CN101615191B - Storage and real-time visualization implementation method of mass cloud data - Google Patents
Storage and real-time visualization implementation method of mass cloud data Download PDFInfo
- Publication number
- CN101615191B CN101615191B CN2009100633728A CN200910063372A CN101615191B CN 101615191 B CN101615191 B CN 101615191B CN 2009100633728 A CN2009100633728 A CN 2009100633728A CN 200910063372 A CN200910063372 A CN 200910063372A CN 101615191 B CN101615191 B CN 101615191B
- Authority
- CN
- China
- Prior art keywords
- cloud data
- file
- data
- image
- cloud
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Image Generation (AREA)
- Processing Or Creating Images (AREA)
Abstract
A storage method of mass cloud data comprises the steps of segmenting a mass cloud data file continuously and mapping the mass cloud data file into a process virtual address space in sequence; analyzing and storing the mass cloud data in a memory, and calculating the minimum external wrapping box of the cloud data; storing the cloud data as a multi-waveband image file when the cloud data in the memory exceeds the residual memory capacity; carrying out multi-segment re-sampling on the multi-waveband image file to construct an image pyramid; and repeating the above steps until the cloud data are all stored as multi-waveband images. A visualization implementation method of mass cloud data comprises the steps of reading all the multi-waveband images and calculating the minimum external wrapping box of all the cloud data, selecting proper level in the image pyramid according to the distance between a view point and the central point of the minimum external wrapping box so as to load data, and rendering the loaded cloud data; and switching the data according to the distance action range of the images of each layer in the image pyramid and the corresponding LOD layer level when the view point moves, and executing real-time rendering.
Description
Technical field
The invention belongs to geospatial information systems technology field, particularly relate to a kind of storage and real time visualized method of magnanimity cloud data.
Background technology
Laser scanner technique is the flourish high precision of the nineties in last century, obtains the new technique of object dimensional geometric data fast, direction by the recording laser pulse and calculate the three-dimensional coordinate that the mistiming of returning (or phase differential) obtains target surface after the target surface reflection, precision generally can reach the millimeter level, its characteristics are that acquisition speed is fast, the data precision height.Therefore, the cloud data that obtains of laser scanning is used widely gradually in fields such as digital historical relic, digital citys.Directly the cloud data based on sampled point has unique advantage in three-dimensional data is handled: only preserve sampled point and not only can reduce storage demand effectively, and the processing that can avoid network forming operation and network topology to concern, this data processing to large complicated scene is particularly important.
Along with the continuous expansion in 3-D scanning The Application of Technology field, and people improve constantly the requirement of scan-data precision, and the cloud data amount that spatial digitizer produces is more and more huger, to such an extent as to the monomer file size of cloud data is generally in the GB level.Because data volume is far longer than memory size, is called " magnanimity cloud data ".For example, in " Michelangelo " project of Stanford Univ USA in 1999, the precision of statue scanning is the highest to have reached 1/4 millimeter, and the original point cloud quantity that famous " David " statue scanning obtains is up to 1,100,000,000.Because memory size is far smaller than and need carries out visual cloud data amount, the cloud data of super large data volume can't once be loaded into internal memory, the requirement that the magnanimity cloud data is difficult to real-time rendering or does not reach real-time, interactive on common computer, therefore, on common computer, realize the real-time visual of magnanimity point cloud is become a problem demanding prompt solution.
What is more important, in real life, the universal time around people observe, the space of human eye and temporal resolution are always limited.Equally, the object in the virtual environment always shows different structures and feature at different level of detail, and therefore level of detail becomes the crucial factor of decision people's spatial cognition.According to the limited ultimate principle of resolution of eye, should neglect those human eyes the details of the space object that can not see, obtain relevant knowledge that true environment is more complete by limited some discrete level of detail.Simultaneously,, therefore, cause the phase mutual interference between points easily, influence effect of visualization because cloud data can not carry out effective occlusion culling when showing in real time.Therefore, (Level of Detail LOD) becomes the work an of necessity for the magnanimity cloud data is set up the level of detail model.
At present, the real time visualized method of magnanimity point cloud mainly contains following two kinds:
(1) based on the cloud data real-time visual of hierarchical structure
These class methods are utilized structures such as Octree or K-D tree that magnanimity cloud data collection is cut apart and are organized as hierarchical structure, to the multiresolution data structure under each piece divided data generated error control, when visual,, the requirement of level of detail plays up according to being extracted suitable hierarchical data, this method can visual magnanimity cloud data, but depend on complex data structures and preprocessing process, need complicated scheduling and controlling mechanism when visual.These class methods all have argumentation in following document: Wagner Toledo Correa.New Techniques for Out-Of-Core Visualization of Large Datasets.PrincetonUniversity, 2004.Matthias Hopf, Michael Luttenberger, Thomas Ertl.HierarchicalSplatting of Scattered 4D Data.IEEE Computer Graphics and Applications, 2004:64-72. put the Meng, look into red refined. draw based on the LOD control and the large-scale three dimensional cloud data of interior external memory scheduling. computer-aided design (CAD) and graphics journal, 2006. road bright moon, He Yongjian. the tissue of three-dimensional magnanimity cloud data and indexing means. Earth Information Science, 2008.
