CN110211219A - A kind of processing method of mass cloud data - Google Patents
A kind of processing method of mass cloud data Download PDFInfo
- Publication number
- CN110211219A CN110211219A CN201910312779.3A CN201910312779A CN110211219A CN 110211219 A CN110211219 A CN 110211219A CN 201910312779 A CN201910312779 A CN 201910312779A CN 110211219 A CN110211219 A CN 110211219A
- Authority
- CN
- China
- Prior art keywords
- leaf node
- octree
- equilibrium
- point
- cloud data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
-
- 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
- G06T17/005—Tree description, e.g. octree, quadtree
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Graphics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Image Generation (AREA)
Abstract
The invention discloses a kind of processing methods of mass cloud data, storage, rendering and capture including point cloud data, wherein storage method stores point cloud data using non-equilibrium Octree, point set in the same leaf node in non-equilibrium Octree is continuously stored in same file, to improve the speed of lookup；And the point cloud data of storage is rendered and captured on the basis of the storage of non-equilibrium Octree, acquisition speed is improved, to improve the efficiency that a cloud shows and captures.
Description
Technical field
The present invention relates to Point Cloud Processing field more particularly to a kind of storages of mass cloud data, rendering and capture
Method.
Background technique
Currently, with the arrival of big data, the continuous upgrading of hardware device obtains product appearance by measuring instrument at present
The point data set on surface is referred to as point cloud, since the point that each scanning obtains includes three-dimensional coordinate, it is also possible to contain
Colouring information or Reflection intensity information, therefore point cloud data is more and more huger, application industry is also more and more extensive therewith, therefore point
The efficiency that cloud shows and captures must also be promoted therewith.
And existing cloud storage format, such as las format, although supporting the storage of mass data, for quick
Rendering, which must additionally establish spatial index, could improve display efficiency, but additionally establishes this method of spatial index and acquire
Point be it is discrete, lead to the speed for reducing acquisition, cause display efficiency and capturing efficiency that cannot improve；And it establishes empty
Between the algorithm that the indexes mode of generally taking Octree or KD-Tree or many algorithms to combine carry out, establish spatial index
Efficiency of algorithm render and capture efficiency be unbalanced, therefore for Fast rendering and capture none well solve
Certainly scheme.
Summary of the invention
For overcome the deficiencies in the prior art, the purpose of the present invention is to provide a kind of data processings of mass cloud data
Method improves the speed of data acquisition using non-equilibrium Octree storage mode, to improve the efficiency that a cloud shows and captures.
The purpose of the present invention adopts the following technical scheme that realization:
A kind of processing method of mass cloud data, comprising:
Step S1: point cloud data is stored using non-equilibrium Octree, by the same leaf node in non-equilibrium Octree
In point set be continuously stored in same file；
Step S2: the maximum rendering points of setting know spy on the basis of step S1 non-equilibrium Octree storage method
The number for determining the point in leaf node vacuates point data in proportion and carries out color applying drawing to it；Wherein particular leaf node is
The leaf node of intersection obtained after carrying out intersection judgement into non-equilibrium Octree；
Step S3: setting captures the precision of ray or bounding box, in the basis of the non-equilibrium Octree storage method of step S1
On closest approach judgement is carried out to the point group of particular leaf node one by one, capture and return to first point for meeting required precision；Its
Middle particular leaf node is the leaf node that intersection obtained after intersection judgement is carried out into non-equilibrium Octree.
Further, the storage method of the step S1 includes:
Step S1.1: will be in the new non-equilibrium Octree of single 3 D coordinate points Incremental insertion；
Step S1.2: recurrence enters non-equilibrium Octree, is found according to the leaf node spatial dimension of non-equilibrium Octree
The leaf node that each insertion coordinate points are belonged in step S1.1；
Step S1.3: the leaf node quantity found in judgment step S1.2 whether be more than non-equilibrium Octree maximum leaf
Child node number, if not exceeded, thening follow the steps S1.4；If being more than, S1.5 is thened follow the steps；
Step S1.4: the coordinate points of all insertions are respectively written into the temporary file or memory of corresponding leaf node
It is stored, executes step S1.6；
Step S1.5: division leaf node simultaneously increases new leaf node newly down, and new coordinate points continue from present node
It transmits and is inserted into new leaf node down；Again it is started the cycle over from step S1.2；
Step S1.6: the point set that the same leaf node in non-equilibrium Octree is stored is continuously stored in same
In file.
