CN106874409A - The storage method and device of cloud data - Google Patents
The storage method and device of cloud data Download PDFInfo
- Publication number
- CN106874409A CN106874409A CN201710044180.7A CN201710044180A CN106874409A CN 106874409 A CN106874409 A CN 106874409A CN 201710044180 A CN201710044180 A CN 201710044180A CN 106874409 A CN106874409 A CN 106874409A
- Authority
- CN
- China
- Prior art keywords
- point
- away
- node
- level
- file
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/13—File access structures, e.g. distributed indices
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Generation (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention belongs to computer information technology field, and in particular to the storage method and device of a kind of cloud data.The storage method of the cloud data that the present invention is provided includes：Calculate the bounding box of cloud data；Carry out recurrence cutting to the bounding box according to octree structure, each level one point of correspondence of the octree structure away from；After each cutting, point according to current level is away from judging whether the point in each subelement space that cutting is obtained belongs to current level, storage in the form of a file belongs to the information of the point of current level, wherein, the node in the current level of correspondence of each subelement space that cutting is obtained, each node one file of correspondence.The storage method and device of the cloud data that the present invention is provided, reduce the amount of storage of cloud data, improve reading, inquiry, the rendering efficiency of cloud data.
Description
Technical field
The present invention relates to computer information technology field, and in particular to the storage method and device of a kind of cloud data.
Background technology
Cloud point data refer to that scanning data is recorded in dots, and each point includes three-dimensional coordinate, and some may contain
There are colouring information (RGB) or Reflection intensity information (Intensity).The data for generally being obtained when being scanned can be according to scanning
Space size be in geometric growth, therefore the usual needs of operation such as storage, inquiry, readings of cloud data are carried out by computer
Substantial amounts of calculating, it is ageing very poor.
The content of the invention
The storage method and device of the cloud data provided for defect of the prior art, the present invention, reduce a cloud
The amount of storage of data, improves reading, inquiry, the rendering efficiency of cloud data.
In a first aspect, a kind of storage method of cloud data that the present invention is provided, including：Calculate the encirclement of cloud data
Box；Recurrence cutting, each level one point of correspondence of the octree structure are carried out to the bounding box according to octree structure
Away from；After each cutting, point according to current level is away from judging whether the point in each subelement space that cutting is obtained belongs to
Current level, in the form of a file storage belongs to the information of the point of current level, wherein, each subelement space that cutting is obtained
A node in the current level of correspondence, each node one file of correspondence.
The storage method of the cloud data that the present embodiment is provided, from root node to all of intermediate node and leaf node,
Comprising the point for not having to repeat in point data, and All Files in the corresponding file of each node, cloud data distribution is solved
When uneven, the very big problem of the quantity variance of point is included after equidistant segmentation in each subelement space.Reduce data storage
Amount of redundancy, save memory space.By point in the way of successively store, can be shown as required when loading and rendering
Precision is successively dispatched, and as the nodes for loading and render are more, renders the effect come just more preferably.
Preferably, also include：Volume according to the total and described bounding box of a cloud determine the point of root node away from；According to root section
The point of point away from successively determine the point of the point of each level away from, next level away from the point less than last layer level away from.
Preferably, the point according to root node away from successively determine the point of each level away from, including：Point of next level etc.
In less than last layer level point away from half.
Preferably, the point of the current level of the basis is away from judging whether the point in each subelement space that cutting is obtained belongs to
In current level, including：Point according to current level is away from by subelement spatial gridding；Point in traversal subelement space, root
The positional information at strong point judges the grid residing for point, if the grid is not occupied, it is determined that the point belongs to current level and mark
Remember that the grid is occupied；The point is put into next level and is judged if the grid has been occupied.
Preferably, after each cutting, also include：Judge current level whether less than default maximum fractionation depth；If
Current level is more than or equal to default maximum fractionation depth, then terminate cutting；If current level is less than default maximum fractionation depth
Degree, then each the subelement space for being obtained to cutting proceeds as follows respectively：The quantity of the point in judgment sub-unit space is
It is no minimum comprising points less than default, if being less than, terminate the cutting to the subelement space.
Preferably, the storage in the form of a file belongs to the information of the point of current level, including：With Lob formatted files
Storage belongs to the information of the point of current level.