(2) the cloud data real-time visual of handling based on vacuate
These class methods have two kinds of implementations, and a kind of is to set the vacuate coefficient before showing, another kind is all data load to be set the vacuate coefficient again show after internal memory.The former is because to be loaded into data in the internal memory are points of reading of vacuate, therefore can not dynamically change the details of demonstration, and the latter needs all data all are loaded into earlier in the middle of the internal memory, when carrying out magnanimity point mysorethorn when visual, often because the memory size deficiency causes data load to be failed.Business software Polyworks commonly used promptly adopts above-mentioned vacuate method to show cloud data.
Summary of the invention
The object of the invention is to be difficult to a real-time visual difficult problem at the magnanimity cloud data, at the characteristics of cloud data, proposes a kind of method of utilizing multi-band image storage cloud data and based on the real time visualized method of this storage means.
A kind of magnanimity cloud data storage means is characterized in that: include following steps,
Step 1.1 is opened in computing machine in the cloud data file, obtains system for computer free memory capacity, and determines the maximum amount of data of each mapping according to this capacity;
Step 1.2, get a segmentation and be mapped to the process virtual address space from the cloud data file, the cloud data that parsing has been shone upon is also stored in internal memory, calculate in the internal memory all minimum outer bounding boxs of cloud datas, and estimate in the internal memory all data volume summations of cloud datas and the cloud data that next time shines upon;
Step 1.3, obtain computing machine system's free memory capacity of this moment, judge whether the data volume summation of estimating exceeds system's free memory capacity this moment, if do not exceed, all whether judging point cloud data file mapping, if then enter step 1.4, otherwise return step 1.2, continue to take off segmentation mapping from the cloud data file; If exceed system's free memory capacity this moment, enter step 1.4;
Step 1.4 is stored as the multi-band image file with the cloud data in the internal memory, the information of a dimension of wherein every wave band storage cloud data;
Step 1.5, the multi-band image file of having stored is carried out multistage resampling, for it makes up image pyramid, and the new multi-band image file that storage has an image pyramid replaces the multi-band image file of having stored, and with the minimum outer bounding box information stores of step 1.2 gained cloud data in the header of new multi-band image file; Discharge the shared internal memory of cloud data; Level in the described image pyramid corresponds respectively to the level of detail model level of cloud data;
Step 1.6, repeating step 1.2～1.5 all is stored as the multi-band image file until the cloud data file.
The present invention also provides corresponding magnanimity cloud data real time visualized method, it is characterized in that: include following steps,
Step 2.1 by reading the minimum outer bounding box information that writes down respectively in whole multi-band image files, is calculated the minimum outer bounding box of whole cloud datas; Minimum outer bounding box according to whole cloud datas calculates the initial viewpoint position;
Step 2.2 is extracted the image pyramid in the multi-band image file, according to the point of actual scanning apart from, screen resolution and resampling parameter, each tomographic image apart from reach in the computed image pyramid;
Step 2.3, according to the distance of initial viewpoint position to the minimum outer bounding box central point of whole cloud datas, according to each tomographic image in the image pyramid apart from reach, select the level in the image pyramid, load the cloud data of relevant details hierarchical model level, and the cloud data that will load is played up; When viewpoint moves, according to each tomographic image in the image pyramid switch the cloud data of corresponding level of detail model level apart from reach, and carry out real-time rendering.
The present invention utilizes the calculator memory residual capacity as constraint, and by the method for segmentation mapping, the magnanimity cloud data that will comprise a plurality of dimensional information is stored as multi-band image; Is that the magnanimity cloud data is set up many details level fast by multi-band image being carried out multistage resampling design of graphics as pyramidal mode, provides the reduced data of the multiresolution of magnanimity cloud data collection to express; By each tomographic image in the computed image pyramid apart from reach, the level that needs in the real-time visual process to load is controlled, realized the real-time visual of the detail of magnanimity cloud data effectively.Magnanimity cloud data storage that the present invention proposes and real time visualized method have method succinct, carry out the high advantage of efficient, solved a difficult problem of on common computer, carrying out magnanimity cloud data real-time visual effectively, be applicable to the cloud data collection of various data scales, be particularly useful for the real-time visual of the cloud data collection of the above data scale of GB level.
