CN114862808B - Determination method, device, equipment and storage medium for precision of dotted line frame - Google Patents

Determination method, device, equipment and storage medium for precision of dotted line frame Download PDF

Info

Publication number
CN114862808B
CN114862808B CN202210542312.XA CN202210542312A CN114862808B CN 114862808 B CN114862808 B CN 114862808B CN 202210542312 A CN202210542312 A CN 202210542312A CN 114862808 B CN114862808 B CN 114862808B
Authority
CN
China
Prior art keywords
dotted line
line frame
length
point cloud
frame
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
Application number
CN202210542312.XA
Other languages
Chinese (zh)
Other versions
CN114862808A (en
Inventor
鲁荣荣
罗玮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ecarx Hubei Tech Co Ltd
Original Assignee
Ecarx Hubei Tech Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ecarx Hubei Tech Co Ltd filed Critical Ecarx Hubei Tech Co Ltd
Priority to CN202210542312.XA priority Critical patent/CN114862808B/en
Publication of CN114862808A publication Critical patent/CN114862808A/en
Application granted granted Critical
Publication of CN114862808B publication Critical patent/CN114862808B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/05—Geographic models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/68—Analysis of geometric attributes of symmetry
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Software Systems (AREA)
  • Quality & Reliability (AREA)
  • Remote Sensing (AREA)
  • Computer Graphics (AREA)
  • Image Analysis (AREA)

Abstract

The application provides a method, a device, equipment and a storage medium for determining the precision of a broken line frame, which relate to the technical field of map construction and comprise the following steps: acquiring point cloud data comprising a first dotted line frame and a second dotted line frame, and determining the point cloud reflectivity contrast ratio taking the longitudinal edge of the first dotted line frame as a boundary according to the point cloud data; determining the reflectivity contrast confidence corresponding to the first dotted line frame according to the point cloud reflectivity contrast; acquiring a length consistency confidence corresponding to the first dotted line frame based on the length of the transverse edge of the first dotted line frame and the length of the transverse edge of the second dotted line frame; and determining the accuracy of the first dotted line frame according to the reflectivity contrast confidence and the length consistency confidence. The method and the device can accurately determine the precision of the dotted line frame, thereby improving the production efficiency of the high-precision map.