Preferably, the file is deposited with the structure being layered, and the node required to meet layering sets up file, institute
Stating file includes the file of the node, the file of the node all child nodes in current layer, an index file；It is described
The quantity of the point that index of the index file comprising all nodes in the file and each node are included；If the file
The node of middle maximum level also includes child node, then also include next layer of file in the file.
Second aspect, a kind of storage device of cloud data that the present invention is provided, including：Bounding box computing module, is used for
Calculate the bounding box of cloud data；Recurrence cutting module is used for, and recurrence cutting is carried out to the bounding box according to octree structure,
The octree structure each level correspondence one point away from；Data memory module, for after each cutting, according to current layer
Away from judging whether the point in each subelement space that cutting is obtained belongs to current level, storage in the form of a file belongs to the point of level
In the information of the point of current level, wherein, the node in the current level of correspondence of each subelement space that cutting is obtained, often
One file of individual node correspondence.
The storage device of the cloud data that the present embodiment is provided, from root node to all of intermediate node and leaf node,
Comprising the point for not having to repeat in point data, and All Files in the corresponding file of each node, cloud data distribution is solved
When uneven, the very big problem of the quantity variance of point is included after equidistant segmentation in each subelement space.Reduce data storage
Amount of redundancy, save memory space.By point in the way of successively store, can be shown as required when loading and rendering
Precision is successively dispatched, and as the nodes for loading and render are more, renders the effect come just more preferably.
Preferably, also it is used for including putting away from computing module：Volume according to the total and described bounding box of a cloud determines root section
Point point away from；Point according to root node determines the point of each level away from the point of next level is away from less than last layer level away from successively
Point away from.
Preferably, in computing module, the point according to root node determines the point of each level away from bag to the point away from successively
Include：The point of next level equal to less than last layer level point away from half.
Brief description of the drawings
The flow chart of the storage method of the cloud data that Fig. 1 is provided by the embodiment of the present invention；
Fig. 2 is the schematic diagram that cutting is carried out by octree structure；
Fig. 3 is for the node of different levels is independent, combine the design sketch for rendering；
Fig. 4 is the example that recurrence cutting is carried out by octree structure；
Fig. 5 is the example that node file is deposited with the structure being layered；
The structured flowchart of the storage device of the cloud data that Fig. 6 is provided by the embodiment of the present invention.
Specific embodiment
The embodiment of technical solution of the present invention is described in detail below in conjunction with accompanying drawing.Following examples are only used for
Technical scheme is clearly illustrated, therefore is intended only as example, and protection of the invention can not be limited with this
Scope.
It should be noted that unless otherwise indicated, technical term used in this application or scientific terminology should be this hair
The ordinary meaning that bright one of ordinary skill in the art are understood.
As shown in figure 1, a kind of storage method of cloud data that the present embodiment is provided, including：
Step S1, calculates the bounding box of cloud data.
Wherein, bounding box is solid that can be comprising all cloud datas and volume minimum, such as spheroid, cube.
Most common OBBs algorithm has AABB bounding boxs (Axis-aligned bounding box), surrounds ball (Sphere), direction
Bounding box OBB (Oriented bounding box) and fixed direction hull FDH (Fixed directions hulls or
k-DOP).To adapt to the cutting of octree structure, preferably using the algorithm of AABB bounding boxs, the algorithm is this area to the present embodiment
General algorithm, will not be repeated here.
Step S2, recurrence cutting, each level correspondence one of octree structure are carried out to bounding box according to octree structure
Individual point away from.
Wherein, octree structure is quad-tree structure extension in three dimensions, is that a kind of description and retrieval are three-dimensional vertical
The tree of the having levels property in body space.Root node with bounding box as octree structure, is eight sons by bounding box cutting
Unitary space, obtains eight nodes of the first level, and for each subelement space is numbered, numbering is every in order to distinguish
Individual sub- unitary space, determines the locus in each subelement space, slit mode referring to Fig. 2, to each subelement space again
Cutting is carried out in the manner described above, the end condition until meeting recurrence cutting.Each node correspondence in octree structure
The sub- unitary space that cutting is obtained.
Wherein, point is put away from the coefficient degree that can represent level midpoint away from being each level midpoint and the minimum range of point.
The point of next level away from less than last layer level point away from, with this reach successively improve data precision purpose.
