CN108241819B - Method and device for identifying pavement marker - Google Patents
Method and device for identifying pavement marker Download PDFInfo
- Publication number
- CN108241819B CN108241819B CN201611201463.XA CN201611201463A CN108241819B CN 108241819 B CN108241819 B CN 108241819B CN 201611201463 A CN201611201463 A CN 201611201463A CN 108241819 B CN108241819 B CN 108241819B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- cloud data
- sample point
- road
- mark
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The application provides a method and a device for identifying a pavement marker, wherein the method comprises the following steps: along a lane line of a road, obtaining sample point cloud data from point cloud data marked as a road surface mark of the road; sequencing the sample point cloud data along the direction of the road to obtain sequenced sample point cloud data; dividing the sorted sample point cloud data into sample point cloud data blocks according to a preset interval distance along the direction of a road; counting each sample point cloud data block to obtain distribution data of the sample point cloud data; and judging whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the road mark with the preset characteristics or not according to the distribution data of the sample point cloud data, and if so, identifying the road mark as the road mark with the preset characteristics. The invention fully utilizes the point cloud distribution characteristics of some road surface marks, and can efficiently and accurately identify the road surface marks with the point cloud characteristics.
Description
Technical Field
The present application relates to the field of electronic map technologies, and in particular, to a method and an apparatus for identifying a road surface marker.
Background
The road surface mark is a mark for transmitting traffic information such as guidance, restriction, warning and the like to traffic participants by lines, arrows, characters, elevation marks, raised road signs, contour marks and the like on the road surface of a road, and has the function of controlling and guiding traffic. Pavement markings are also known as pavement markings, and the like.
In the fields of electronic maps, automatic driving, and the like, various pavement markers need to be identified. For example, in the process of making or updating an electronic map, road surface marking information on roads, such as lane lines, deceleration marks, pedestrian crossings, arrows and the like, needs to be obtained, and the position accuracy is generally required to be less than 10 cm.
At present, the rough methods for identifying the pavement marks mainly comprise three methods, and each method has characteristics.
The first is manual identification. By collecting video data collected by the vehicle, pavement markings are manually found in the video data, and then the type is marked. The method needs manual operation in the whole process, has low efficiency and can not meet the processing requirement of mass data.
The second is automatic recognition by a pattern matching algorithm. Information such as geometric dimensions, reflectivity, dot density, and the like of various road surface markings is previously arranged in a program, the type of the road surface marking is determined by a model matching method, and the type with the highest probability is output. In the method, the pattern matching result depends on the accuracy of the model library, and the size information of some models has similarity, for example, the deceleration mark is similar to the size information of the models such as Chinese characters, a vehicle distance confirmation line and the like, so the identification accuracy is not high.
And the third is automatic recognition by an image matching algorithm. The data is converted into an image by using an image matching method, and then the image is identified by using a model library of the image. This method is inefficient for some road markings, for example, for a deceleration marked image composed of a large amount of point cloud data, since each pixel of the image is processed, the program execution efficiency is low.
Disclosure of Invention
One of the technical problems to be solved by the present application is to provide a method and an apparatus for identifying a road surface mark, which efficiently identify the road surface mark by using the distribution characteristics of point cloud data.
According to one embodiment of the present application, there is provided a method of identifying a pavement marking, the method including the steps of: along a lane line of a road, obtaining sample point cloud data from point cloud data marked as a road surface mark of the road; sequencing the sample point cloud data along the direction of the road to obtain sequenced sample point cloud data; dividing the sorted sample point cloud data into sample point cloud data blocks according to a preset interval distance along the direction of a road; counting each sample point cloud data block to obtain distribution data of the sample point cloud data; and judging whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the road mark with the preset characteristics or not according to the distribution data of the sample point cloud data, and if so, identifying the road mark as the road mark with the preset characteristics.
According to another embodiment of the present application, there is provided a pavement marking recognition apparatus including: an acquisition unit configured to acquire sample point cloud data from point cloud data of one road surface marker marked as a road along a lane line of the road; the sorting unit is used for sorting the sample point cloud data along the direction of the road to obtain sorted sample point cloud data; the dividing unit is used for dividing the sorted sample point cloud data into sample point cloud data blocks according to a preset interval distance along the direction of a road; the statistical unit is used for carrying out statistics on each sample point cloud data block to obtain the distribution data of the sample point cloud data; and the identification unit is used for judging whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the road mark with the preset characteristics or not according to the distribution data of the sample point cloud data, and if so, identifying the road mark as the road mark with the preset characteristics.
The embodiment of the invention fully utilizes the point cloud distribution characteristics of some pavement markers, sequences and counts the point cloud data marked as the pavement markers to obtain the distribution rule of the point cloud data, and judges whether the pavement markers have preset characteristics or not according to the distribution rule. The scheme of the invention has automatic identification without manual operation, so the cost is very low and the practicability is very strong. The invention fully utilizes the point cloud distribution characteristics of the pavement markers, and can efficiently and accurately identify the pavement markers with preset characteristics.