Further, point data new in the step S1.1 is accompanied with attributive character, and the attributive character includes but not
It is limited to color, intensity.
Further, the storage mode of the step S1.6 is to judge whether the stored memory of point set to be stored exceeds
Point set is stored entirely in next file by the free memory capacity of current file if being more than；If not exceeded, then will
The point set is stored in current file, and the node inside leaf is kept continuously to store together.
Further, the rendering method of the step S2 includes:
Step S2.1: according to different hardware equipment performance, maximum rendering points are set；
Step S2.2: intersection judgement is carried out into non-equilibrium Octree, obtains the leaf node of intersection；
Step S2.3: the leaf node midpoint group of intersection is returned to according to the Coutinuous store rule of the non-equilibrium Octree of step S1
First address, therefrom obtain current leaf node point number；
Step S2.4: the number of point is vacuated into points in conjunction with the maximum rendering points being arranged according to step S2.1 in proportion
According to, and it is drawn.
Further, attribute filtering is carried out to point data after point data is vacuated in the step S2.4.
Further, the attribute filtering includes but is not limited to elevation filtering, Intensity attribute filtering, color attribute filtering.
Further, the method for catching in the step S3 includes:
Step S3.1: precision is arranged in the incoming ray or bounding box captured；
Step S3.2: intersection operation is carried out with the non-equilibrium Octree of memory mapping, obtains the leaf node of intersection；
Step S3.3: the point group of leaf node is carried out one by one according to the Coutinuous store rule of the non-equilibrium Octree of step S1
Closest approach judgement, until capturing first meets the closest approach of required precision, and is returned.
Compared with prior art, the beneficial effects of the present invention are:
Point cloud data is handled using non-equilibrium Octree algorithm and memory mapping mode, point cloud data is continuously deposited
Storage reduces the space of leaf node number and storage, improves search speed；And to point on the basis of the storage of non-equilibrium Octree
Cloud data are rendered and are captured, and can further promote the speed of data acquisition, to improve the efficiency that a cloud shows and captures.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of the storage method of mass cloud data of the present invention；
Fig. 2 is a kind of flow diagram of the rendering method of mass cloud data of the present invention；
Fig. 3 is a kind of flow diagram of the method for catching of mass cloud data of the present invention.
Specific embodiment
In the following, being described further in conjunction with attached drawing and specific embodiment to the present invention, it should be noted that not
Under the premise of conflicting, new implementation can be formed between various embodiments described below or between each technical characteristic in any combination
Example.
Embodiment one
A kind of storage method of mass cloud data, as shown in Figure 1, comprising:
Step S1.1: will be in the new non-equilibrium Octree of single 3 D coordinate points Incremental insertion；Wherein new three-dimensional coordinate
Point further includes colouring information, strength information, wherein colouring information is to obtain colour by camera other than with geometric position
Then image assigns the colouring information of the pixel of corresponding position to corresponding point in point cloud, and the acquisition of strength information is laser
Scanner, which receives, decorates collected echo strength, the Facing material of strength information and target, roughness, incident angular direction and
The generation energy of instrument, optical maser wavelength are related.
Step S1.2: recurrence enters non-equilibrium Octree, is found according to the leaf node spatial dimension of non-equilibrium Octree
The leaf node that each insertion coordinate points are belonged in step S1.1；
The non-equilibrium Octree is similar and Octree, leaf node number can be zero to eight any one, it is non-equilibrium
Octree reduces the space of leaf node number and storage, and fast and easy is searched, and can significantly improve search speed.