Description

Determination method, device, equipment and storage medium for precision of dotted line frame
Technical Field
The present application relates to the field of map construction technologies, and in particular, to a method, an apparatus, a device, and a storage medium for determining accuracy of a dashed frame.
Background
With the development of technology, the application of high-precision maps is becoming more and more widespread. In the process of constructing a high-precision map, because some components in the acquisition system have unavoidable errors, such as deviation of calibration parameters, ranging errors of equipment such as a camera and a laser radar, and bottlenecks of a perception segmentation model, how to extract a high-precision dotted line frame from an image and a laser point cloud becomes a challenge.
At present, in the drawing process of the high-precision map, a manual interaction link is generally added, and the boundaries of the low-precision dashed frame are checked and corrected one by one in a manual drawing checking mode so as to meet the requirement of the high-precision map on element precision. However, the above-described method of manually inspecting the broken line frame greatly reduces the production efficiency of the high-precision map.
Disclosure of Invention
The application provides a method, a device, equipment and a storage medium for determining the precision of a broken line frame, which are used for solving the problem that the production efficiency of a high-precision map is greatly reduced by manually checking the broken line frame.
In a first aspect, the present application provides a method for determining the accuracy of a dashed box, including:
acquiring point cloud data comprising a dotted line frame, wherein the dotted line frame comprises a first dotted line frame and a second dotted line frame, the middle points of two longitudinal edges of the first dotted line frame form a target line segment, the center distance between the second dotted line frame and the first dotted line frame is smaller than or equal to a first distance threshold, and the projection distance between the center of the second dotted line frame and the target line segment is smaller than or equal to a second distance threshold;
Determining a point cloud reflectivity contrast ratio taking the longitudinal edge of the first dotted line frame as a boundary according to the point cloud data;
determining the reflectivity contrast confidence corresponding to the first dotted line frame according to the point cloud reflectivity contrast;
acquiring a length consistency confidence corresponding to the first dotted line frame based on the length of the transverse edge of the first dotted line frame and the length of the transverse edge of the second dotted line frame, wherein the length of the transverse edge of the first dotted line frame is greater than the length of the longitudinal edge of the first dotted line frame;
and determining the accuracy of the first dotted line frame according to the reflectivity contrast confidence and the length consistency confidence.
Optionally, determining, according to the point cloud data, a point cloud reflectivity contrast with a longitudinal edge of the first dashed box as a boundary line includes: acquiring midpoint coordinates corresponding to the midpoint according to the point cloud data; according to the midpoint coordinates, acquiring a target point cloud with a distance from the midpoint being smaller than or equal to a third distance threshold value through a preset search algorithm, wherein the third distance threshold value is larger than the length of the longitudinal edge of the first dotted line frame; and determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud.
Optionally, determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud includes: determining a first area and a second area which are used for taking two transverse sides of a first dotted line frame as boundaries and taking the longitudinal sides as boundaries and contain target point clouds for each of the two longitudinal sides, wherein the first area is a rectangular area contained in the first dotted line frame, and the second area is a rectangular area which is not contained in the first dotted line frame; acquiring a first average reflectivity intensity value corresponding to the first region according to the reflectivity intensity value of the target point cloud contained in the first region; acquiring a second average reflectivity intensity value corresponding to the second region according to the reflectivity intensity value of the target point cloud contained in the second region; the point cloud reflectance contrast is determined as the difference between the first average reflectance intensity value and the second average reflectance intensity value.
Optionally, determining the confidence level of the reflectivity contrast corresponding to the first dotted line box according to the reflectivity contrast of the point cloud includes: determining a target point cloud reflectivity contrast according to the point cloud reflectivity contrast taking each of two longitudinal edges of the first dotted line frame as a boundary; and determining the reflectivity contrast confidence coefficient through a first preset function according to the reflectivity contrast of the cloud of the target point and the contrast threshold.
Optionally, based on the length of the lateral edge of the first dashed frame and the length of the lateral edge of the second dashed frame, obtaining the length consistency confidence corresponding to the first dashed frame includes: acquiring a reflectivity contrast confidence coefficient corresponding to the second dotted line frame; acquiring a weighted average length according to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence corresponding to the second dotted line frame; determining the length deviation of the first dashed frame as the difference between the length of the transverse edge of the first dashed frame and the weighted average length; and acquiring the confidence coefficient of the length consistency through a second preset function according to the length deviation and a deviation sensitivity factor, wherein the deviation sensitivity factor is used for representing the sensitivity degree to the deviation.
Optionally, determining the accuracy of the first dashed box according to the reflectivity contrast confidence and the length consistency confidence includes: weighting and fusing the reflectivity contrast confidence coefficient and the length consistency confidence coefficient to obtain a target confidence coefficient; if the target confidence coefficient is greater than or equal to the confidence coefficient threshold value, determining the precision of the first dotted line frame to be high; if the target confidence level is less than the confidence level threshold, determining that the accuracy of the first dashed box is low.
In a second aspect, the present application provides a determination apparatus for determining accuracy of a dashed frame, including:
the first acquisition module is used for acquiring point cloud data comprising a dotted line frame, the dotted line frame comprises a first dotted line frame and a second dotted line frame, the middle points of two longitudinal sides of the first dotted line frame form a target line segment, the center distance between the second dotted line frame and the first dotted line frame is smaller than or equal to a first distance threshold, and the projection distance from the center of the second dotted line frame to the target line segment is smaller than or equal to a second distance threshold;
the first determining module is used for determining the point cloud reflectivity contrast ratio taking the longitudinal edge of the first dotted line frame as a boundary according to the point cloud data;
the second determining module is used for determining the reflectivity contrast confidence coefficient corresponding to the first dotted line frame according to the point cloud reflectivity contrast;
The second acquisition module is used for acquiring the length consistency confidence corresponding to the first dotted frame based on the length of the transverse edge of the first dotted frame and the length of the transverse edge of the second dotted frame, wherein the length of the transverse edge of the first dotted frame is greater than the length of the longitudinal edge of the first dotted frame;
and the third determining module is used for determining the accuracy of the first dotted line frame according to the reflectivity contrast confidence and the length consistency confidence.
Optionally, the first determining module is specifically configured to: acquiring midpoint coordinates corresponding to the midpoint according to the point cloud data; according to the midpoint coordinates, acquiring a target point cloud with a distance from the midpoint being smaller than or equal to a third distance threshold value through a preset search algorithm, wherein the third distance threshold value is larger than the length of the longitudinal edge of the first dotted line frame; and determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud.
Optionally, the first determining module is specifically configured to, when determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud: determining a first area and a second area which are used for taking two transverse sides of a first dotted line frame as boundaries and taking the longitudinal sides as boundaries and contain target point clouds for each of the two longitudinal sides, wherein the first area is a rectangular area contained in the first dotted line frame, and the second area is a rectangular area which is not contained in the first dotted line frame; acquiring a first average reflectivity intensity value corresponding to the first region according to the reflectivity intensity value of the target point cloud contained in the first region; acquiring a second average reflectivity intensity value corresponding to the second region according to the reflectivity intensity value of the target point cloud contained in the second region; the point cloud reflectance contrast is determined as the difference between the first average reflectance intensity value and the second average reflectance intensity value.
Optionally, the second determining module is specifically configured to: determining a target point cloud reflectivity contrast according to the point cloud reflectivity contrast taking each of two longitudinal edges of the first dotted line frame as a boundary; and determining the reflectivity contrast confidence coefficient through a first preset function according to the reflectivity contrast of the cloud of the target point and the contrast threshold.
Optionally, the second obtaining module is specifically configured to: acquiring a reflectivity contrast confidence coefficient corresponding to the second dotted line frame; acquiring a weighted average length according to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence corresponding to the second dotted line frame; determining the length deviation of the first dashed frame as the difference between the length of the transverse edge of the first dashed frame and the weighted average length; and acquiring the confidence coefficient of the length consistency through a second preset function according to the length deviation and a deviation sensitivity factor, wherein the deviation sensitivity factor is used for representing the sensitivity degree to the deviation.
Optionally, the third determining module is specifically configured to: weighting and fusing the reflectivity contrast confidence coefficient and the length consistency confidence coefficient to obtain a target confidence coefficient; if the target confidence coefficient is greater than or equal to the confidence coefficient threshold value, determining the precision of the first dotted line frame to be high; if the target confidence level is less than the confidence level threshold, determining that the accuracy of the first dashed box is low.
In a third aspect, the present application provides an electronic device comprising: a processor, a memory communicatively coupled to the processor;
the memory stores computer-executable instructions;
the processor executes computer-executable instructions stored in the memory to implement a method for determining the accuracy of a dashed box according to the first aspect of the present application.
In a fourth aspect, the present application provides a computer readable storage medium having stored therein computer program instructions which, when executed by a processor, implement a method for determining the accuracy of a dashed box according to the first aspect of the present application.