It will be appreciated by those of ordinary skill in the art that although the following detailed description will proceed with reference being made to illustrative embodiments, the present application is not intended to be limited to these embodiments.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is a flow chart of a method of identifying pavement markings according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a lateral deceleration mark according to an embodiment of the present application;
FIG. 3 is a schematic illustration of a longitudinal deceleration mark according to an embodiment of the present application;
FIG. 4 is a flow chart of a method of ordering sample point cloud data along a road direction according to an embodiment of the application;
FIG. 5 is a schematic diagram of a sample point cloud region obtained according to an embodiment of the present application;
FIGS. 6(a), 6(b) and 6(c) are schematic diagrams of three pavement markers to be identified according to the embodiment of the application;
FIG. 7 is a schematic view of a sliding window according to an embodiment of the present application;
FIG. 8 is a graphical illustration of statistical information for the three pavement markings to be identified of FIG. 6 according to an embodiment of the present disclosure;
fig. 9 is a schematic structural view of a pavement marking recognition apparatus according to an embodiment of the present application.
It will be appreciated by those of ordinary skill in the art that although the following detailed description will proceed with reference being made to illustrative embodiments, the present application is not intended to be limited to these embodiments. Rather, the scope of the application is broad and is intended to be defined only by the claims that follow.
Detailed Description
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel, concurrently, or simultaneously. In addition, the order of the operations may be re-arranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
The computer equipment comprises user equipment and network equipment. Wherein the user equipment includes but is not limited to computers, smart phones, PDAs, etc.; the network device includes, but is not limited to, a single network server, a server group consisting of a plurality of network servers, or a Cloud Computing (Cloud Computing) based Cloud consisting of a large number of computers or network servers, wherein Cloud Computing is one of distributed Computing, a super virtual computer consisting of a collection of loosely coupled computers. The computer equipment can be independently operated to realize the application, and can also be accessed into a network to realize the application through the interactive operation with other computer equipment in the network. The network in which the computer device is located includes, but is not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a VPN network, and the like.
It should be noted that the user equipment, the network device, the network, etc. are only examples, and other existing or future computer devices or networks may also be included in the scope of the present application, if applicable, and are included by reference.
The methods discussed below, some of which are illustrated by flow diagrams, may be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware or microcode, the program code or code segments to perform the necessary tasks may be stored in a machine or computer readable medium such as a storage medium. The processor(s) may perform the necessary tasks.
Specific structural and functional details disclosed herein are merely representative and are provided for purposes of describing example embodiments of the present application. This application may, however, be embodied in many alternate forms and should not be construed as limited to only the embodiments set forth herein.
It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element may be termed a second element, and, similarly, a second element may be termed a first element, without departing from the scope of example embodiments. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may be present. In contrast, when an element is referred to as being "directly connected" or "directly coupled" to another element, there are no intervening elements present. Other words used to describe the relationship between elements (e.g., "between" versus "directly between", "adjacent" versus "directly adjacent to", etc.) should be interpreted in a similar manner.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used herein, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be noted that, in some alternative implementations, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed substantially concurrently, or the figures may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
The technical solution of the present application is further described in detail below with reference to the accompanying drawings.
Fig. 1 is a flowchart of a method for identifying a pavement marking according to an embodiment of the present application, the method mainly including the steps of:
s101: along a lane line of a road, obtaining sample point cloud data from point cloud data marked on one road surface of the road;
the point cloud data (point cloud data) refers to data recorded in the form of points by scanning or the like, and each point includes three-dimensional position coordinates, and some may include color information (RGB) or reflection Intensity information (Intensity).
After the point cloud data along the road is acquired, the classification of the point cloud data can be marked in advance in a manual or machine marking mode. For example, point cloud data of roads, point cloud data of pavement markings, point cloud data of bridges, point cloud data of road accessories such as signal lights, and the like. The invention processes the point cloud data marked as the road surface marks, wherein a plurality of road surface marks are arranged on one road, and each road surface mark corresponds to some point cloud data.
S102: sequencing the sample point cloud data along the direction of the road to obtain sequenced sample point cloud data;
s103: dividing the sorted sample point cloud data into sample point cloud data blocks according to a preset interval distance along the direction of a road;
s104: counting each sample point cloud data block to obtain distribution data of the sample point cloud data;
s105: and judging whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the road surface marker with the preset characteristics or not according to the distribution data of the sample point cloud data, and if so, identifying the road surface marker as the road surface marker with the preset characteristics.
For a further understanding of the present application, the following schemes are described in further detail.
According to the embodiment of the invention, the point cloud distribution characteristics of some pavement markers are fully utilized, and the pavement markers with specific characteristics are identified according to the repeatability rule of point cloud data.