Step S3, after each cutting, point according to current level is away from judging in each subelement space that cutting is obtained
Point whether belong to current level, in the form of a file storage belong to current level point information, wherein, it is every that cutting is obtained
A node in the current level of individual sub- unitary space correspondence, each node one file of correspondence.
Wherein, belonging to the point of current level will not participate in judging again in the dicing process of next level.For being not belonging to
The point of current level, is judged again in the dicing process of next level.
Wherein, the information of point includes point coordinate in three dimensions, can also include that the colouring information of point or reflection are strong
Degree information etc..
Wherein, for no longer carrying out the subfile that the node of cutting will not be set up under its node, for the section not comprising point
Point will not set up corresponding file, can so reduce the amount of storage of file, and effectiveness of retrieval is improved when data are called in loading.
The storage method of the cloud data that the present embodiment is provided, from root node to all of intermediate node and leaf node,
Comprising the point for not having to repeat in point data, and All Files in the corresponding file of each node, cloud data distribution is solved
When uneven, the very big problem of the quantity variance of point is included after equidistant segmentation in each subelement space.Reduce data storage
Amount of redundancy, save memory space.By point in the way of successively store, can be shown as required when loading and rendering
Precision is successively dispatched, and as the nodes for loading and render are more, renders the effect come just more preferably.
In order to determine the point of each level away from the method that the present embodiment is provided also includes：According to cloud sum and a bounding box
Volume determine the point of root node away from, point according to root node away from successively determine the point of the point of each level away from, next level away from
Less than last layer level point away from.
Wherein, point cloud sum is the total quantity at bounding box midpoint.Volume according to a cloud sum and bounding box determines root section
Point point away from specific method be：Obtain point cloud sum and calculate the volume of bounding box, the volume of bounding box is divided by a cloud sum
Obtain the average external volume of a single point, by the average external volume of a single point open cube root obtain the point of root node away from.Such as, bounding box
Volume is 1 cubic metre, and point cloud sum is 1000 points, then be every cubic decimeter containing a point, point away from for one decimeter, i.e., 0.1 meter.
Certainly, calculating point during, can according to the actual requirements be multiplied by coefficient, with choose suitable point away from.
Wherein, the point according to root node away from successively determine the point of each level away from, including：The point of next level equal to less than
Last layer level point away from half.For example, root node midpoint spacing is 1.0, then its child node (i.e. next layer of root node
Level) point away from be 0.5, referring specifically to Fig. 3, wherein (a) is the effect that the point in root node is rendered, (b) is root section
The effect that is rendered of point in one child node of point, substantially be can be seen that from figure, and the lower cloud data of level is rendered more
Coarse, level loading higher and the point for rendering are more, render the effect come just more preferably, and (c) and (d) is by root node in Fig. 3
Rendering effect when being combined with child node.
On the basis of above-mentioned any means embodiment, in step S3, point according to current level is obtained away from judging cutting
Each subelement space in point whether belong to current level, including：Each the subelement space obtained to cutting is entered respectively
The following operation of row：
Step S31, point according to current level is away from by subelement spatial gridding.
Wherein, gridding refers to that subelement space is divided into multiple to put away from the cube for the length of side.
Step S32, the point in traversal subelement space, the positional information according to point judges the grid residing for point, if the net
Lattice are not occupied, it is determined that the point belongs to current level and indicia grid is occupied；The point is put into if the grid has been occupied
Next level is judged.
Wherein, step S32 is substantially to choose a point in each grid to be stored as the point of current level, is
Raising treatment effeciency, it is not necessary to travel through institute in subelement space a little, only need to travel through each grid, one is chosen from grid
Individual point.
The cutting of Octree is a recursive procedure, it is necessary to set end condition or segmentation limiting factor, in each cutting
Afterwards, the end condition for judging whether to meet recurrence cutting is also needed, on the basis of above-mentioned any means embodiment, is judged whether full
The end condition of sufficient recurrence cutting is specifically included：
Whether step S10, judge current level less than default maximum fractionation depth.
Wherein, used as a kind of tree construction, the depth of tree is the number of times split to Octree.Maximum fractionation depth is cutting
Maximum level.
Step S20, if current level is more than or equal to default maximum fractionation depth, terminates cutting.