Step S1.3: the leaf node quantity found in judgment step S1.2 whether be more than non-equilibrium Octree maximum leaf
Child node number, if not exceeded, thening follow the steps S1.4；If being more than, S1.5 is thened follow the steps；
Step S1.4: the coordinate points of all insertions are respectively written into the temporary file or memory of corresponding leaf node
It is stored, memory and speed can be balanced, execute step S1.6 thereafter；
Step S1.5: division leaf node simultaneously increases new leaf node newly down, and new coordinate points continue from present node
It transmits and is inserted into new leaf node down；Again it is started the cycle over from step S1.2；
Step S1.6: the point set that the same leaf node in non-equilibrium Octree is stored is continuously stored in same
In file.Wherein storage mode is to judge whether the stored memory of point set to be stored exceeds the free memory appearance of current file
Amount, if being more than, point set is stored entirely in next file；It ought be above if not exceeded, being then stored in the point set
In part.Purpose is that the point set for being stored the same leaf node is put together, facilitates and is written in same file below；
The point set Coutinuous store of leaf node can be convenient for the quick reading of data by the way of the present embodiment, to improve subsequent
Rendering and the efficiency captured.
Embodiment two
A kind of rendering method of mass cloud data is to store point cloud data using non-equilibrium Octree in embodiment one
Rendering processing is carried out in structure basis, as shown in Fig. 2, the rendering method includes:
Step S2.1: according to different hardware equipment performance, maximum rendering points X is set；Due to the property of different hardware equipment
Can be different, therefore different values is set, while reaching fast browsing, non-Caton, the more points of display as far as possible are to reach rendering
Best effect out, to improve rendering efficiency.
Step S2.2: intersection judgement is carried out into non-equilibrium Octree, obtains the leaf node of intersection；
Step S2.3: the leaf node midpoint group of intersection is returned to according to the Coutinuous store rule of the non-equilibrium Octree of step S1
First address, therefrom obtain current leaf node point number；
Since the rendering of the present embodiment is on condition that using the Coutinuous store side using non-equilibrium Octree in embodiment one
Formula carries out storage processing to data, and the number of the point of leaf node storage and the head of point group can be found according to the rule of Coutinuous store
The first address of address, the point group is the position of first point of leaf node, can be from first point according to the rule of Coutinuous store
Position extend to all the points of current leaf node, to know all point sets of the leaf node, and then obtain the leaf
The number Y of all the points of node.
Step S2.4: the number of point is vacuated into points in conjunction with the maximum rendering points being arranged according to step S2.1 in proportion
According to, and it is drawn.
The formula for vacuating ratio is the points X=that the number Y/ reagent rendering of all the points of leaf node needs to vacuate
Z can extract X point according to the spacing of Z according to the ratio of vacuating and be rendered, to improve the performance and display effect of display.
It is vacuated using a cloud, precentagewise sampling simplifies mass cloud data, so that data processing amount becomes smaller, to accelerate
Data processing time.
It is vacuated in the step S2.4 and attribute filtering is carried out to point data after point data, the attribute filtering includes but not
It is limited to elevation filtering, Intensity attribute filtering, color attribute filtering, to improve display effect.
Embodiment three
A kind of method for catching of mass cloud data is to store point cloud data using non-equilibrium Octree in embodiment one
Capture processing is carried out in structure basis, can further improve capture velocity on the basis of quickly reading data, as shown in figure 3,
The method for catching includes:
Step S3.1: precision is arranged in the incoming ray or bounding box captured；
Step S3.2: intersection operation is carried out with the non-equilibrium Octree of memory mapping, obtains the leaf node of intersection；
Step S3.3: the point group of leaf node is carried out one by one according to the Coutinuous store rule of the non-equilibrium Octree of step S1
Closest approach judgement, until capturing first meets the closest approach of required precision, and is returned.
Closest approach judgement is judged the closest approach of single leaf node, if judging, closest approach is unsatisfactory for precision and wants
It asks, then the closest approach continued in next leaf node calculates, as long as encountering the node for meeting required precision, generation
Table captures, that is, can return to first node for meeting required precision, completes capture operation.
The above embodiment is only the preferred embodiment of the present invention, and the scope of protection of the present invention is not limited thereto,
The variation and replacement for any unsubstantiality that those skilled in the art is done on the basis of the present invention belong to institute of the present invention
Claimed range.