The technical scheme provided by the embodiment of the invention is particularly suitable for identifying the traffic marking road surface marks with point cloud characteristics. The traffic marking road surface mark having the point cloud characteristic means a marking mark composed of a plurality of repeated point cloud data or a marking mark composed of a plurality of regularly presented point cloud data. These markings are typically bar-shaped and appear repeatedly, such as deceleration markings, crosswalks, etc. For example, deceleration markers are often placed on toll booths, exit ramps, or other road segments requiring deceleration of a vehicle, and may be classified as lateral deceleration markers and longitudinal deceleration markers. Referring to fig. 2, a schematic diagram of a lateral deceleration mark is shown. The lateral deceleration marks are generally white dashed lines in the form of single, double and triple dashed lines, arranged perpendicular to the direction of travel. Referring to fig. 3, a schematic diagram of a longitudinal deceleration mark is shown. The longitudinal deceleration mark is a group of diamond-shaped block dashed lines parallel to the boundary line of the roadway, a transition section (such as 30 meters) is arranged at the initial position of the longitudinal deceleration mark of the roadway, and the diamond-shaped block dashed lines are changed from narrow to wide and the width is changed from narrow to wide (such as 10 centimeters to 30 centimeters). It can be seen that, regardless of the transverse deceleration mark and the longitudinal deceleration mark, the point cloud data includes repeated or regular points. Therefore, the lateral deceleration mark or the longitudinal deceleration mark is a typical road surface mark having a point cloud characteristic.
It should be noted that, although only the transverse deceleration mark and the longitudinal deceleration mark are described as typical road surface marks having a point cloud characteristic, those skilled in the art should understand that the road surface marks having a point cloud characteristic are not limited thereto, and include other road surface marks having a point cloud characteristic that are not listed in the past or will appear in the future.
The steps of the foregoing method are described in detail below.
The point cloud is a massive point set which expresses the space distribution and the surface characteristic of the target under the same space reference system, if the processing workload of the point cloud data is huge, the efficiency is low, and the embodiment of the invention can reduce the data volume to be processed by only selecting partial point cloud data of the pavement marker, thereby improving the efficiency.
In a specific implementation, in step S101, sample point cloud data is acquired from point cloud data of a road surface marker marked as a road along a lane line of the road.
Generally, point cloud data with typical characteristics of the pavement marker is selected as sample point cloud data. For example, for the transverse deceleration mark, because the transverse deceleration mark is generally formed by a plurality of dotted lines perpendicular to the driving direction, the point cloud data in the transverse deceleration mark, which are located in a certain width range at two sides of the center line of the road, can be selected as the sample point cloud data according to the distribution characteristics of the point cloud data in the transverse deceleration mark. Accordingly, along the lane line of the road, the following specific implementation manner can be adopted to obtain the sample point cloud data from the point cloud data marked as the road surface of the road:
acquiring the central line of a lane line of a road;
and acquiring point cloud data of which the distance from the central line of the lane line is not more than a preset distance threshold value from the point cloud data of one pavement marker marked as the road as sample point cloud data.
If the total number of lanes of the road is an odd number, the lane line is preferably the lane line of the middle lane, if the total number of lanes of the road is an even number, the preferred lane number is the lane line of the lane in which the total number of lanes is divided by 2, and for the country with the driving position on the left side, the lane numbers are numbered from left to right; for countries with the driving position on the right, the lane numbers are numbered from right to left. Therefore, the center line of the lane acquired by the invention can be used as the center line of the road.
Then, in step S102, the sample point cloud data is sorted along the direction of the road to obtain sorted sample point cloud data.
Specifically, referring to fig. 4 and 5, the sample point cloud data may be sorted along the road direction by the following steps:
s401: selecting one sample point cloud data from the sample point cloud data as reference point cloud data;
as shown in fig. 5, the point cloud data in the rectangular frame is sample point cloud data, and any one of the point cloud data is selected as reference point cloud data in step S401.
S402: selecting a target position point from the central line of a lane line of a road, wherein the target position point is positioned outside a sample point cloud data coverage area;
s403: acquiring the distance from each sample point cloud data to a target position point;
s404: along the direction of the road, if the target position point is positioned behind the reference point cloud data, sequencing the sample point cloud data according to the sequence that the distance from the target position point is from small to large to obtain the sequenced sample point cloud data;
s405: and along the direction of the road, if the target position point is positioned in front of the reference point cloud data, sequencing the sample point cloud data according to the sequence of the distances from the target position point from large to small to obtain the sequenced sample point cloud data.
As shown in fig. 5, the central line arrow indicates the direction of the road, and the target position point is located behind the reference point cloud data, so that the sample point cloud data needs to be sorted in the order of increasing the distance from the target position point in step S404. Otherwise, sorting is performed according to step 405.
Next, in step S103, the sorted sample point cloud data is divided into sample point cloud data blocks according to a preset interval distance along the direction of the road.
If the pavement marks which are in the shape of horizontal stripes and appear repeatedly are identified, preferably, the sorted sample point cloud data can be divided into more than four sample point cloud data blocks according to a preset interval distance.
In specific implementation, the sorted sample point cloud data can be sequentially scanned by using a sliding window to obtain a sample point cloud data block. The method adopts a sliding window scanning mode, and aims to control a window to sequentially scan the sorted sample point cloud data, wherein the scanning mode comprises two layers of meanings, namely, the scanning sequence is certain (along a certain direction), and secondly, the sliding window is used for carrying out multiple segmented block scanning on the sorted sample point cloud data. Therefore, a plurality of sample point cloud data blocks in the same direction are obtained, and a basis is provided for subsequent statistics and judgment of whether each sample point cloud data block has certain repeatability/regularity.