In a fifth aspect, the present application provides a computer program product comprising a computer program which when executed by a processor implements a method of determining the accuracy of a dashed box according to the first aspect of the present application.
According to the method, the device, the equipment and the storage medium for determining the precision of the dotted line frame, the point cloud reflectivity contrast taking the longitudinal edge of the first dotted line frame as a boundary is determined according to the point cloud data comprising the first dotted line frame and the second dotted line frame, and further the reflectivity contrast confidence coefficient corresponding to the first dotted line frame is determined, namely, a first association relation is established between the precision of the dotted line frame and the surrounding physical environment characteristics; meanwhile, according to a second dotted line frame which is the same as the first dotted line frame, acquiring the length consistency confidence coefficient corresponding to the first dotted line frame, namely establishing a second association relation between the precision of the dotted line frame and the surrounding physical environment characteristics; and determining the accuracy of the first dotted line frame according to the reflectivity contrast confidence and the length consistency confidence. According to the method and the device, the accuracy of the dotted line frame is mapped into the confidence value according to the corresponding reflectivity contrast confidence and length consistency confidence of the first dotted line frame, so that the quick quality inspection and fixed-point overhaul of the dotted line frame are realized based on the confidence, the accuracy of the dotted line frame can be accurately determined, the production efficiency of a high-accuracy map is improved, and the competitiveness of a high-accuracy map product is further improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions of the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort to a person skilled in the art.
FIG. 1 is a schematic diagram of a qualified dashed box and three unqualified dashed boxes according to an embodiment of the present application;
fig. 2 is a schematic diagram of an application scenario provided in an embodiment of the present application;
FIG. 3 is a flowchart of a method for determining accuracy of a dashed box according to an embodiment of the present application;
FIG. 4 is a flowchart of a method for determining accuracy of a dashed box according to another embodiment of the present application;
fig. 5 (a) is a schematic diagram of a first dashed box containing point cloud data according to an embodiment of the present application;
FIG. 5 (b) is a schematic diagram illustrating the acquisition of the point cloud reflectivity contrast with the boundary line V1V2 of the first dashed box according to an embodiment of the present application;
FIG. 5 (c) is a schematic diagram illustrating the acquisition of the point cloud reflectivity contrast with the boundary line V3V4 of the first dashed box according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a second dashed box provided in an embodiment of the present application and assembled with the first dashed box;
FIG. 7 (a) is a schematic diagram illustrating distribution of confidence values of a dashed box in a scene according to an embodiment of the present application;
FIG. 7 (b) is a schematic diagram illustrating distribution of confidence values of a dashed box in a scene according to another embodiment of the present application;
FIG. 8 is a schematic diagram of a device for determining accuracy of a dashed box according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In the technical scheme of the application, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the information such as financial data or user data are in accordance with the regulations of related laws and regulations, and the public welfare is not violated.
The dashed box is a very common ground traffic sign, and most roads contain the dashed box and occupy a relatively high area. At present, in the drawing process of the high-precision map, a manual interaction link is generally added, and the boundaries of the low-precision dashed frame are checked and corrected one by one in a manual drawing checking mode so as to meet the requirement of the high-precision map on element precision. Specifically, all spliced three-dimensional point clouds and the dotted line frame are overlapped and checked together by manpower, and whether the precision of the dotted line frame meets the standard is judged by observing whether the boundary of the dotted line frame just frames a corresponding highlight region in the three-dimensional point clouds. Exemplary, fig. 1 is a schematic diagram of one qualified precision dashed box and three unqualified precision dashed boxes according to an embodiment of the present application, as shown in fig. 1, the qualified precision dashed box 101 is a high-precision dashed box, and the boundary of the dashed box 101 just frames a corresponding highlight region in the three-dimensional point cloud; the dashed boxes 102, 103 and 104 with unqualified precision are all low-precision dashed boxes, and the boundaries of the dashed boxes 102, 103 and 104 are too long to precisely frame the corresponding highlight areas in the three-dimensional point cloud. However, the above-described method of manually inspecting the broken line frame greatly reduces the production efficiency of the high-precision map.
Based on the above problems, the application provides a method, a device, equipment and a storage medium for determining the precision of a dotted line frame, which are characterized in that the precision of the dotted line frame is associated with the surrounding physical environment characteristics, and the physical environment characteristics are mapped into confidence values by utilizing a reasonable nonlinear function, so that the quick quality inspection and fixed-point maintenance of the dotted line frame are realized based on the confidence level, the production efficiency of the elements of the dotted line frame is accelerated, and the production efficiency of a high-precision map and the competitiveness of a high-precision map product are further improved.
In the following, first, an application scenario of the solution provided by the present application is illustrated.
Fig. 2 is a schematic diagram of an application scenario provided in an embodiment of the present application. As shown in fig. 2, in the present application scenario, the server 202 acquires the position information of the autonomous vehicle 201 in real time, and sends a high-precision map within a preset geographical range to the autonomous vehicle 201 based on the position information of the autonomous vehicle 201. The autonomous vehicle 201 automatically recognizes a broken line on the road 203 according to the high-precision map, and runs on the corresponding lane. Wherein the high-precision map is generated by the server 202. In the process of constructing the high-precision map by the server 202, the server 202 determines the precision of the dotted line frame according to the point cloud data comprising the dotted line frame, and performs fixed-point maintenance on the low precision so as to meet the requirement of the high-precision map on the element precision. For how the server 202 determines the accuracy of the dotted frame from the point cloud data containing the dotted frame, see the schemes of the embodiments described below.
It should be noted that fig. 2 is only a schematic diagram of an application scenario provided by the embodiment of the present application, and the embodiment of the present application does not limit the devices included in fig. 2, and does not limit the positional relationship between the devices in fig. 2. For example, in the application scenario shown in fig. 2, a data storage device may be further included, where the data storage device may be an external memory with respect to the server 202, or may be an internal memory integrated into the server 202.
Next, the dotted line is introduced by a specific embodimentA method for determining frame accuracy. It should be noted that, the scheme of the determination method for the precision of the dashed box provided by the embodiment of the application is input as a plurality of spliced three-dimensional point clouds and vectorization results of the dashed box on the corresponding road section. Illustratively, a number of stitched three-dimensional point cloud sets P are represented as: p= { P 1 ,P 2 ,…,P N Each three-dimensional point cloud (e.g. P N ) Contains a number of three-dimensional coordinate points with reflectivity intensity values (such as obtained by lidar scanning). The set of dashed boxes Q for all the accuracies to be determined is denoted as: q= { B 1 ,B 2 ,…,B i ,…B M }, wherein B is i A dashed box representing each precision to be determined, each dashed box being rectangular and being generally represented by 5 points, the order of the 5 points being stored clockwise or counterclockwise, the beginning and end being identical, each of the 5 points being understood as B i Is defined by the vertex of (a); each of the 5 points is a point in three-dimensional space and can be represented by three-dimensional coordinates; specifically, B i ＝{v i1 ,v i2 ,v i3 ,v i4 ,v i5 -wherein each point is represented as: v ij ＝{x ij ,y ij ,z ij I.e. each point is represented by three-dimensional coordinates. In the embodiment of the application, any one of the dashed frames B in the dashed frame set Q with the precision to be determined i For example, a scheme of a method for determining accuracy of a dashed box provided by the embodiment of the application is described.
Fig. 3 is a flowchart of a method for determining accuracy of a dashed box according to an embodiment of the present application. The method of the embodiment of the application can be applied to the electronic equipment, and the electronic equipment can be a server or a server cluster and the like. As shown in fig. 3, the method of the embodiment of the present application includes:
s301, acquiring point cloud data comprising a dotted line frame, wherein the dotted line frame comprises a first dotted line frame and a second dotted line frame.
The middle points of the two longitudinal sides of the first dotted line frame form a target line segment, the center distance between the second dotted line frame and the first dotted line frame is smaller than or equal to a first distance threshold, and the projection distance between the center of the second dotted line frame and the target line segment is smaller than or equal to a second distance threshold.
In the embodiment of the present application, the point cloud data including the dashed box may be input by the user to the electronic device executing the embodiment of the method, or may be sent by other devices to the electronic device executing the embodiment of the method. In this step, the first dotted frame is a dotted frame B of the set of dotted frames Q, whose accuracy is to be determined i The method comprises the steps of carrying out a first treatment on the surface of the Since the dotted lines on the road are usually printed according to a certain rule, for example, the dotted lines within a certain range of the same lane need to keep the same length and the same interval, the second dotted line box can be understood as a dotted line box of the same group as the first dotted line box. Illustratively, one can gather p= { P from the three-dimensional point cloud described above 1 ,P 2 ,…,P N Acquiring point cloud data comprising a first dotted frame and a second dotted frame, e.g. using P k And (3) representing. Wherein P is k The first dotted line frame is a dotted line frame B i ；P k The second dashed frame and the first dashed frame B i The center distance of the second dashed box to the midpoint of the two longitudinal sides of the first dashed box is less than or equal to a first distance threshold, the first distance threshold being indicated as d, for example, and the second distance threshold being indicated as epsilon, for example. Specifically, the acquired plurality of second dashed-line box sets H are expressed as h= { B, for example i1 ,B i2 ,…,B ij ,…,B in Second dashed box B for any one of the second dashed box sets H ij It satisfies the following two conditions:
(1)||(center(B i )-center(B ij ) D is equal to or less than d, and represents a second dotted line box B ij And a first dotted line frame B i Is less than or equal to a first distance threshold d;
Wherein center (B) i )＝0.25(v i1 ,v i2 ,v i3 ,v i4 )，center(B ij )＝0.25(v ij1 ,v ij2 ,v ij3 ,v ij4 ) That is, four vertices of each dotted frame are subjected to coordinate averaging to obtain coordinates of the center of the dotted frame.
(2)proj(v i ,center(B ij ) Epsilon is not less than and indicates a second dotted line box B ij The projection distance of a target line segment formed by the center of the first dashed line frame to the middle point of the two longitudinal sides is smaller than or equal to a second distance threshold epsilon;
wherein, it is assumed that a first dotted line frame B i The three-dimensional coordinates of the midpoints of the two longitudinal edges of (2) are respectively v s1 And v s2 And then v is expressed as i ＝v s1 -v s2 Representing a first dashed box B i A target line segment formed by the midpoints of the two longitudinal edges of (a), which may also be referred to as a directed line segment; epsilon can be preset as required, for example, a value smaller than half the width of the lane can be adopted, and specifically, for example, the value is 1 meter; for condition (2), it is assumed that, illustratively, center (B) is represented by p ij ) Then
S302, determining the point cloud reflectivity contrast ratio taking the longitudinal edge of the first dotted line frame as a boundary according to the point cloud data.
For example, referring to fig. 1, the boundary of the qualified dashed-line frame 101 is just the boundary of the high-reflectivity and low-reflectivity point cloud, so that the degree of precision of the dashed-line frame has a strong correlation with the degree of contrast of the reflectivity of the point cloud on both sides of the boundary of the dashed-line frame, and the degree of contrast of the reflectivity of the point cloud on the boundary of the dashed-line frame can be used as an evaluation index of the degree of precision of the dashed-line frame. In this step, after the point cloud data including the dotted line frame is obtained, the point cloud reflectance contrast ratio using the longitudinal side of the first dotted line frame as the boundary line may be determined based on the three-dimensional coordinates and the reflectance intensity value of each point in the point cloud data. It will be appreciated that for both longitudinal sides of the first dashed box, a point cloud reflectivity contrast with each longitudinal side as a boundary can be obtained. For how to determine the point cloud reflectivity contrast with the longitudinal edge of the first dashed frame as the boundary according to the point cloud data, reference may be made to the following embodiments, which will not be repeated here.
S303, determining the confidence coefficient of the reflectivity contrast corresponding to the first dotted line frame according to the reflectivity contrast of the point cloud.
In this step, after the point cloud reflectance contrast with the longitudinal side of the first dotted frame as the boundary is obtained, the reflectance contrast confidence corresponding to the first dotted frame may be determined according to the point cloud reflectance contrast. For how to determine the confidence of the reflectivity contrast corresponding to the first dashed box according to the point cloud reflectivity contrast, reference may be made to the following embodiments, which are not described herein.
S304, based on the length of the transverse edge of the first dotted line frame and the length of the transverse edge of the second dotted line frame, the length consistency confidence corresponding to the first dotted line frame is obtained.
Wherein the length of the lateral side of the first dashed frame is greater than the length of the longitudinal side of the first dashed frame.
It will be appreciated that the reflectivity contrast confidence is an indication of the accuracy of the individual dashed boxes from a local environmental point of view, and that since the dashed lines on a road are typically printed according to a certain rule, for example, the dashed lines within a certain range of the same lane need to be kept at the same length, the same spacing. Therefore, the consistency of the lengths of the same group of dashed boxes can be used as another index reflecting the precision of the dashed boxes. In this step, the first dotted frame is illustratively rectangular, the lateral side of the first dotted frame may be understood as the length of the rectangle, and the longitudinal side of the first dotted frame may be understood as the width of the rectangle, and the length of the lateral side of the first dotted frame is greater than the length of the longitudinal side of the first dotted frame. In this step, the length consistency confidence corresponding to the first dashed frame may be obtained based on the length of the lateral edge of the first dashed frame and the length of the lateral edge of the second dashed frame. For how to obtain the length consistency confidence corresponding to the first dashed frame based on the length of the lateral edge of the first dashed frame and the length of the lateral edge of the second dashed frame, reference may be made to the subsequent embodiments, which are not described herein.
S305, determining the accuracy of the first dotted line frame according to the reflectivity contrast confidence and the length consistency confidence.
In this step, after the confidence level of the reflectivity contrast and the confidence level of the length consistency corresponding to the first dashed box are obtained, the accuracy of the first dashed box may be determined according to the confidence level of the reflectivity contrast and the confidence level of the length consistency. For example, corresponding weights may be respectively assigned to the reflectivity contrast confidence coefficient and the length consistency confidence coefficient corresponding to the first dashed box, so as to obtain a final target confidence coefficient, and further determine the accuracy of the first dashed box according to the size of the target confidence coefficient. For how to determine the accuracy of the first dashed box according to the confidence of the reflectivity contrast and the confidence of the length consistency, reference may be made to the following embodiments, which are not described herein.
It is understood that for each of the dashed boxes to be determined in the dashed box set Q, the accuracy of each of the dashed boxes may be determined in turn by performing the steps S301 to S305. After the precision of the dashed frame is determined, the high-precision dashed frame can be inspected without checking, and the low-precision dashed frame can be subjected to fixed-point maintenance, namely, the low-precision dashed frame boundary is corrected, so that the requirement of the high-precision map on the element precision is met. Based on the determination method of the precision of the dotted line frame provided by the embodiment of the application, the quality inspection of the dotted line frame can be improved by 20 to 30 percent, so that the production efficiency of the high-precision map can be improved.
According to the method for determining the accuracy of the dotted line frame, which is provided by the embodiment of the application, according to the point cloud data comprising the first dotted line frame and the second dotted line frame, the point cloud reflectivity contrast taking the longitudinal edge of the first dotted line frame as a boundary is determined, and further the reflectivity contrast confidence corresponding to the first dotted line frame is determined, namely, a first association relationship is established between the accuracy of the dotted line frame and the surrounding physical environment characteristics; meanwhile, according to a second dotted line frame which is the same as the first dotted line frame, acquiring the length consistency confidence coefficient corresponding to the first dotted line frame, namely establishing a second association relation between the precision of the dotted line frame and the surrounding physical environment characteristics; and determining the accuracy of the first dotted line frame according to the reflectivity contrast confidence and the length consistency confidence. According to the embodiment of the application, the accuracy of the dotted line frame is mapped into the confidence value according to the corresponding reflectivity contrast confidence and length consistency confidence of the first dotted line frame, so that the quick quality inspection and fixed-point maintenance of the dotted line frame are realized based on the confidence, the accuracy of the dotted line frame can be accurately determined, the production efficiency of a high-accuracy map is improved, and the competitiveness of a high-accuracy map product is further improved.
Fig. 4 is a flowchart of a method for determining accuracy of a dashed box according to another embodiment of the present application. On the basis of the above embodiment, the method for determining the precision of the dashed box in the embodiment of the present application is further described.
As shown in fig. 4, the method of the embodiment of the present application may include:
s401, acquiring point cloud data comprising a dotted line frame, wherein the dotted line frame comprises a first dotted line frame and a second dotted line frame.
The middle points of the two longitudinal sides of the first dotted line frame form a target line segment, the center distance between the second dotted line frame and the first dotted line frame is smaller than or equal to a first distance threshold, and the projection distance between the center of the second dotted line frame and the target line segment is smaller than or equal to a second distance threshold.
A specific description of this step may be referred to as a related description of S301 in the embodiment shown in fig. 3, which is not repeated here.
In the embodiment of the present application, the step S302 in fig. 3 may further include the following three steps S402 to S404:
and S402, acquiring midpoint coordinates corresponding to midpoints of two longitudinal sides of the first dotted line frame according to the point cloud data.
Fig. 5 (a) is a schematic diagram of a first dashed box included in point cloud data according to an embodiment of the present application, as shown in fig. 5 (a), a first dashed box B with accuracy to be determined i Respectively with V at four vertexes of (2) 1 、V 2 、V 3 And V 4 Indicated as if side V 1 V 2 Is less than the length of the side V 2 V 3 Length of (V) then side 1 V 2 With edge V 3 V 4 Is a first dotted line frame B i Is a short side; if edge V 1 V 2 Is longer than the edge V 2 V 3 Length of (V) then side 1 V 4 With edge V 2 V 3 Is a first dotted line frame B i Is provided. FIG. 5 (b) shows an embodiment of the present application for acquiring a first dashed boxSide V 1 V 2 As shown in fig. 5 (B), the point cloud reflectivity contrast is a schematic diagram of the boundary line, and is based on a first dotted line box B to be determined in accuracy shown in fig. 5 (a) i By edge V 1 V 2 Is a first dotted line frame B i And one short side of (C) is a side V 1 V 2 The midpoint coordinate corresponding to the midpoint of (a) is v s1 ＝0.5(v i1 +v i2 ) I.e. V in FIG. 5 (b) s1 . FIG. 5 (c) shows an embodiment of the present application for acquiring the edge V of the first dashed box 3 V 4 As shown in fig. 5 (c), the point cloud reflectivity contrast is schematically represented as a boundary line, and is based on a first dotted line box B to be determined in accuracy shown in fig. 5 (a) i By edge V 3 V 4 Is a first dotted line frame B i And one short side of (C) is a side V 3 V 4 The midpoint coordinate corresponding to the midpoint of (a) is v s2 ＝0.5(v i3 +v i4 ) I.e. V in FIG. 5 (c) s2 。
S403, acquiring a target point cloud with the distance from the midpoint being smaller than or equal to a third distance threshold value through a preset search algorithm according to the midpoint coordinates.
Wherein the third distance threshold is greater than the length of the longitudinal edge of the first dashed box.