Step S30, if current level is less than default maximum fractionation depth, each the subelement space obtained to cutting
Proceed as follows respectively：Whether the quantity of the point in judgment sub-unit space is minimum comprising points less than default, if being less than,
Then terminate the cutting in sub-unit space.
Wherein, it is minimum comprising counting by the minimum points included in single subelement space, that is, set minimum comprising points
It is N, when the quantity of the point in When subunits space is less than or equal to N, just terminates the cutting to the subelement space, otherwise, continues
Cutting is carried out to it.
Above-mentioned two cutting end condition does not have situation a little suitable for subelement space, to the empty node of no
Continue cutting be it is insignificant, to institute either with or without comprising put node we it is all read as sky, file is not generated and is deposited
The data for storing can be so compressed by storage.It is because cloud data is the surface of object and generally simply right every time
The object of single angle measures range finding, and it is all empty that this allows for most of space, so, the node of the Octree of majority
Also will only individual other child node.By defining two end conditions of cutting simultaneously, can avoid creating complete eight fork
Tree, i.e., all leaf nodes in one Octree are all in same layer, and all of remaining node has eight child nodes.Cause
A bit, after using above two end condition, memory data output is reduced, the storage efficiency and retrieval that improve cloud data are imitated
Rate.Wherein, leaf node refers to the node not comprising child node.Fig. 4 is the example that recurrence cutting is carried out by octree structure, one
It is total to cutting twice, cutting altogether is three layers, and the part for having point data in space is represented with grey, the part for not having point data is used
White represents that only reservation includes the node of point data in structure tree.
For the amount of storage of further compressed data, on the basis of above-mentioned any means embodiment, in step S3, with text
The method that is preferable to carry out that the form storage of part belongs to the information of the point of current level includes：Belonged to Lob formatted file storages and worked as
The information of the point of preceding level.Lob formatted files store the information of cloud data in the form of binary stream, reduce data volume
Product, accelerates storage and the access speed of data.In addition, for most of data, 2 byte have been enough to express data
Precision, while other information a little, such as vector can also be represented, so the present embodiment selection only stores each with 2 byte
The coordinate of point, and stored with Lob forms.
Cloud data can index file in storing process, facilitate the lookup of node.In order to reduce index file, improve
Recall precision, on the basis of above-mentioned any means embodiment, file is deposited with the structure being layered, and is required to meet layering
Node set up file, file includes the file of the node, the file of the node all child nodes in current layer, one
Individual index file；The quantity of the point that index of the index file comprising all nodes in this document folder and each node are included；If
The node of maximum level also includes child node in this document folder, then also include next layer of file in this document folder.
Wherein, the depth of file hierarchy depends on the value of used layering step pitch.Layering requirement defines every
It is layered (as corresponding node sets up file) every a how many level, layering step pitch is the interval number of levels of layering, if
Layering step pitch is 5, then be divided into one layer at interval of 5 levels, the child node of memory node and lower 5 levels of node in file
Information, the index file records of file 6 hierarchical structures of level (own node and ensuing 5 levels).
Wherein, a node of Octree includes 8 child nodes, the numbering of this 8 child nodes is respectively 0,1,2,3,4,
5th, 6,7, the order of specific numbering can self-defining, once it is determined that number order, then the numbering of all child nodes is all in the order
Carry out.It is the numbering name file with the level where it and in its father node for the node of each Octree,
For example：It is root node (correspondence level 0 level, the level only includes a root node), the entitled l0 tables of file that the entitled l of file is represented
Show first child node (correspondence the first level) of root node, the entitled l03 of file represents the of first child node of root node
Four child nodes (correspondence third layer level).As shown in figure 5, knot of the corresponding file of all nodes according to layering step pitch to be layered
Structure is stored under data catalogues.
Wherein, the length of each index is 5bytes (byte) in index files (index file), and 1byte is used for describing
The child node that this node is included, for example：In l.index files 00000011 refer to comprising present node the 0th child node and
1st child node；Other 4bytes is used for storing the quantity at the node midpoint.
Wherein, the index of each node is deposited in index files according to the mode of depth-first, i.e., is first deposited according to tree construction
Then all nodes on one branch road of storage store the next node of branch road again until leaf node.
It is assumed that point cloud sum is 100000, for convenience of describing, if root node name is l.