Sliding window (Sliding window) is a flow control technique. In the embodiment of the invention, based on the multithreading principle of the sliding window, the length and the step length of the sliding window are set, and the sorted sample point cloud data are sequentially scanned according to the step length along a certain direction. And after scanning, obtaining a sample point cloud data block.
Then, in step S104, each sample point cloud data block is counted to obtain distribution data of the sample point cloud data.
Specifically, the statistics of the sample point cloud data block can be realized by the following steps:
step 1041, counting the number of sample point cloud data in each sample point cloud data block, if the number is greater than a preset number threshold, marking the sample point cloud data block as data, otherwise, marking the sample point cloud data block as blank;
1042, traversing the marking result of each sample point cloud data block, and recording the continuous occurrence times of the ith data mark L One when the ith data mark occurs continuouslyiAnd recording the number L Zero of consecutive occurrences of the jth blank mark when the jth occurrence of the consecutive blank markjWherein, L Onei、LZerojAnd i and j are natural numbers which are greater than or equal to 1 and are distribution data of the sample point cloud data.
Since there are multiple sample point cloud data blocks, in order to improve the processing efficiency, the foregoing step 1041 may be performed in parallel on the sample point cloud data blocks, that is, similar to the "multithreading" principle. Specifically, a plurality of data buffers equal to the number of the sample point cloud data blocks may be set, and the data buffers are used to store data in each sample point cloud data block, and then the step 1041 is executed in parallel to complete statistics on the sample point cloud data blocks quickly. For example, if the sample point cloud data selected by the transverse deceleration mark is scanned, and the length of the area covered by the sample point cloud data along the direction of the road is 2 meters, and meanwhile, the length of the sliding window is set to be 0.1 meter and the window step length is set to be 0.1 meter, then the sliding window needs to be slid 20 times (2/0.1 equals 20) to complete the scanning of the sample point cloud data with the length of 2 meters, so as to obtain 20 sample point cloud data blocks. Correspondingly, 20 data buffer areas are set for data storage, specifically, an array with the length of 20 can be set as the data buffer area, and storage of all sample point cloud data blocks is completed.
A specific implementation of the step 104 is described by way of example. Assuming that the sample point cloud data blocks are 10 blocks in total, the sequence number is 1-10, and the number statistical results are shown in table 1, wherein 0 represents no data, and 1 represents data. The result obtained by traversing the number statistical result is as follows:
Lone1＝4，Lone2＝2，LZero1＝2，LZero2＝2。
| numbering | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| Number of | 0 | 0 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 |
Finally, in step S105, it is determined whether the distribution rule of the sample point cloud data satisfies the distribution rule of the point cloud data of the road surface marker with the preset characteristic according to the distribution information of the sample point cloud data.
For example, pavement markings of predetermined characteristics are striped and repeating pavement markings. For example, crosswalks, zebra crossings, etc. all belong to this pavement marking. The focus of the present invention is to identify cross-striped and repetitive pavement markings.
Specifically, the identification of the pavement marking of the preset characteristic may be achieved by:
determine each L Onei+1And L Zeroj+1Whether the following formula is satisfied:
OneLength*Lself-adaption1<LOnei+1<OneLength*Lself-adaption2；
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength*Lself-adaption2；
and if the two parameters are met, determining that the road surface mark meets the preset cross-stripe-shaped repeated road surface mark, wherein One L ength is L One1, Zero L ength is L Zero1, L self-adaptation 1, L self-adaptation 2 are preset adaptive parameters, and L self-adaptation 1< L self-adaptation 2.
In practical applications, L self-adaptation 1 is 0.5, and L self-adaptation 2 is 1.5, which are only preferred embodiments and not limitations on the values of the adaptive parameters of the present invention, and those skilled in the art can take the values of the adaptive parameters according to actual needs.
Continuing with the foregoing example, One L ength ═ L One1＝4，ZeroLength＝LZero1At this time, L one is equal to 222 is less than One L ength 3/2, equal to One L ength/2, L Zero22 is smaller than Zero L ength 3/2 and larger than Zero L ength/2, so that the pavement markings shown by way of example do not belong to the group of cross-striped and recurring pavement markings.
Therefore, the embodiment of the invention fully utilizes the point cloud distribution characteristics of some pavement markers, orders and counts the point cloud data marked as the pavement markers to obtain the distribution rule of the point cloud data, and judges whether the pavement markers have the preset characteristics according to the distribution rule. The scheme of the invention has automatic identification without manual operation, so the cost is very low and the practicability is very strong. The invention fully utilizes the point cloud distribution characteristics of some road surface marks, and can efficiently and accurately identify the road surface marks with the point cloud characteristics.
Embodiments of the present invention will now be described with reference to the identification of a particular lateral deceleration mark.
Referring to fig. 6, a schematic diagram of three pavement markings to be identified is shown, wherein fig. 6(a) and 6(b) are two types of pavement deceleration markings and fig. 6(c) is an optional additional pavement marking. The following example recognizes these three kinds of road surface markings and compares the recognition results.
The identification of the transverse deceleration mark mainly comprises the following five steps.