Claims (8)
1. a kind of processing method of mass cloud data characterized by comprising
Step S1: storing point cloud data using non-equilibrium Octree, will be in the same leaf node in non-equilibrium Octree
Point set is continuously stored in same file；
Step S2: the maximum rendering points of setting know particular leaf on the basis of step S1 non-equilibrium Octree storage method
The number of point in child node vacuates point data in proportion and carries out color applying drawing to it；Wherein particular leaf node is to enter
Non-equilibrium Octree carries out the leaf node of intersection obtained after intersection judgement；
Step S3: setting captures the precision of ray or bounding box, on the basis of step S1 non-equilibrium Octree storage method by
A point group to particular leaf node carries out closest approach judgement, captures and returns to first point for meeting required precision；It is wherein special
Determining leaf node is the leaf node that intersection obtained after intersection judgement is carried out into non-equilibrium Octree.
2. the processing method of mass cloud data according to claim 1, which is characterized in that the storage side of the step S1
Method includes:
Step S1.1: will be in the new non-equilibrium Octree of single 3 D coordinate points Incremental insertion；
Step S1.2: recurrence enters non-equilibrium Octree, finds step according to the leaf node spatial dimension of non-equilibrium Octree
The leaf node that each insertion coordinate points are belonged in S1.1；
Step S1.3: the leaf node quantity found in judgment step S1.2 whether be more than non-equilibrium Octree maximum leaf section
Point number, if not exceeded, thening follow the steps S1.4；If being more than, S1.5 is thened follow the steps；
Step S1.4: the coordinate points of all insertions are respectively written into the temporary file or memory of corresponding leaf node and are carried out
Storage executes step S1.6；
Step S1.5: division leaf node simultaneously increases new leaf node newly down, and new coordinate points continue down from present node
It transmits and is inserted into new leaf node；Again it is started the cycle over from step S1.2；
Step S1.6: the point set that the same leaf node in non-equilibrium Octree is stored is continuously stored in same file
In.
3. the processing method of mass cloud data according to claim 2, which is characterized in that in the step S1.1 newly
Point data is accompanied with attributive character, and the attributive character includes but is not limited to color, intensity.
4. the processing method of mass cloud data according to claim 2, which is characterized in that the storage of the step S1.6
Mode is to judge whether the stored memory of point set to be stored exceeds the free memory capacity of current file, will if being more than
Point set is stored entirely in next file；If not exceeded, then the point set is stored in current file.
5. the processing method of mass cloud data according to claim 1, which is characterized in that the rendering side of the step S2
Method includes:
Step S2.1: according to different hardware equipment performance, maximum rendering points are set；
Step S2.2: intersection judgement is carried out into non-equilibrium Octree, obtains the leaf node of intersection；
Step S2.3: the head of the leaf node midpoint group of intersection is returned to according to the Coutinuous store rule of the non-equilibrium Octree of step S1
Address therefrom obtains the number of the point of current leaf node；
Step S2.4: vacuating point data in conjunction with the maximum rendering points being arranged according to step S2.1 for the number of point in proportion,
And it is drawn.
6. the processing method of mass cloud data according to claim 5, which is characterized in that vacuated in the step S2.4
Attribute filtering is carried out to point data after point data.
7. the processing method of mass cloud data according to claim 6, which is characterized in that attribute filtering include but
It is not limited to elevation filtering, Intensity attribute filtering, color attribute filtering.