The third distance threshold may be determined experimentally, for example, denoted by r, which is greater than the short side length of the first dashed box. The preset searching algorithm is, for example, using point cloud data P k A K-dimensional search tree (kd-Tree) is constructed for input for fast querying of nearest neighbors. In this step, after the midpoint coordinates corresponding to the midpoints of the two longitudinal sides of the first dashed box are obtained, v s1 In the center, searching a target point cloud with the distance smaller than or equal to r through the constructed kd-tree, and recording as a set G= { G v s1 R is equal to or less than R, g is P k The method comprises the steps of carrying out a first treatment on the surface of the Based on the mode of obtaining the set G, v s2 In the center, the cloud of target points with the distance less than or equal to r can be searched through the constructed kd-tree.
S404, determining the point cloud reflectivity contrast taking the longitudinal edge of the first dotted line frame as a boundary according to the reflectivity intensity value of the target point cloud.
In this step, after the target point cloud is obtained, the point cloud reflectivity contrast with the longitudinal edge of the first dotted frame as the boundary may be determined according to the reflectivity intensity value of the target point cloud.
Further, optionally, determining, according to the reflectance intensity value of the target point cloud, a point cloud reflectance contrast with a longitudinal edge of the first dashed line frame as a boundary line includes: determining a first area and a second area which are used for taking two transverse sides of a first dotted line frame as boundaries and taking the longitudinal sides as boundaries and contain target point clouds for each of the two longitudinal sides, wherein the first area is a rectangular area contained in the first dotted line frame, and the second area is a rectangular area which is not contained in the first dotted line frame; acquiring a first average reflectivity intensity value corresponding to the first region according to the reflectivity intensity value of the target point cloud contained in the first region; acquiring a second average reflectivity intensity value corresponding to the second region according to the reflectivity intensity value of the target point cloud contained in the second region; the point cloud reflectance contrast is determined as the difference between the first average reflectance intensity value and the second average reflectance intensity value.
Illustratively, referring to FIG. 5 (B), a first dashed box B i Is a group of boundaries by the two long sides of V 1 V 2 Generating two regions S for dividing line in1 And S is equal to out1 ；S in1 Is a first dotted line frame B i Small rectangular area of the interior, S out1 Is symmetrical and not included in the first dotted line box B i Outer region of (S) in1 And S is equal to out1 Sharing one edge, i.e. edge V 1 V 2 ，S in1 And S is equal to out1 The length of the other side of (2) is denoted by r; fall in set G at S in1 The points in the region are denoted as G in1 Fall to S out1 The points in the region are denoted as G out1 G can be obtained separately in1 And G out1 Average reflectivity intensity value I of a medium three-dimensional point cloud in1 And I out1 ；I in1 I is the first average reflectivity intensity value corresponding to the first region, I out1 I.e. the second average reflectance intensity value corresponding to the second region, typically I in1 ≥I out1 That is, the average reflectance intensity value of the points inside the dashed box is higher than the average reflectance intensity value of the external ground background; according to I in1 And I out1 Can be obtained with a short side V 1 V 2 The point cloud reflectivity contrast for the boundary is ΔI 1 ＝I in1 -I out1 . Based on the obtained delta I 1 With reference to 5 (c), a frame B with a first dotted line can be obtained i Another short side V of (2) 3 V 4 Two areas S being dividing lines in2 And S is equal to out2 Further, S can be obtained in2 Corresponding first average reflectance intensity value I in2 S and S out2 Corresponding second average reflectance intensity value I out2 The method comprises the steps of carrying out a first treatment on the surface of the According to I in2 And I out2 Can be obtained with a short side V 3 V 4 The point cloud reflectivity contrast for the boundary is ΔI 2 ＝I in2 -I out2 。
In the embodiment of the present application, the step S303 in fig. 3 may further include two steps S405 and S406 as follows:
s405, determining the target point cloud reflectivity contrast according to the point cloud reflectivity contrast taking each of the two longitudinal sides of the first dotted line frame as a boundary.
In this step, the higher the point cloud reflectance contrast, the higher the accuracy of specifying the boundary line, and therefore, based on the example of step S404, Δi is obtained 1 And DeltaI 2 After that, ΔI can be calculated 1 And DeltaI 2 As the first dotted line box B i Target point cloud reflectance contrast, Δi i ＝min(ΔI 1 ,ΔI 2 )。
S406, determining the reflectivity contrast confidence coefficient corresponding to the first dotted line frame through a first preset function according to the reflectivity contrast of the target point cloud and the contrast threshold.
In this step, the contrast threshold is e.g. τ I It is indicated that different contrast thresholds may be set according to different lidars. Illustratively, based on a number of statistical experimental contrast analyses, when the point cloud reflectance contrast is greater than the contrast threshold τ I And when the map is displayed, the dotted line frame has higher precision and meets the drawing standard of the high-precision map. After the target point cloud reflectivity contrast delta I is obtained i Then, the cloud reflectivity contrast delta I of the target point can be obtained i And a contrast threshold τ I Determining the reflectivity contrast confidence corresponding to the first dotted line box through the following first preset function:
it should be noted that, the present application does not specifically limit the first preset function, and the first preset function is just like a function with monotonically increasing value range of 0 to 1.
In the embodiment of the present application, the step S304 in fig. 3 may further include four steps S407 to S410 as follows:
s407, obtaining the reflectivity contrast confidence corresponding to the second dotted line box.
Illustratively, on the basis of the above-described embodiments, the second dashed box B is for any one of the second dashed box sets H ij The second dashed frame B can be obtained according to steps S402 to S406 ij The corresponding set of reflectivity contrast confidence values is specifically represented by Γ= { ρ Ii1 ，ρ Ii2 ，...，ρ Iin And the representation is performed.
S408, acquiring a weighted average length according to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence corresponding to the second dotted line frame.
FIG. 6 is a schematic diagram of a second dashed box of the same group as the first dashed box according to an embodiment of the present application, as shown in FIG. 6, showing a plurality of dashed boxes, wherein the dashed box B i0 I.e. the first dashed box B of the accuracy to be determined i And dashed line box B i1 Dashed line box B i2 Dashed line box B i3 And a dashed line frame B i4 Namely, with the first dotted line frame B i0 And the second dotted line boxes in the same group are separated from the centers of the first dotted line boxes by a distance smaller than or equal to the first distance threshold d. In this step, can rootAccording to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence coefficient corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence coefficient corresponding to the second dotted line frame, the weighted average length L of the same group of dotted line frames can be obtained through the following formula:
wherein, length (B) k ) Indicated by a dotted line frame B i0 To B in The length of the lateral edge of each dashed box; ρ Ik Indicated by a dotted line frame B i0 To B in The confidence of the reflectivity contrast corresponding to each dashed box.
Referring to fig. 6, a weighted average length L is shown.
S409, determining the length deviation of the first dotted line frame as the difference between the length of the transverse edge of the first dotted line frame and the weighted average length.
Illustratively, after the weighted average length is obtained, the length deviation of the first dashed box may be determined as the difference between the length of the lateral edge of the first dashed box and the weighted average length, that is:
ΔL＝|L-length(B i )|
referring to fig. 6, the length deviation Δl of the first dashed box is shown.
S410, acquiring the length consistency confidence coefficient corresponding to the first dotted line frame through a second preset function according to the length deviation and the deviation sensitivity factor.
Wherein the deviation sensitivity factor is used to characterize the degree of sensitivity to deviation.
According to the assumption that the length of the first dotted frame is equal to the length of the print, the smaller the length deviation of the first dotted frame is, the higher the accuracy of the first dotted frame is, so that the length consistency confidence corresponding to the first dotted frame can be calculated by adopting a monotonically decreasing function. Specifically, according to the length deviation and deviation sensitivity factor of the first dashed frame, the length consistency confidence ρ corresponding to the first dashed frame is obtained through the following second preset function l ：
ρ l ＝exp(-ΔL/σ)
Wherein sigma represents a deviation sensitive factor, the larger the value of the sigma is, the less sensitive to deviation is, and when the sigma value is 0.5 through experimental analysis, the better correlation can be obtained.
It should be noted that, optionally, the second preset function is not limited in particular, and the second preset function may be a function similar to a monotonically decreasing value range from 0 to 1.
In the embodiment of the present application, the step S305 in fig. 3 may further include three steps S411 to S413 as follows:
s411, carrying out weighted fusion on the reflectivity contrast confidence coefficient and the length consistency confidence coefficient to obtain a target confidence coefficient.
Illustratively, when a reflectivity contrast confidence ρ corresponding to the first dashed box is obtained I And length consistency confidence ρ l The reflectivity contrast confidence ρ can then be calculated I And length consistency confidence ρ l And carrying out weighted fusion to obtain a target confidence coefficient rho as follows: ρ=ωρ I +(1-ω)ρ l Wherein ω is a weighting coefficient; due to ρ I 、ρ l And ω are both values between 0 and 1, so the final target confidence ρ ε [0,1 ]]The larger the value of the target confidence, the higher the accuracy of the dashed box; as can be obtained through experimental analysis, when the omega value is 0.7, the obtained target confidence coefficient has better correlation with the precision of the dotted line box. Fig. 7 (a) is a schematic diagram of a distribution of confidence values of a dashed box in a scene according to an embodiment of the present application, and fig. 7 (b) is a schematic diagram of a distribution of confidence values of a dashed box in a scene according to another embodiment of the present application. As shown in fig. 7 (a) and 7 (b), target confidence levels respectively corresponding to different dotted boxes are shown.
Optionally, a geometric weighted average manner may be further adopted to perform weighted fusion on the reflectivity contrast confidence coefficient and the length consistency confidence coefficient corresponding to the first dashed line frame, so as to obtain the target confidence coefficient, where the ratio of the reflectivity contrast confidence coefficient and the length consistency confidence coefficient corresponding to the first dashed line frame needs to be controlled.
And S412, if the target confidence coefficient is greater than or equal to the confidence coefficient threshold value, determining the precision of the first dotted line frame to be high.
And S413, if the target confidence coefficient is smaller than the confidence coefficient threshold value, determining that the precision of the first dotted line frame is low.
Illustratively, the confidence threshold may be reasonably set according to the accuracy requirements, such as 0.8. After the target confidence coefficient corresponding to the first dotted line frame is obtained, if the target confidence coefficient is greater than or equal to 0.8, determining the precision of the first dotted line frame to be high; if the target confidence is less than 0.8, the accuracy of the first dashed box is determined to be low.