Considering the characteristic that the transverse deceleration marks are repeatedly distributed by multiple sections of dotted lines, all the points of the pavement marks do not need to be used, and only one section of data in the middle is needed, as shown in fig. 5, the sample point cloud area schematic diagram of the transverse deceleration marks is shown, in this example, points in a certain area in the middle of a road are selected as sample point cloud data. Assuming that the lane line width is W and the distance of the point Pi from the right lane line is Di, the point Pi satisfying the following condition may be taken as sample point cloud data:
|W/2-Di|<l
where l is a preset distance threshold, for example, taking l equal to 0.2, i.e., | W/2-Di|<0.2, the distance from the sample point cloud data to the lane line of the road is less than 0.2, that is, the width of the coverage area of the sample point cloud data shown in fig. 5 is 0.4, that is, point clouds within 0.2 meters around the center line of the road are taken as the sample point cloud data.
And 2, sequencing the sample point cloud data along the road direction.
Screening out sample point cloud data PiThen, density clustering and noise removal processing can be performed on the screened points, for example, clustering is performed by using a DBSCAN method; the clustered points are then sorted such that the points follow the roadThe directions are arranged. Finally, a set of points is defined as P ═ P1,P2,…,Pn}。
And 3, scanning the sorted sample point cloud data by using a sliding window to obtain a sample point cloud data block.
The invention uses a sliding window method to count the point cloud distribution information. Referring to fig. 7, a sliding window schematic is shown, which is divided into 6 blocks, i.e., if data can be scanned, the sliding window can acquire 6 sample point cloud data blocks at a time. In this example, the window slides in the direction of travel of the vehicle, the window being 0.1 meter long by a sliding step of 0.1 meter. Setting Window buffer size to Dist (P)1,Pn) 2+1, where Dist represents the distance of the two points projected on the centerline.
And 4, counting the quantity information of the point clouds based on the sample point cloud data blocks.
And traversing the sample point cloud data block, and rounding each point Pi with the buffer subscript of Dist (Pi, P1) × 10. And after traversing is completed, recording the number of points of each buffer area. Referring to fig. 8, a statistical information diagram for the three pavement markings of fig. 6 is shown. In fig. 8, the abscissa is the buffer subscript and the ordinate is the number of dots. As can be seen from the statistical results in fig. 8, the pavement marking in fig. 6(a) has complete data statistics and shows a repeated occurrence rule, the pavement marking in fig. 6(b) has approximately half of the data statistics and also shows a repeated occurrence rule, and the pavement marking in fig. 6(c) does not show the repeated occurrence or regularity.
And 5, analyzing the statistical result and judging whether the periodic rule is met.
The point density of the transverse deceleration mark accords with the periodic characteristics, but specific period parameters are unknown, and the period parameters of transverse deceleration marks in different forms (the transverse deceleration marks in the forms of single dotted lines, double dotted lines and three dotted lines) are inconsistent, so that global parameters cannot be set for judgment. In the invention, a self-adaptive method is used for pre-estimating the period parameters of the statistical information, and then the whole is judged whether to meet the periodic rule.
And (3) carrying out binarization processing on the buffer area, wherein the number of points is higher than a certain amount (such as 5) and is 1, and the number of points is less than or equal to 5 and is 0, traversing the buffer area, recording the length of the first continuous 1 as One L ength, and recording the length of the first continuous 0 as Zero L ength.
The buffer is traversed twice, recording the length of each successive 1 or 0, noted L One and L Zero, respectively, for any L Onei+1And L Zeroj+1And if all of them are satisfied:
OneLength*Lself-adaption1<LOnei+1<OneLength*Lself-adaption2
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength*Lself-adaption2
the periodic signature is met and identified as a transverse deceleration mark type, otherwise as another type.
L self-adaptation 1 and L self-adaptation 2 are adaptive parameters set empirically in advance, values of which define an adaptation space, and L self-adaptation 1< L self-adaptation 2, for example, L self-adaptation 1 is 0.5 and L self-adaptation 2 is 1.5.
In this example, as a result of the above determination, it is determined that both fig. 6(a) and 6(b) are correctly determined as the lateral deceleration mark; fig. 6(c) is determined to be a non-lateral deceleration flag or output unknown type.
The embodiment of the present application provides a pavement marking recognition apparatus corresponding to the above pavement marking recognition method, as shown in fig. 9, the pavement marking recognition apparatus is a schematic structural diagram, and the apparatus mainly includes the following units:
an obtaining unit 901, configured to obtain sample point cloud data from point cloud data marked as a road surface of a road along a lane line of the road;
a sorting unit 902, configured to sort the sample point cloud data along the direction of the road, so as to obtain sorted sample point cloud data;
a dividing unit 903, configured to divide the sorted sample point cloud data into sample point cloud data blocks according to a preset interval distance along a road direction;
a counting unit 904, configured to count each sample point cloud data block to obtain distribution data of the sample point cloud data;
the identification unit 905 is configured to determine whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the road surface marker with the preset characteristic according to the distribution data of the sample point cloud data, and if yes, identify the road surface marker as the road surface marker with the preset characteristic.