In the first order：
Node l1 includes 10000 points, and node l4 includes 50000 points, and node l7 includes 40000 points.
In the second level：
The child node l11 of node l1 includes 3000 points, and the son section l14 of node l1 includes 7000 points.
The child node l41 of node l4 includes 10000 points, and child node l44 includes 20000 points, and child node l45 is included
20000 points.
The child node l77 of node l7 includes 40000 points.
Below by taking Fig. 4 as an example, the storage mode of index files is illustrated.Assuming that layering step pitch is 5, then all of node is all
In the file of root node, the relation of all nodes is all stored in the middle of index files according to the mode of depth-first for storage,
Then the information of index files storage is：
10010010 000000000000000000000000000100000 (corresponding node l)
00010010000000000000000000000000000010000 (corresponding node l1)
00000000000000000000000000000000000003000 (corresponding node l11)
00000000000000000000000000000000000007000 (corresponding node l14)
00110010000000000000000000000000000050000 (corresponding node l4)
00000000000000000000000000000000000010000 (corresponding node l41)
00000000000000000000000000000000000020000 (corresponding node l44)
00000000000000000000000000000000000020000 (corresponding node l45)
10000000000000000000000000000000000010000 (corresponding node l7)
00000000000000000000000000000000000040000 (corresponding node l77)
Above-mentioned is to understand that just the form as is stated for convenience, in index files is several during actual storage
According to be without any form and annotate.
Assuming that layering step pitch be 5, be layered structure deposited the step of be：First, it is root node (i.e. level 0
Level) file is set up, file l is named as convenience of description, by the text of all nodes in root node and its 5 later levels
Part is stored in file l, while including an index file in file l, the index file is used for institute in storage folder l
The quantity of the point that the index and each node for having node are included.Then, if the node of the layer 5 level in file l has son
Node, then set up the corresponding file of the node in file l, is named with the filename of the child node, e.g., the node
The entitled l02666 of file, the then entitled l02666 of corresponding file；Included in file l02666：Node l02666 with
And the file of the child node of lower 5 levels of node l02666, an index file；The index file is used for storage folder
The quantity of the point that the index of all nodes and each node are included in l02666.By that analogy, all nodes that go directly all are stored
To in corresponding file.
If cloud data amount is less than normal, when being sliced into most fine level and also not meeting layering and require, the density of cloud has been put
Through that can control in the range of a limitation of amount, then without continuing down cutting.For great cloud of data volume, layer is sliced into
Level has met layering and has required, but the data volume that now node is included is still very big, then also need to carry out cutting to node,
A file can be created this when, the corresponding lob files of the node below it are stored under this file, and are this
Individual node builds index files.So for the point cloud less than normal to data volume, treatment finally can result be that only one of which is deposited
The index files of root node are placed on, and multiple index files are had for the cloud data of magnanimity, satisfaction point is stored in respectively
Under the node that layer is required.The mode of index is set up in this layering, reduces index file quantity, is favorably improved recall precision.
After Point Cloud Processing is finished, the relation of all nodes is all placed on index files according to the mode of depth-first
It is central, when calling data, index files are directly read, every five bytes are exactly a nodal information, according to this information
Just clearly know the child node information below it, build the relation tree construction of egress, and looked into without by the way of traversal
Look for many times over, the recall precision of index is put forward significantly.Rendering layer can do corresponding control according to the different display mechanisms that renders, assorted
Which node data is situation render.Rendering layer is general according to current view point position and distance, and which level judgement should render
Node.
Based on the storage method identical inventive concept with above-mentioned cloud data, this implementation additionally provides a kind of cloud data
Storage device, as shown in fig. 6, including：Bounding box computing module, the bounding box for calculating cloud data；Recurrence dividing die
Block is used for, and recurrence cutting is carried out to bounding box according to octree structure, each level one point of correspondence of octree structure away from；Number
According to memory module, for after each cutting, point according to current level is away from judging in each subelement space that cutting is obtained
Point whether belong to current level, in the form of a file storage belong to current level point information, wherein, it is every that cutting is obtained
A node in the current level of individual sub- unitary space correspondence, each node one file of correspondence.
