the rod-shaped ground object provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the method for extracting and classifying the rod-shaped ground object.

It should be noted that, in the embodiment of the device for extracting and classifying rod-shaped ground objects, the included units and modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.

EXAMPLE five

Fig. 8 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present invention. FIG. 8 illustrates a block diagram of an exemplary electronic device 12 suitable for use in implementing embodiments of the present invention. The electronic device 12 shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention.

As shown in FIG. 8, electronic device 12 is embodied in the form of a general purpose computing device. The components of electronic device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that couples various system components including the system memory 28 and the processing unit 16.

Bus 18 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Electronic device 12 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by electronic device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)30 and/or cache memory 32. The electronic device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 8, and commonly referred to as a "hard drive"). Although not shown in FIG. 8, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 by one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.

A program/utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 42 generally carry out the functions and/or methodologies of the described embodiments of the invention.

Electronic device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), with one or more devices that enable a user to interact with electronic device 12, and/or with any devices (e.g., network card, modem, etc.) that enable electronic device 12 to communicate with one or more other computing devices. Such communication may be through an input/output (I/O) interface 22. Also, the electronic device 12 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet) via the network adapter 20. As shown, the network adapter 20 communicates with other modules of the electronic device 12 via the bus 18. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

The processing unit 16 executes various functional applications and data processing by running a program stored in the system memory 28, for example, implementing the steps of a method for extracting and classifying a rod-shaped feature provided by the embodiment of the present invention, the method including:

performing space growth of laser points based on the edge point subset corresponding to each rod-shaped ground object, and determining the geometric attribute information corresponding to each rod-shaped ground object based on the grown laser point subset and the edge point subset;

Of course, those skilled in the art can understand that the processor can also implement the technical solution of the method for extracting and classifying the rod-shaped ground object provided by any embodiment of the present invention.

EXAMPLE six

The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the method for extracting and classifying a rod-shaped feature according to any of the embodiments of the present invention, the method comprising:

Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer-readable storage medium may be, for example but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It will be understood by those skilled in the art that the modules or steps of the invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of computing devices, and optionally they may be implemented by program code executable by a computing device, such that it may be stored in a memory device and executed by a computing device, or it may be separately fabricated into various integrated circuit modules, or it may be fabricated by fabricating a plurality of modules or steps thereof into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for extracting and classifying rod-shaped ground objects is characterized by comprising the following steps:

2. The method of claim 1, wherein the detecting whether each laser point in the laser point cloud data is an edge point of a rod-shaped terrain comprises:

detecting whether each laser point meets a preset edge point condition or not according to the forward distance and the backward distance corresponding to each laser point in the laser point cloud data, and taking the laser points meeting the preset edge point condition as edge points of the rod-shaped ground object;

3. The method according to claim 2, wherein the detecting whether each laser point satisfies a preset edge point condition according to the forward distance and the backward distance corresponding to each laser point in the laser point cloud data, and using the laser point satisfying the preset edge point condition as an edge point of the rod-shaped ground object comprises:

taking a first laser point scanned by laser in the laser point cloud data as a current laser point;

detecting whether the current laser point meets the preset left edge point condition or not according to the current forward distance and the current backward distance corresponding to the current laser point;

if the current laser point meets the preset left edge point condition, determining the spatial distance between the current laser point and a next laser point on the same scanning line, and detecting whether the spatial distance meets the preset distance condition;

if the spatial distance meets the preset distance condition, detecting whether the next laser point meets the preset right-side edge point condition or not according to the current forward distance and the current backward distance corresponding to the next laser point;

and if the latter laser point meets the preset right-side edge point condition, taking the current laser point and the latter laser point as edge points of the rod-shaped ground object, taking the next laser point of the latter laser point as the current laser point based on the laser scanning sequence, returning to execute the current forward distance and the current backward distance corresponding to the current laser point, and detecting whether the current laser point meets the preset left-side edge point condition.

4. The method of claim 3, wherein the detecting whether the current laser point satisfies the preset left edge point condition according to the current forward distance and the current backward distance corresponding to the current laser point comprises:

if the current forward distance corresponding to the current laser point is greater than a first preset distance and the current backward distance corresponding to the current laser point is less than a second preset distance, determining that the current laser point meets the preset left edge point condition;

whether the condition of presetting right side edge point is satisfied to the current forward distance and the current backward distance that correspond according to a back laser point, detect back laser point includes:

if the current backward distance corresponding to the latter laser point is greater than the first preset distance and the current forward distance corresponding to the latter laser point is less than the second preset distance, determining that the latter laser point meets the preset right-side edge point condition;

detecting whether the space distance meets a preset distance condition or not, including:

if the spatial distance is smaller than or equal to a third preset distance, determining that the spatial distance meets a preset distance condition; wherein the third preset distance is set based on a maximum diameter of the rod-shaped ground object.