It is understood that for each of the dashed boxes to be determined in the dashed box set Q, the corresponding precision of each of the dashed boxes may be determined in turn by performing the steps of S401 to S413. After the precision of the dashed frame is determined, the high-precision dashed frame can be inspected without checking, and the low-precision dashed frame can be subjected to fixed-point maintenance, namely, the low-precision dashed frame boundary is corrected, so that the requirement of the high-precision map on the element precision is met.
According to the method for determining the precision of the dotted line frame, which is provided by the embodiment of the application, the dotted line frame comprises a first dotted line frame and a second dotted line frame by acquiring the point cloud data comprising the dotted line frame; according to the point cloud data, acquiring midpoint coordinates corresponding to the midpoints of the two longitudinal sides of the first dotted line frame, acquiring a target point cloud with a distance from the midpoint being smaller than or equal to a third distance threshold value through a preset search algorithm according to the midpoint coordinates, and determining the point cloud reflectivity contrast ratio taking the longitudinal side of the first dotted line frame as a boundary according to the reflectivity intensity value of the target point cloud; determining the target point cloud reflectivity contrast according to the point cloud reflectivity contrast taking each of two longitudinal edges of a first dotted line frame as a boundary, and determining the reflectivity contrast confidence corresponding to the first dotted line frame through a first preset function according to the target point cloud reflectivity contrast and a contrast threshold, namely establishing a first association relation between the precision of the dotted line frame and the surrounding physical environment characteristics; acquiring a weighted average length according to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence corresponding to the second dotted line frame by acquiring the reflectivity contrast confidence corresponding to the second dotted line frame; determining the length deviation of the first dashed frame as the difference between the length of the transverse edge of the first dashed frame and the weighted average length; acquiring the length consistency confidence coefficient corresponding to the first dotted line frame through a second preset function according to the length deviation and the deviation sensitive factor, namely establishing a second association relation between the precision of the dotted line frame and the surrounding physical environment characteristics; carrying out weighted fusion on the reflectivity contrast confidence coefficient and the length consistency confidence coefficient to obtain a target confidence coefficient, and determining the precision of the first dotted line frame to be high if the target confidence coefficient is greater than or equal to a confidence coefficient threshold value; if the target confidence level is less than the confidence level threshold, determining that the accuracy of the first dashed box is low. According to the embodiment of the application, the accuracy of the dotted line frame is mapped into the confidence value according to the corresponding reflectivity contrast confidence and length consistency confidence of the first dotted line frame, so that the quick quality inspection and fixed-point maintenance of the dotted line frame are realized based on the confidence, the production efficiency of the high-accuracy map can be improved, and the competitiveness of high-accuracy map products can be improved.
The following are examples of the apparatus of the present application that may be used to perform the method embodiments of the present application. For details not disclosed in the embodiments of the apparatus of the present application, please refer to the embodiments of the method of the present application.
Fig. 8 is a schematic structural diagram of a determination device for precision of a dashed box according to an embodiment of the present application, as shown in fig. 8, a determination device 800 for precision of a dashed box according to an embodiment of the present application includes: a first acquisition module 801, a first determination module 802, a second determination module 803, a second acquisition module 804, and a third determination module 805. Wherein:
the first obtaining module 801 is configured to obtain point cloud data including a dashed frame, where the dashed frame includes a first dashed frame and a second dashed frame, a midpoint of two longitudinal sides of the first dashed frame forms a target line segment, a center distance between the second dashed frame and the first dashed frame is less than or equal to a first distance threshold, and a projection distance between a center of the second dashed frame and the target line segment is less than or equal to a second distance threshold.
A first determining module 802 is configured to determine, according to the point cloud data, a point cloud reflectivity contrast with a longitudinal edge of the first dashed box as a boundary.
The second determining module 803 is configured to determine, according to the point cloud reflectivity contrast, a reflectivity contrast confidence level corresponding to the first dashed box.
The second obtaining module 804 is configured to obtain a length consistency confidence corresponding to the first dashed frame based on a length of a lateral edge of the first dashed frame and a length of a lateral edge of the second dashed frame, where the length of the lateral edge of the first dashed frame is greater than the length of the longitudinal edge of the first dashed frame.
A third determining module 805 is configured to determine the accuracy of the first dashed box according to the reflectivity contrast confidence level and the length consistency confidence level.
In some embodiments, the first determining module 802 may be specifically configured to: acquiring midpoint coordinates corresponding to the midpoint according to the point cloud data; according to the midpoint coordinates, acquiring a target point cloud with a distance from the midpoint being smaller than or equal to a third distance threshold value through a preset search algorithm, wherein the third distance threshold value is larger than the length of the longitudinal edge of the first dotted line frame; and determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud.
Optionally, the first determining module 802 may be specifically configured to, when configured to determine the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud: determining a first area and a second area which are used for taking two transverse sides of a first dotted line frame as boundaries and taking the longitudinal sides as boundaries and contain target point clouds for each of the two longitudinal sides, wherein the first area is a rectangular area contained in the first dotted line frame, and the second area is a rectangular area which is not contained in the first dotted line frame; acquiring a first average reflectivity intensity value corresponding to the first region according to the reflectivity intensity value of the target point cloud contained in the first region; acquiring a second average reflectivity intensity value corresponding to the second region according to the reflectivity intensity value of the target point cloud contained in the second region; the point cloud reflectance contrast is determined as the difference between the first average reflectance intensity value and the second average reflectance intensity value.
In some embodiments, the second determining module 803 may be specifically configured to: determining a target point cloud reflectivity contrast according to the point cloud reflectivity contrast taking each of two longitudinal edges of the first dotted line frame as a boundary; and determining the reflectivity contrast confidence coefficient through a first preset function according to the reflectivity contrast of the cloud of the target point and the contrast threshold.
In some embodiments, the second obtaining module 804 may be specifically configured to: acquiring a reflectivity contrast confidence coefficient corresponding to the second dotted line frame; acquiring a weighted average length according to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence corresponding to the second dotted line frame; determining the length deviation of the first dashed frame as the difference between the length of the transverse edge of the first dashed frame and the weighted average length; and acquiring the confidence coefficient of the length consistency through a second preset function according to the length deviation and a deviation sensitivity factor, wherein the deviation sensitivity factor is used for representing the sensitivity degree to the deviation.
In some embodiments, the third determining module 805 may be specifically configured to: weighting and fusing the reflectivity contrast confidence coefficient and the length consistency confidence coefficient to obtain a target confidence coefficient; if the target confidence coefficient is greater than or equal to the confidence coefficient threshold value, determining the precision of the first dotted line frame to be high; if the target confidence level is less than the confidence level threshold, determining that the accuracy of the first dashed box is low.
The device of the present embodiment may be used to execute the technical solution of any of the above-described method embodiments, and its implementation principle and technical effects are similar, and are not described herein again.
Fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device may be provided as a server or computer, for example. Referring to fig. 9, the electronic device 900 includes a processing component 901 that further includes one or more processors, and memory resources represented by memory 902 for storing instructions, such as applications, executable by the processing component 901. The application program stored in the memory 902 may include one or more modules each corresponding to a set of instructions. Further, the processing component 901 is configured to execute instructions to perform any of the method embodiments described above.
The electronic device 900 may also include a power component 903 configured to perform power management of the electronic device 900, a wired or wireless network interface 904 configured to connect the electronic device 900 to a network, and an input output (I/O) interface 905. The electronic device 900 may operate based on an operating system stored in memory 902, such as Windows Server, mac OS XTM, unixTM, linuxTM, freeBSDTM, or the like.
The application also provides a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and when the processor executes the computer execution instructions, the scheme of the method for determining the precision of the above dotted line frame is realized.
The application also provides a computer program product comprising a computer program which, when executed by a processor, implements the aspects of the method of determining the accuracy of a dashed box as above.
The computer readable storage medium described above may be implemented by any type of volatile or non-volatile memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk. A readable storage medium can be any available medium that can be accessed by a general purpose or special purpose computer.
An exemplary readable storage medium is coupled to the processor such the processor can read information from, and write information to, the readable storage medium. In the alternative, the readable storage medium may be integral to the processor. The processor and the readable storage medium may reside in an application specific integrated circuit (Application Specific Integrated Circuits, ASIC). It is also possible that the processor and the readable storage medium are present as discrete components in the determination means of the accuracy of the dashed box.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the method embodiments described above may be performed by hardware associated with program instructions. The foregoing program may be stored in a computer readable storage medium. The program, when executed, performs steps including the method embodiments described above; and the aforementioned storage medium includes: various media that can store program code, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the application.