The storage device of the cloud data that the present embodiment is provided, from root node to all of intermediate node and leaf node,
Comprising the point for not having to repeat in point data, and All Files in the corresponding file of each node, cloud data distribution is solved
When uneven, the very big problem of the quantity variance of point is included after equidistant segmentation in each subelement space.Reduce data storage
Amount of redundancy, save memory space.By point in the way of successively store, can be shown as required when loading and rendering
Precision is successively dispatched, and as the nodes for loading and render are more, renders the effect come just more preferably.
Wherein, also including putting away from computing module, the point for determining root node according to the volume of a cloud sum and bounding box
Away from；Point according to root node away from successively determine the point of the point of each level away from, next level away from the point less than last layer level away from.
Wherein, point is in computing module, point according to root node away from successively determine the point of each level away from, including：It is next
The point of level equal to less than last layer level point away from half.
Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations；To the greatest extent
Pipe has been described in detail with reference to foregoing embodiments to the present invention, it will be understood by those within the art that：Its according to
The technical scheme described in foregoing embodiments can so be modified, or which part or all technical characteristic are entered
Row equivalent；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme, it all should cover in the middle of the scope of claim of the invention and specification.
Claims (10)
1. a kind of storage method of cloud data, it is characterised in that including：
Calculate the bounding box of cloud data；
Recurrence cutting, each level one point of correspondence of the octree structure are carried out to the bounding box according to octree structure
Away from；
After each cutting, point according to current level is away from judging whether the point in each subelement space that cutting is obtained belongs to
Current level, in the form of a file storage belongs to the information of the point of current level, wherein, each subelement space that cutting is obtained
A node in the current level of correspondence, each node one file of correspondence.
2. method according to claim 1, it is characterised in that also include：
Volume according to the total and described bounding box of a cloud determine the point of root node away from；
Point according to root node away from successively determine the point of the point of each level away from, next level away from the point less than last layer level away from.
3. method according to claim 2, it is characterised in that the point according to root node is away from successively determining each level
Point away from, including：The point of next level equal to less than last layer level point away from half.
4. method according to claim 1, it is characterised in that the point of the current level of basis is away from judging what cutting was obtained
Whether the point in each subelement space belongs to current level, including：
Each the subelement space obtained to cutting proceeds as follows respectively：
Point according to current level is away from by subelement spatial gridding；
Point in traversal subelement space, the positional information according to point judges the grid residing for point, if the grid is not occupied,
Then determine that the point belongs to current level and marks the grid to be occupied；The point is put into next layer if the grid has been occupied
Level is judged.
5. method according to claim 1, it is characterised in that after each cutting, also include：
Judge current level whether less than default maximum fractionation depth；
If current level is more than or equal to default maximum fractionation depth, terminate cutting；
If current level is less than default maximum fractionation depth, each the subelement space obtained to cutting carries out as follows respectively
Operation：Whether the quantity of the point in judgment sub-unit space is minimum comprising points less than default, if being less than, terminates to described
The cutting in subelement space.
6. method according to claim 1, it is characterised in that the storage in the form of a file belongs to the point of current level
Information, including：
Belong to the information of the point of current level with Lob formatted file storages.
7. method according to any one of claim 1 to 6, it is characterised in that the file is carried out with the structure being layered
Storage, the node required to meet layering sets up file, and the file includes the file of the node, and the node is current
The file of all child nodes, an index file in layer；Index of the index file comprising all nodes in the file
And the quantity of point that each node is included；If the node of maximum level also includes child node, the text in the file
Also include next layer of file in part folder.
8. a kind of storage device of cloud data, it is characterised in that including：
Bounding box computing module, the bounding box for calculating cloud data；
Recurrence cutting module is used for, and recurrence cutting is carried out to the bounding box according to octree structure, the octree structure
Each level correspondence one point away from；
Data memory module, for after each cutting, point according to current level is away from judging each subelement that cutting is obtained
Whether the point in space belongs to current level, and storage in the form of a file belongs to the information of the point of current level, wherein, cutting is obtained
The node in the current level of correspondence of each subelement space for arriving, each node one file of correspondence.
9. device according to claim 8, it is characterised in that also including putting away from computing module, be used for：
Volume according to the total and described bounding box of a cloud determine the point of root node away from；
Point according to root node away from successively determine the point of the point of each level away from, next level away from the point less than last layer level away from.