Preferably, the obtaining unit 901 is specifically configured to: acquiring the central line of a lane line of a road; and acquiring point cloud data of which the distance from the central line of the lane line is not more than a preset distance threshold value from the point cloud data of one pavement marker marked as the road as sample point cloud data.
Preferably, the sorting unit 902 is specifically configured to: selecting one sample point cloud data from the sample point cloud data as reference point cloud data; selecting a target position point from the central line of the lane line of the road, wherein the target position point is positioned outside the coverage area of the sample point cloud data; obtaining the distance from each sample point cloud data to the target position point; along the direction of the road, if the target position point is positioned behind the reference point cloud data, sequencing the sample point cloud data according to the sequence that the distance from the target position point to the target position point is from small to large to obtain sequenced sample point cloud data; and along the direction of the road, if the target position point is positioned in front of the reference point cloud data, sequencing the sample point cloud data according to the sequence of the distances from the target position point to the target position point from large to small to obtain the sequenced sample point cloud data.
Preferably, the statistical unit 904 is specifically configured to count the number of sample point cloud data in each sample point cloud data block, mark the sample point cloud data block as data if the number is greater than a preset number threshold, or mark the sample point cloud data block as blank if the number is greater than the preset number threshold, traverse the marking result of each sample point cloud data block, and record L One times of continuous occurrence of the ith data mark when the ith continuous data mark occursiAnd recording when a consecutive blank mark appears j times, the j times of the consecutive blank marksNumber of times L ZerojThe L Onei、LZerojAnd i and j are natural numbers which are greater than or equal to 1 and are distribution data of the sample point cloud data.
Preferably, the road surface marks with the preset characteristics are horizontal and repeated road surface marks, and the identification unit 905 is specifically configured to judge each L Onei+1And L Zeroj+1Whether the following formula is satisfied:
OneLength*Lself-adaption1<LOnei+1<OneLength*Lself-adaption2；
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength*Lself-adaption2；
if the two road surface marks meet the preset cross-stripe-shaped repeated road surface mark, determining that the road surface mark meets the preset cross-stripe-shaped repeated road surface mark, wherein One L ength-L One1，ZeroLength＝LZero1L self-adaptation 1, L self-adaptation 2 are preset adaptive parameters, L self-adaptation 1<Lself-adaption2。
It should be noted that the present application may be implemented in software and/or a combination of software and hardware, for example, implemented using Application Specific Integrated Circuits (ASICs), general purpose computers or any other similar hardware devices. In one embodiment, the software programs of the present application may be executed by a processor to implement the steps or functions described above. Likewise, the software programs (including associated data structures) of the present application may be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Additionally, some of the steps or functions of the present application may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.
In addition, some of the present application may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or techniques in accordance with the present application through the operation of the computer. Program instructions which invoke the methods of the present application may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the present application comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or a solution according to the aforementioned embodiments of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Claims (10)
1. A method of identifying a pavement marking, the method comprising:
along a lane line of a road, obtaining sample point cloud data from point cloud data marked as a road surface mark of the road;
sequencing the sample point cloud data along the direction of the road to obtain sequenced sample point cloud data;
dividing the sorted sample point cloud data into sample point cloud data blocks according to a preset interval distance along the direction of a road;
counting each sample point cloud data block to obtain distribution data of the sample point cloud data;
and judging whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the pavement marker with the preset characteristics or not according to the distribution data of the sample point cloud data, and if so, identifying the pavement marker as the pavement marker with the preset characteristics.
2. The method of claim 1, wherein obtaining sample point cloud data from point cloud data along a lane line of a road marked as a pavement marker of the road comprises:
acquiring the central line of a lane line of a road;
and acquiring point cloud data of which the distance from the central line of the lane line is not more than a preset distance threshold value from the point cloud data of one pavement marker marked as the road as sample point cloud data.
3. The method of claim 2, wherein the ordering the sample point cloud data in the direction of the road specifically comprises:
selecting one sample point cloud data from the sample point cloud data as reference point cloud data;
selecting a target position point from the central line of the lane line of the road, wherein the target position point is positioned outside the coverage area of the sample point cloud data;
obtaining the distance from each sample point cloud data to the target position point;
along the direction of the road, if the target position point is positioned behind the reference point cloud data, sequencing the sample point cloud data according to the sequence that the distance from the target position point to the target position point is from small to large to obtain sequenced sample point cloud data;
and along the direction of the road, if the target position point is positioned in front of the reference point cloud data, sequencing the sample point cloud data according to the sequence of the distances from the target position point to the target position point from large to small to obtain the sequenced sample point cloud data.
4. The method according to any one of claims 1 to 3, wherein the performing statistics on each sample point cloud data block to obtain the distribution data of the sample point cloud data specifically comprises:
counting the number of sample point cloud data in each sample point cloud data block, if the number is larger than a preset number threshold value, marking the sample point cloud data block as data, otherwise, marking the sample point cloud data block as blank;
traversing the marking result of each sample point cloud data, and recording the continuous occurrence times of the ith data mark L One when the ith data mark appears continuouslyiAnd recording the number L Zero of consecutive occurrences of the jth blank mark when the jth occurrence of the consecutive blank markj；
The L Onei、LZerojAnd i and j are natural numbers which are greater than or equal to 1 and are distribution data of the sample point cloud data.