Claims (6)

1. A method for determining the accuracy of a dashed box, comprising:
acquiring point cloud data comprising a dotted line frame, wherein the dotted line frame comprises a first dotted line frame and a second dotted line frame, the midpoints of two longitudinal sides of the first dotted line frame form a target line segment, the center distance between the second dotted line frame and the first dotted line frame is smaller than or equal to a first distance threshold, and the projection distance from the center of the second dotted line frame to the target line segment is smaller than or equal to a second distance threshold;
According to the point cloud data, determining the point cloud reflectivity contrast ratio taking the longitudinal edge of the first dotted line frame as a boundary;
determining the reflectivity contrast confidence corresponding to the first dotted line frame according to the point cloud reflectivity contrast;
acquiring a length consistency confidence corresponding to the first dotted frame based on the length of the transverse edge of the first dotted frame and the length of the transverse edge of the second dotted frame, wherein the length of the transverse edge of the first dotted frame is greater than the length of the longitudinal edge of the first dotted frame;
determining the accuracy of the first dashed box according to the reflectivity contrast confidence and the length consistency confidence;
the determining, according to the point cloud data, a point cloud reflectivity contrast ratio using a longitudinal edge of the first dashed line frame as a boundary line includes:
acquiring midpoint coordinates corresponding to the midpoints according to the point cloud data;
according to the midpoint coordinates, acquiring a target point cloud with a distance from the midpoint being smaller than or equal to a third distance threshold value through a preset search algorithm, wherein the third distance threshold value is larger than the length of the longitudinal edge of the first dotted frame;
determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud;
The determining the reflectivity contrast confidence corresponding to the first dotted line frame according to the point cloud reflectivity contrast includes:
determining a target point cloud reflectivity contrast according to the point cloud reflectivity contrast taking each of two longitudinal edges of the first dotted line frame as a boundary;
determining the reflectivity contrast confidence coefficient through a first preset function according to the target point cloud reflectivity contrast and a contrast threshold;
based on the length of the lateral edge of the first dashed frame and the length of the lateral edge of the second dashed frame, obtaining the length consistency confidence corresponding to the first dashed frame includes:
acquiring the reflectivity contrast confidence corresponding to the second dotted line frame;
acquiring a weighted average length according to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence corresponding to the second dotted line frame;
determining a length deviation of the first dashed box as a difference between the length of the lateral edge of the first dashed box and the weighted average length;
and acquiring the confidence coefficient of the length consistency through a second preset function according to the length deviation and the deviation sensitivity factor, wherein the deviation sensitivity factor is used for representing the sensitivity degree to deviation.
2. The method for determining the accuracy of the dashed box according to claim 1, wherein said determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud comprises:
determining, for each of the two longitudinal sides, a first region and a second region including a target point cloud, the first region being a rectangular region included in the first dashed frame, the second region being a rectangular region not included in the first dashed frame, the first region being bounded by the two lateral sides of the first dashed frame and the longitudinal sides being a boundary line;
acquiring a first average reflectivity intensity value corresponding to the first region according to the reflectivity intensity value of the target point cloud contained in the first region;
acquiring a second average reflectivity intensity value corresponding to the second region according to the reflectivity intensity value of the target point cloud contained in the second region;
determining the point cloud reflectance contrast as a difference between the first average reflectance intensity value and the second average reflectance intensity value.
3. The method of determining the precision of a dashed box according to any of claims 1 to 2, characterized in that the determining the precision of the first dashed box from the reflectivity-contrast confidence and the length-consistency confidence comprises:
Weighting and fusing the reflectivity contrast confidence coefficient and the length consistency confidence coefficient to obtain a target confidence coefficient;
if the target confidence coefficient is greater than or equal to a confidence coefficient threshold value, determining that the precision of the first dotted frame is high;
and if the target confidence coefficient is smaller than the confidence coefficient threshold value, determining that the precision of the first dotted line frame is low.
4. A determination apparatus for accuracy of a dashed box, comprising:
the first acquisition module is used for acquiring point cloud data comprising a dotted line frame, the dotted line frame comprises a first dotted line frame and a second dotted line frame, the middle points of two longitudinal sides of the first dotted line frame form a target line segment, the center distance between the second dotted line frame and the first dotted line frame is smaller than or equal to a first distance threshold, and the projection distance from the center of the second dotted line frame to the target line segment is smaller than or equal to a second distance threshold;
the first determining module is used for determining the point cloud reflectivity contrast ratio taking the longitudinal edge of the first dotted line frame as a boundary according to the point cloud data;
the second determining module is used for determining the reflectivity contrast confidence corresponding to the first dotted line frame according to the reflectivity contrast of the point cloud;
The second acquisition module is used for acquiring the length consistency confidence corresponding to the first dotted line frame based on the length of the transverse edge of the first dotted line frame and the length of the transverse edge of the second dotted line frame, and the length of the transverse edge of the first dotted line frame is larger than the length of the longitudinal edge of the first dotted line frame;
a third determining module, configured to determine an accuracy of the first dashed frame according to the reflectivity contrast confidence level and the length consistency confidence level;
the first determining module is specifically configured to obtain a midpoint coordinate corresponding to the midpoint according to the point cloud data; according to the midpoint coordinates, acquiring a target point cloud with a distance from the midpoint being smaller than or equal to a third distance threshold value through a preset search algorithm, wherein the third distance threshold value is larger than the length of the longitudinal edge of the first dotted frame; determining the point cloud reflectivity contrast according to the reflectivity intensity value of the target point cloud;
the second determining module is specifically configured to determine a target point cloud reflectivity contrast according to a point cloud reflectivity contrast taking each of two longitudinal sides of the first dashed frame as a boundary; determining the reflectivity contrast confidence coefficient through a first preset function according to the target point cloud reflectivity contrast and a contrast threshold;
The second obtaining module is specifically configured to obtain a reflectivity contrast confidence corresponding to the second dashed box; acquiring a weighted average length according to the length of the transverse edge of the first dotted line frame, the reflectivity contrast confidence corresponding to the first dotted line frame, the length of the transverse edge of the second dotted line frame and the reflectivity contrast confidence corresponding to the second dotted line frame; determining a length deviation of the first dashed box as a difference between the length of the lateral edge of the first dashed box and the weighted average length; and acquiring the confidence coefficient of the length consistency through a second preset function according to the length deviation and the deviation sensitivity factor, wherein the deviation sensitivity factor is used for representing the sensitivity degree to deviation.
5. An electronic device, comprising: a processor, and a memory communicatively coupled to the processor;
the memory stores computer-executable instructions;
the processor executes the computer-executable instructions stored in the memory to implement the method of determining the precision of a dashed box as claimed in any one of claims 1 to 3.
6. A computer-readable storage medium, in which computer program instructions are stored which, when executed by a processor, implement the method of determining the accuracy of a dashed box as claimed in any of claims 1 to 3.
CN202210542312.XA 2022-05-18 2022-05-18 Determination method, device, equipment and storage medium for precision of dotted line frame Active CN114862808B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210542312.XA CN114862808B (en) 2022-05-18 2022-05-18 Determination method, device, equipment and storage medium for precision of dotted line frame