10. device according to claim 9, it is characterised in that the point in computing module, point according to root node away from
Successively determine the point of each level away from, including：The point of next level equal to less than last layer level point away from half.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710044180.7A CN106874409B (en) | 2017-01-19 | 2017-01-19 | The storage method and device of point cloud data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710044180.7A CN106874409B (en) | 2017-01-19 | 2017-01-19 | The storage method and device of point cloud data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106874409A true CN106874409A (en) | 2017-06-20 |
| CN106874409B CN106874409B (en) | 2019-11-19 |
Family
ID=59158895
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710044180.7A Active CN106874409B (en) | 2017-01-19 | 2017-01-19 | The storage method and device of point cloud data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106874409B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107886561A (en) * | 2017-11-30 | 2018-04-06 | 公安部物证鉴定中心 | Storage scheduling and display methods based on scene of a crime cloud data |
| CN108615254A (en) * | 2018-03-28 | 2018-10-02 | 广州市本真网络科技有限公司 | Point cloud rendering intent, system and device based on the quantization of tree lattice vector |
| CN108804714A (en) * | 2018-06-30 | 2018-11-13 | 浙江同创空间技术有限公司 | Point cloud data storage method and device |
| CN110928878A (en) * | 2019-11-19 | 2020-03-27 | 国网通用航空有限公司 | HDFS-based point cloud data processing method and device |
| KR20200037442A (en) * | 2017-08-30 | 2020-04-08 | 삼성전자주식회사 | METHOD AND APPARATUS FOR POINT-CLOUD STREAMING |
| CN111061820A (en) * | 2018-10-17 | 2020-04-24 | 宝马股份公司 | Method and apparatus for storing high-precision map |
| CN112465949A (en) * | 2020-11-26 | 2021-03-09 | 陕西铁道工程勘察有限公司 | Large-scale point cloud data organization method based on nested octree |
| CN113434514A (en) * | 2021-07-19 | 2021-09-24 | 中海油能源发展装备技术有限公司 | Voxelization index and output method of offshore oil and gas field point cloud model |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090060345A1 (en) * | 2007-08-30 | 2009-03-05 | Leica Geosystems Ag | Rapid, spatial-data viewing and manipulating including data partition and indexing |
| CN101615191A (en) * | 2009-07-28 | 2009-12-30 | 武汉大学 | The storage of magnanimity cloud data and real time visualized method |
| WO2011153624A2 (en) * | 2010-06-11 | 2011-12-15 | Ambercore Software Inc. | System and method for manipulating data having spatial coordinates |
| CN102682103A (en) * | 2012-04-28 | 2012-09-19 | 北京建筑工程学院 | Three-dimensional space index method aiming at massive laser radar point cloud models |
| CN105808672A (en) * | 2016-03-01 | 2016-07-27 | 重庆市勘测院 | Browser based mass three-dimensional point cloud data release method |
| CN106095907A (en) * | 2016-06-08 | 2016-11-09 | 江西理工大学 | Based on the laser point cloud data management method that Octree is integrated with three-dimensional R star tree |
-
2017
- 2017-01-19 CN CN201710044180.7A patent/CN106874409B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090060345A1 (en) * | 2007-08-30 | 2009-03-05 | Leica Geosystems Ag | Rapid, spatial-data viewing and manipulating including data partition and indexing |
| CN101615191A (en) * | 2009-07-28 | 2009-12-30 | 武汉大学 | The storage of magnanimity cloud data and real time visualized method |
| WO2011153624A2 (en) * | 2010-06-11 | 2011-12-15 | Ambercore Software Inc. | System and method for manipulating data having spatial coordinates |
| CN102682103A (en) * | 2012-04-28 | 2012-09-19 | 北京建筑工程学院 | Three-dimensional space index method aiming at massive laser radar point cloud models |
| CN105808672A (en) * | 2016-03-01 | 2016-07-27 | 重庆市勘测院 | Browser based mass three-dimensional point cloud data release method |
| CN106095907A (en) * | 2016-06-08 | 2016-11-09 | 江西理工大学 | Based on the laser point cloud data management method that Octree is integrated with three-dimensional R star tree |
Non-Patent Citations (1)
| Title |
|---|
| 律帅 等: "《基于数据类型转换的点云快速有损压缩算法》", 《图学学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20200037442A (en) * | 2017-08-30 | 2020-04-08 | 삼성전자주식회사 | METHOD AND APPARATUS FOR POINT-CLOUD STREAMING |