5. The method according to claim 4, wherein the pavement marker with the preset characteristics is a horizontal stripe-shaped and repeated pavement marker, and the determining whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the pavement marker with the preset characteristics according to the distribution information of the sample point cloud data specifically comprises:
determine each L Onei+1And L Zeroj+1Whether the following formula is satisfied:
OneLength*Lself-adaption1<LOnei+1<OneLength*Lself-adaption2；
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength*Lself-adaption2；
if the two road surface marks meet the preset cross-stripe-shaped repeated road surface mark, determining that the road surface mark meets the preset cross-stripe-shaped repeated road surface mark, wherein One L ength-L One1，ZeroLength＝LZero1L self-adaptation 1, L self-adaptation 2 are preset adaptive parameters, L self-adaptation 1<Lself-adaption2。
6. An apparatus for identifying pavement markings, the apparatus comprising:
an acquisition unit configured to acquire sample point cloud data from point cloud data of one road surface marker marked as a road along a lane line of the road;
the sorting unit is used for sorting the sample point cloud data along the direction of the road to obtain sorted sample point cloud data;
the dividing unit is used for dividing the sorted sample point cloud data into sample point cloud data blocks according to a preset interval distance along the direction of a road;
the statistical unit is used for carrying out statistics on each sample point cloud data block to obtain the distribution data of the sample point cloud data;
and the identification unit is used for judging whether the distribution rule of the sample point cloud data meets the distribution rule of the point cloud data of the road mark with the preset characteristics or not according to the distribution data of the sample point cloud data, and if so, identifying the road mark as the road mark with the preset characteristics.
7. The apparatus of claim 6, wherein the obtaining unit is specifically configured to: acquiring the central line of a lane line of a road; and acquiring point cloud data of which the distance from the central line of the lane line is not more than a preset distance threshold value from the point cloud data of one pavement marker marked as the road as sample point cloud data.
8. The apparatus of claim 7, wherein the ordering unit is specifically configured to: selecting one sample point cloud data from the sample point cloud data as reference point cloud data; selecting a target position point from the central line of the lane line of the road, wherein the target position point is positioned outside the coverage area of the sample point cloud data; obtaining the distance from each sample point cloud data to the target position point; along the direction of the road, if the target position point is positioned behind the reference point cloud data, sequencing the sample point cloud data according to the sequence that the distance from the target position point to the target position point is from small to large to obtain sequenced sample point cloud data; and along the direction of the road, if the target position point is positioned in front of the reference point cloud data, sequencing the sample point cloud data according to the sequence of the distances from the target position point to the target position point from large to small to obtain the sequenced sample point cloud data.
9. The device according to any One of claims 6 to 8, wherein the statistical unit is specifically configured to count the number of sample point cloud data in each sample point cloud data block, mark the sample point cloud data block as data if the number is greater than a preset number threshold, or mark the sample point cloud data block as blank if the number is greater than the preset number threshold, traverse the marking result of each sample point cloud data block, and record L One times of continuous occurrence of the ith data mark when the ith occurrence of the continuous data mark occursiAnd recording the number L Zero of consecutive occurrences of the jth blank mark when the jth occurrence of the consecutive blank markjThe L Onei、LZerojAnd i and j are natural numbers which are greater than or equal to 1 and are distribution data of the sample point cloud data.
10. The device according to claim 9, characterized in that the pavement marking of preset characteristics is a cross-bar and repetitive pavement marking, the identification unit being particularly adapted to:
determine each L Onei+1And L Zeroj+1Whether the following formula is satisfied:
OneLength*Lself-adaption1<LOnei+1<OneLength*Lself-adaption2
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength*Lself-adaption2
if the two road surface marks meet the preset cross-stripe-shaped repeated road surface mark, determining that the road surface mark meets the preset cross-stripe-shaped repeated road surface mark, wherein One L ength-L One1，ZeroLength＝LZero1L self-adaptation 1, L self-adaptation 2 are preset adaptive parameters, L self-adaptation 1<Lself-adaption2。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611201463.XA CN108241819B (en) | 2016-12-23 | 2016-12-23 | Method and device for identifying pavement marker |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611201463.XA CN108241819B (en) | 2016-12-23 | 2016-12-23 | Method and device for identifying pavement marker |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108241819A CN108241819A (en) | 2018-07-03 |
| CN108241819B true CN108241819B (en) | 2020-07-28 |
Family
ID=62703214
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611201463.XA Active CN108241819B (en) | 2016-12-23 | 2016-12-23 | Method and device for identifying pavement marker |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108241819B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111174777A (en) * | 2018-11-09 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Positioning method and device and electronic equipment |