5. The method of claim 3, further comprising:

if the current laser point does not meet the preset left edge point condition, based on the laser scanning sequence, taking the next laser point of the current laser point as the current laser point, returning to execute the current forward distance and the current backward distance corresponding to the current laser point, and detecting whether the current laser point meets the preset left edge point condition;

if the spatial distance does not meet the preset distance condition, based on the laser scanning sequence, taking the next laser point of the next laser point as the current laser point, returning to execute the current forward distance and the current backward distance corresponding to the current laser point, and detecting whether the current laser point meets the preset left edge point condition;

and if the latter laser point does not meet the preset right edge point condition, based on the laser scanning sequence, taking the next laser point of the latter laser point as the latter laser point, returning to execute the operation of determining the spatial distance between the current laser point and the latter laser point on the same scanning line, and detecting whether the spatial distance meets the preset distance condition.

6. The method of claim 1, wherein the performing cluster segmentation on the edge point set to determine a corresponding edge point subset for each rod-shaped feature comprises:

creating an empty edge point subset, and moving a first edge point in the edge point subset into the edge point subset;

taking the first edge point which is remained at present in the edge point set as the current edge point;

determining a plane distance between the current edge point and each edge point in each edge point subset which is created currently, and determining a minimum plane distance corresponding to each edge point subset based on each plane distance;

and based on the minimum plane distance and the fourth preset distance corresponding to each edge point subset, moving the current edge point into the corresponding edge point subset, and returning to execute the operation of taking the current remaining first edge point in the edge point set as the current edge point until the edge point set is an empty set.

7. The method according to claim 6, wherein the moving the current edge point into the corresponding edge point subset based on the minimum planar distance and the fourth preset distance corresponding to each edge point subset comprises:

detecting whether a target edge point subset with the minimum plane distance smaller than or equal to a fourth preset distance exists;

if yes, moving the current edge point into the target edge point subset;

if not, an empty edge point subset is created, and the current edge point is moved into the created edge point subset.

8. The method of claim 1, wherein before performing cluster segmentation on the edge point set and determining a corresponding edge point subset for each rod-shaped feature, the method further comprises:

and filtering the interference points in the edge point set based on a plane projection mode of the two-dimensional point density to obtain a filtered edge point set.

9. The method according to claim 8, wherein the filtering the interference points in the edge point set based on the planar projection mode of the two-dimensional point density to obtain a filtered edge point set comprises:

projecting each edge point in the edge point set to a horizontal plane, and determining the number of points contained in a preset range corresponding to each edge point;

and filtering each edge point with the point number smaller than a preset two-dimensional point density threshold value from the edge point set to obtain a filtered edge point set.

10. The method of claim 1, wherein after performing cluster segmentation on the edge point set and determining a subset of edge points corresponding to each rod-shaped feature, the method further comprises:

and screening each edge point subset based on a preset rod height threshold and a preset plane projection length threshold to obtain screened edge point subsets.

11. The method according to any one of claims 1-10, wherein the spatially growing the laser points based on the subset of edge points corresponding to each rod-shaped feature, and determining the geometric attribute information corresponding to each rod-shaped feature based on the grown subset of laser points and the subset of edge points comprises:

growing the edge point subset corresponding to each rod-shaped ground object upwards in space;

and determining a first length of each rod-shaped ground object along the direction perpendicular to the driving direction of the vehicle, a second length of each rod-shaped ground object along the driving direction of the vehicle and a third length of each rod-shaped ground object along the vertical direction on the basis of the grown laser point subset and the edge point subset.

12. The method according to any one of claims 1-10, wherein the determining the specific category corresponding to each rod-shaped feature based on the geometric attribute information corresponding to each rod-shaped feature comprises:

if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the trunk, determining that the concrete category of the rod-shaped ground object is the trunk;

if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the street lamp, determining that the specific type of the rod-shaped ground object is the street lamp;

if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the telegraph pole, determining that the specific type of the rod-shaped ground object is the telegraph pole;

and if the geometric attribute information corresponding to the rod-shaped ground object meets the shape characteristics of the traffic sign post, determining that the specific type of the rod-shaped ground object is the traffic sign post.

13. An extraction and classification device for rod-shaped ground objects, which is characterized by comprising:

14. An electronic device, characterized in that the electronic device comprises:

one or more processors;

a memory for storing one or more programs;

when executed by the one or more processors, cause the one or more processors to implement the method of extracting and classifying a rod-shaped feature of any one of claims 1-12.

15. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a method for extracting and classifying a rod-shaped feature according to any one of claims 1 to 12.