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210542312.XA CN114862808B (en) 2022-05-18 2022-05-18 Determination method, device, equipment and storage medium for precision of dotted line frame

Publications (2)

Publication Number Publication Date
CN114862808A CN114862808A (en) 2022-08-05
CN114862808B true CN114862808B (en) 2023-11-17

Family

ID=82639924

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210542312.XA Active CN114862808B (en) 2022-05-18 2022-05-18 Determination method, device, equipment and storage medium for precision of dotted line frame

Country Status (1)

Country Link
CN (1) CN114862808B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3620817A1 (en) * 2018-09-07 2020-03-11 Baidu Online Network Technology (Beijing) Co., Ltd. Method and apparatus for generating object detection box, device, storage medium, and vehicle
CN111487641A (en) * 2020-03-19 2020-08-04 福瑞泰克智能系统有限公司 Method and device for detecting object by using laser radar, electronic equipment and storage medium
CN113607185A (en) * 2021-10-08 2021-11-05 禾多科技（北京）有限公司 Lane line information display method, lane line information display device, electronic device, and computer-readable medium
CN113945937A (en) * 2021-09-16 2022-01-18 阿里巴巴达摩院(杭州)科技有限公司 Precision detection method, device and storage medium
CN114076936A (en) * 2020-08-20 2022-02-22 北京万集科技股份有限公司 Precision evaluation method and device of combined calibration parameters, server and computer readable storage medium
WO2022083178A1 (en) * 2020-10-21 2022-04-28 湖北亿咖通科技有限公司 Laser point cloud processing method, electronic apparatus and storage medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3620817A1 (en) * 2018-09-07 2020-03-11 Baidu Online Network Technology (Beijing) Co., Ltd. Method and apparatus for generating object detection box, device, storage medium, and vehicle
CN111487641A (en) * 2020-03-19 2020-08-04 福瑞泰克智能系统有限公司 Method and device for detecting object by using laser radar, electronic equipment and storage medium
CN114076936A (en) * 2020-08-20 2022-02-22 北京万集科技股份有限公司 Precision evaluation method and device of combined calibration parameters, server and computer readable storage medium
WO2022083178A1 (en) * 2020-10-21 2022-04-28 湖北亿咖通科技有限公司 Laser point cloud processing method, electronic apparatus and storage medium
CN113945937A (en) * 2021-09-16 2022-01-18 阿里巴巴达摩院(杭州)科技有限公司 Precision detection method, device and storage medium
CN113607185A (en) * 2021-10-08 2021-11-05 禾多科技（北京）有限公司 Lane line information display method, lane line information display device, electronic device, and computer-readable medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
多源点云数据融合的建筑物精细化建模;王树臻;郑国强;王光生;胡玉民;张德怀;亓伟;;测绘通报(08);全文 *

Also Published As

Publication number Publication date
CN114862808A (en) 2022-08-05

Similar Documents

Publication Publication Date Title
US20210287037A1 (en) Object detection method and apparatus, electronic device, and storage medium
US20130050476A1 (en) Structured-Light Based Measuring Method and System
CN111080662A (en) Lane line extraction method and device and computer equipment
EP3621041B1 (en) Three-dimensional representation generating system
US11466992B2 (en) Method, apparatus, device and medium for detecting environmental change
CN113804100B (en) Method, device, equipment and storage medium for determining space coordinates of target object
US20230206500A1 (en) Method and apparatus for calibrating extrinsic parameter of a camera
CN113705669A (en) Data matching method and device, electronic equipment and storage medium
US20170161946A1 (en) Stochastic map generation and bayesian update based on stereo vision
CN114863380A (en) Lane line identification method and device and electronic equipment
CN113074634B (en) Rapid phase matching method, storage medium and three-dimensional measurement system
WO2022099620A1 (en) Three-dimensional point cloud segmentation method and apparatus, and mobile platform
CN114862808B (en) Determination method, device, equipment and storage medium for precision of dotted line frame
CN111813882B (en) Robot map construction method, device and storage medium
CN110673114B (en) Method and device for calibrating depth of three-dimensional camera, computer device and storage medium
CN113808142B (en) Ground identification recognition method and device and electronic equipment
CN113822892B (en) Evaluation method, device and equipment of simulated radar and computer storage medium
CN114611635A (en) Object identification method and device, storage medium and electronic device
CN115797310A (en) Method for determining inclination angle of photovoltaic power station group string and electronic equipment
CN115471574A (en) External parameter determination method and device, storage medium and electronic device
CN114782496A (en) Object tracking method and device, storage medium and electronic device
CN114595238A (en) Vector-based map processing method and device
CN113360593A (en) Sensor data processing method and device
CN114897968B (en) Method and device for determining vehicle vision, computer equipment and storage medium
CN117541895A (en) Pavement arrow precision evaluation method, device, equipment and storage medium

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
GR01 Patent grant
GR01 Patent grant