| CN109343021B (en) * | 2018-12-03 | 2023-04-11 | 北京遥感设备研究所 | Spot mark condensation method |
| CN110458083B (en) * | 2019-08-05 | 2022-03-25 | 武汉中海庭数据技术有限公司 | Lane line vectorization method, device and storage medium |
| CN112825196B (en) * | 2019-11-20 | 2023-04-25 | 阿里巴巴集团控股有限公司 | Method and device for determining road sign, storage medium and editing platform |
| KR20210102182A (en) * | 2020-02-07 | 2021-08-19 | 선전 센스타임 테크놀로지 컴퍼니 리미티드 | Road marking recognition method, map generation method, and related products |
| CN112069899A (en) * | 2020-08-05 | 2020-12-11 | 深兰科技（上海）有限公司 | Road shoulder detection method and device and storage medium |
| CN113570004B (en) * | 2021-09-24 | 2022-01-07 | 西南交通大学 | Riding hot spot area prediction method, device, equipment and readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390169A (en) * | 2013-07-19 | 2013-11-13 | 武汉大学 | Sorting method of vehicle-mounted laser scanning point cloud data of urban ground objects |
| CN104197897A (en) * | 2014-04-25 | 2014-12-10 | 厦门大学 | Urban road marker automatic sorting method based on vehicle-mounted laser scanning point cloud |
| CN105069395A (en) * | 2015-05-17 | 2015-11-18 | 北京工业大学 | Road marking automatic identification method based on terrestrial three-dimensional laser scanning technology |
| CN105551082A (en) * | 2015-12-02 | 2016-05-04 | 百度在线网络技术（北京）有限公司 | Method and device of pavement identification on the basis of laser-point cloud |
-
2016
- 2016-12-23 CN CN201611201463.XA patent/CN108241819B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390169A (en) * | 2013-07-19 | 2013-11-13 | 武汉大学 | Sorting method of vehicle-mounted laser scanning point cloud data of urban ground objects |
| CN104197897A (en) * | 2014-04-25 | 2014-12-10 | 厦门大学 | Urban road marker automatic sorting method based on vehicle-mounted laser scanning point cloud |
| CN105069395A (en) * | 2015-05-17 | 2015-11-18 | 北京工业大学 | Road marking automatic identification method based on terrestrial three-dimensional laser scanning technology |
| CN105551082A (en) * | 2015-12-02 | 2016-05-04 | 百度在线网络技术（北京）有限公司 | Method and device of pavement identification on the basis of laser-point cloud |
Non-Patent Citations (1)
| Title |
|---|
| Lane Recognition Using On-vehicle LIDAR;Takashi Ogawa等;《2006 IEEE Intelligent Vehicles Symposium》;20060615;第540-545页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108241819A (en) | 2018-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108241819B (en) | Method and device for identifying pavement marker | |
| Biswas et al. | An automatic traffic density estimation using Single Shot Detection (SSD) and MobileNet-SSD | |
| CN105574543B (en) | A kind of vehicle brand type identifier method and system based on deep learning | |
| CN112016605B (en) | Target detection method based on corner alignment and boundary matching of bounding box | |
| US8433099B2 (en) | Vehicle discrimination apparatus, method, and computer readable medium storing program thereof | |
| CN108845569A (en) | Generate semi-automatic cloud method of the horizontal bend lane of three-dimensional high-definition mileage chart | |
| Feng et al. | Mixed road user trajectory extraction from moving aerial videos based on convolution neural network detection | |
| CN110619279A (en) | Road traffic sign instance segmentation method based on tracking | |
| EP4120123A1 (en) | Scan line-based road point cloud extraction method | |
| Borkar et al. | An efficient method to generate ground truth for evaluating lane detection systems | |
| CN113011331B (en) | Method and device for detecting whether motor vehicle gives way to pedestrians, electronic equipment and medium | |
| CN107808524B (en) | Road intersection vehicle detection method based on unmanned aerial vehicle | |
| CN111554105A (en) | Intelligent traffic identification and statistics method for complex traffic intersection | |
| CN104317583A (en) | Road congestion optimization algorithm based on grid theory | |
| Zhang et al. | End to end video segmentation for driving: Lane detection for autonomous car | |
| Huang et al. | Stereovision-based object segmentation for automotive applications | |
| CN110796230A (en) | Method, equipment and storage medium for training and using convolutional neural network | |
| Poggenhans et al. | A universal approach to detect and classify road surface markings | |
| CN111797738A (en) | Multi-target traffic behavior fast extraction method based on video identification | |
| Revilloud et al. | A lane marker estimation method for improving lane detection | |
| CN113257005B (en) | Traffic flow statistical method based on correlation measurement | |
| Balci et al. | Front-View Vehicle Damage Detection using Roadway Surveillance Camera Images. | |
| Yu et al. | Efficient lane detection using deep lane feature extraction method | |
| Khosravi et al. | Vehicle speed and dimensions estimation using on-road cameras by identifying popular vehicles | |
| CN111325811B (en) | Lane line data processing method and processing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200506 Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Alibaba (China) Co.,Ltd. Address before: 102200, No. 18, No., Changsheng Road, Changping District science and Technology Park, Beijing, China. 1-5 Applicant before: AUTONAVI SOFTWARE Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230511 Address after: 102299 floor 1-5, block B1, 18 Changsheng Road, science and Technology Park, Changping District, Beijing Patentee after: AUTONAVI SOFTWARE Co.,Ltd. Address before: 310052 room 508, 5th floor, building 4, No. 699 Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Patentee before: Alibaba (China) Co.,Ltd. |
|
| TR01 | Transfer of patent right |