8. the processing method of mass cloud data according to claim 1, which is characterized in that the capture in the step S3
Method includes:
Step S3.1: precision is arranged in the incoming ray or bounding box captured；
Step S3.2: intersection operation is carried out with the non-equilibrium Octree of memory mapping, obtains the leaf node of intersection；
Step S3.3: the point group of leaf node is carried out one by one according to the Coutinuous store rule of the non-equilibrium Octree of step S1 nearest
Point judgement, until capturing first closest approach for meeting required precision, and is returned.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910312779.3A CN110211219A (en) | 2019-04-18 | 2019-04-18 | A kind of processing method of mass cloud data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910312779.3A CN110211219A (en) | 2019-04-18 | 2019-04-18 | A kind of processing method of mass cloud data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110211219A true CN110211219A (en) | 2019-09-06 |
Family
ID=67785485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910312779.3A Pending CN110211219A (en) | 2019-04-18 | 2019-04-18 | A kind of processing method of mass cloud data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110211219A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115063496A (en) * | 2022-08-19 | 2022-09-16 | 北京山维科技股份有限公司 | Method and device for rapidly processing point cloud data |
| CN115063497A (en) * | 2022-08-19 | 2022-09-16 | 北京山维科技股份有限公司 | Point cloud data processing method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281654A (en) * | 2008-05-20 | 2008-10-08 | 上海大学 | Method for processing cosmically complex three-dimensional scene based on eight-fork tree |
| CN107798721A (en) * | 2016-08-31 | 2018-03-13 | 北京四维图新科技股份有限公司 | A kind of processing method and processing device of cloud data and point cloud rendering intent and device |
| US20190018680A1 (en) * | 2017-07-12 | 2019-01-17 | Topcon Positioning Systems, Inc. | Point cloud data method and apparatus |
| EP3467778A1 (en) * | 2017-10-06 | 2019-04-10 | Thomson Licensing | A method and apparatus for encoding/decoding the geometry of a point cloud representing a 3d object |
-
2019
- 2019-04-18 CN CN201910312779.3A patent/CN110211219A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101281654A (en) * | 2008-05-20 | 2008-10-08 | 上海大学 | Method for processing cosmically complex three-dimensional scene based on eight-fork tree |
| CN107798721A (en) * | 2016-08-31 | 2018-03-13 | 北京四维图新科技股份有限公司 | A kind of processing method and processing device of cloud data and point cloud rendering intent and device |
| US20190018680A1 (en) * | 2017-07-12 | 2019-01-17 | Topcon Positioning Systems, Inc. | Point cloud data method and apparatus |
| EP3467778A1 (en) * | 2017-10-06 | 2019-04-10 | Thomson Licensing | A method and apparatus for encoding/decoding the geometry of a point cloud representing a 3d object |
Non-Patent Citations (1)
| Title |
|---|
| 徐鹏: ""海量三维点云数据的组织与可视化研究"", 《中国优秀硕士学位论文全文数据库基础科学辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115063496A (en) * | 2022-08-19 | 2022-09-16 | 北京山维科技股份有限公司 | Method and device for rapidly processing point cloud data |
| CN115063497A (en) * | 2022-08-19 | 2022-09-16 | 北京山维科技股份有限公司 | Point cloud data processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109029253B (en) | Package volume measuring method and system, storage medium and mobile terminal | |
| EP3570253B1 (en) | Method and device for reconstructing three-dimensional point cloud | |
| CN108804714A (en) | Point cloud data storage method and device | |
| US7421125B1 (en) | Image analysis, editing and search techniques | |
| CN107247926B (en) | A kind of human body detecting method and device | |
| Richter et al. | Out-of-core real-time visualization of massive 3D point clouds | |
| EP0299769A2 (en) | Method for generating images using ray-tracing | |
| WO2023024482A1 (en) | Interior structured reconstruction method and apparatus, and computer-readable storage medium | |
| CN104408055B (en) | Storage method and device for laser radar point cloud data | |
| CN110211219A (en) | A kind of processing method of mass cloud data | |
| CN205352391U (en) | Entity measurement system based on ground three -dimensional laser scanning | |
| CN109146991A (en) | Picture format conversion method, device, equipment and storage medium | |
| CN108340405A (en) | A kind of robot three-dimensional scanning system and method | |
| CN110648359A (en) | Fruit target positioning and identifying method and system | |
| CN114821571A (en) | Point cloud processing method for power cable identification and reconstruction | |
| CN115063497B (en) | Point cloud data processing method and device | |
| Klein et al. | View planning for the 3d modelling of real world scenes | |
| CN110472085A (en) | 3-D image searching method, system, computer equipment and storage medium | |
| EP4202858A1 (en) | Method and device for cargo counting, computer equipment, and storage medium | |
| Sun et al. | Triangular mesh construction based on point cloud matrix and edge feature extraction | |
| CN113592976A (en) | Map data processing method and device, household appliance and readable storage medium | |
| CN111429570B (en) | Method and system for realizing modeling function based on 3D camera scanning | |
| CN107147624B (en) | Panoramic picture processing method, display device and playback equipment | |
| Stipes et al. | 4D scan registration with the SR-3000 LIDAR | |
| Park et al. | Voxel-based volume modelling of individual trees using terrestrial laser scanners |
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 | ||
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20231013 |