| KR102602380B1 (en) * | 2017-08-30 | 2023-11-16 | 삼성전자주식회사 | METHOD AND APPARATUS FOR POINT-CLOUD STREAMING |
| CN107886561A (en) * | 2017-11-30 | 2018-04-06 | 公安部物证鉴定中心 | Storage scheduling and display methods based on scene of a crime cloud data |
| CN107886561B (en) * | 2017-11-30 | 2021-11-09 | 公安部物证鉴定中心 | Crime scene point cloud data-based storage scheduling and display method |
| CN108615254A (en) * | 2018-03-28 | 2018-10-02 | 广州市本真网络科技有限公司 | Point cloud rendering intent, system and device based on the quantization of tree lattice vector |
| CN108804714A (en) * | 2018-06-30 | 2018-11-13 | 浙江同创空间技术有限公司 | Point cloud data storage method and device |
| CN111061820A (en) * | 2018-10-17 | 2020-04-24 | 宝马股份公司 | Method and apparatus for storing high-precision map |
| CN110928878A (en) * | 2019-11-19 | 2020-03-27 | 国网通用航空有限公司 | HDFS-based point cloud data processing method and device |
| CN112465949A (en) * | 2020-11-26 | 2021-03-09 | 陕西铁道工程勘察有限公司 | Large-scale point cloud data organization method based on nested octree |
| CN113434514A (en) * | 2021-07-19 | 2021-09-24 | 中海油能源发展装备技术有限公司 | Voxelization index and output method of offshore oil and gas field point cloud model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106874409B (en) | 2019-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106874409A (en) | The storage method and device of cloud data | |
| Douglas | Experiments to locate ridges and channels to create a new type of digital elevation model | |
| CN103425772B (en) | A kind of mass data inquiry method with multidimensional information | |
| US20200142384A1 (en) | Method of printing a 3d model from point cloud data | |
| CN102306180A (en) | Modeling method based on mass laser radar grid point cloud data | |
| CN107330214A (en) | Spatial configuration optimal method based on discretization Yu heuristic evolution algorithm | |
| CN105867310B (en) | The numerical-control processing method and system of T Spline Models are realized based on OCC | |
| JP2004164596A (en) | Method and system for traversing n-dimensional bi-tree | |
| CN102289991A (en) | Visual-variable-based automatic classification and configuration method of map lettering | |
| Paul et al. | Percolation threshold, Fisher exponent, and shortest path exponent for four and five dimensions | |
| KR102268676B1 (en) | 3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions | |
| Becker et al. | Spatial priority search: an access technique for scaleless maps | |
| CN109118588B (en) | Automatic color LOD model generation method based on block decomposition | |
| CN107206692B (en) | Three-dimensional object representation | |
| CN113868476A (en) | Octree point cloud preprocessing method based on local density | |
| CN113094463A (en) | Unstructured point cloud storage method, device, equipment and medium | |
| CN110222447A (en) | A kind of adaptive meshing algorithm method and system based on Octree | |
| CN110738726B (en) | Robot vision-guided three-dimensional object reconstruction method based on octree | |
| CN112380306A (en) | Adaptive correction method for Kergin spatial interpolation considering distribution balance | |
| Buchanan | Simulating wood using a voxel approach | |
| CN107194994A (en) | A kind of method and device without the demarcation surface points cloud data reconstruction face of cylinder | |
| EP3234924A1 (en) | Three-dimensional object representation | |
| CN103699569B (en) | A kind of index structure and indexing means | |
| Gallo et al. | Fuzzy B-splines: A surface model encapsulating uncertainty | |
| Elfick | Contouring by use of a triangular mesh |
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 | ||
| CP03 | Change of name, title or address |
Address after: 215010 Room 302, room 101, building 1, No.18, Daoyuan Road, science and Technology City, high tech Zone, Suzhou City, Jiangsu Province Patentee after: Zhongketuxin (Suzhou) Technology Co.,Ltd. Address before: 215000 Suzhou hi tech Zone, Jiangsu Province, No. 78 Patentee before: SUZHOU ZHONGKE TUXIN NETWORK TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address |