Description of drawings
Fig. 1 overview flow chart of the present invention;
The magnanimity cloud data Stored Procedure figure of Fig. 2 embodiment of the invention;
Fig. 3 magnanimity cloud data of the present invention segmentation mapping method;
Fig. 4 the present invention has the multi-band image structural drawing of image pyramid structure;
Fig. 5 screen prjection principle schematic;
The real-time visual process flow diagram of magnanimity cloud data in Fig. 6 embodiment of the invention.
Embodiment
In the technical solution of the present invention, after the magnanimity cloud data is stored as the multiband graphic file, by the processing of multiband graphic file being realized the real-time visual of magnanimity cloud data.These two processes can separately be implemented, and store the magnanimity cloud data by the geodata supplier, realize real-time visual when the user need check virtual image.For the ease of understanding the present invention, Fig. 1 provides and has utilized computer technology, realizes that the overall procedure of robotization processing is as follows:
The storage of magnanimity cloud data realizes including following steps,
Step 1.1 is opened in computing machine in the cloud data file, obtains system for computer free memory capacity, and determines the maximum amount of data of each mapping according to this capacity;
Step 1.2, get a segmentation and be mapped to the process virtual address space from the cloud data file, the cloud data that parsing has been shone upon is also stored in internal memory, calculate in the internal memory all minimum outer bounding boxs of cloud datas, and estimate in the internal memory all data volume summations of cloud datas and the cloud data that next time shines upon;
Step 1.3, obtain computing machine system's free memory capacity of this moment, judge whether the data volume summation of estimating exceeds system's free memory capacity this moment, if do not exceed, all whether judging point cloud data file mapping, if then enter step 1.4, otherwise return step 1.2, continue to take off segmentation mapping from the cloud data file; If exceed system's free memory capacity this moment, enter step 1.4;
Step 1.4 is stored as the multi-band image file with the cloud data in the internal memory, the information of a dimension of wherein every wave band storage cloud data;
Step 1.5, the multi-band image file of having stored is carried out multistage resampling, for it makes up image pyramid, and the new multi-band image file that storage has an image pyramid replaces the multi-band image file of having stored, and with the minimum outer bounding box information stores of step 1.2 gained cloud data in the header of new multi-band image file; Discharge the shared internal memory of cloud data.Set up image pyramid by multistage resampling, the level in the image pyramid is represented different sharpness, has been equivalent to set up the level of detail model of cloud data.Therefore the level in the image pyramid corresponds respectively to the level of detail model level of cloud data.
Step 1.6, repeating step 1.2～1.5 all is stored as the multi-band image file until the cloud data file.
Multi-band image all is with the document form storage, so be called the multi-band image file, simply is denoted as multi-band image among the figure.It should be noted that step 1.3 judges whether predicted data amount summation exceeds system free memory capacity this moment, is not system's free memory capacity that step 1.1 is obtained.System's free memory capability value that step 1.1 is obtained is because the consumption of program run is inequality with the free memory capability value that obtains in the step 1.3.Need computing machine to distribute continuous memory block during routine processes, and relate to the problem that internal memory reclaims and reallocates, so will obtain the free memory capacity again.Step 1.6 determines whether to circulate repeating step 1.2～1.5, can adopt the cloud data file whether all to be stored as the Rule of judgment of multi-band image file.Certainly, if the data volume summation of judge estimating in step 1.3 do not exceed this moment system's free memory capacity and the cloud data file all shine upon, enter step 1.4 then, can omit herein and repeat to judge, direct termination routine gets final product.
Magnanimity cloud data real-time visual realizes including following steps,
Step 2.1 by reading the minimum outer bounding box information that writes down respectively in whole multi-band image files, is calculated the minimum outer bounding box (abbreviating the outer bounding box size of total minimum among Fig. 1 as) of whole cloud datas; Minimum outer bounding box according to whole cloud datas calculates the initial viewpoint position;
Step 2.2 is extracted the image pyramid in the multi-band image file, according to the point of actual scanning apart from, screen resolution and resampling parameter, each tomographic image apart from reach in the computed image pyramid;
Step 2.3, according to the distance of initial viewpoint position to the minimum outer bounding box central point of whole cloud datas, according to each tomographic image in the image pyramid apart from reach, select the level in the image pyramid, load the cloud data (can abbreviate initial LOD hierarchical data as) of relevant details hierarchical model level, and the cloud data that will load is played up; When viewpoint moves, according to each tomographic image in the image pyramid switch the cloud data of corresponding level of detail model level apart from reach, and carry out real-time rendering.Switching is exactly to delete former LOD hierarchical data, loads new LOD hierarchical data, and playing up then, loading data gets final product.
Describe technical scheme of the present invention in detail below in conjunction with drawings and Examples:
Referring to Fig. 2, the principle of magnanimity cloud data storage means of the present invention is that the original magnanimity cloud data that will include N (N 〉=1) dimension information is stored as T (T 〉=1) image that includes N wave band, the multi-band image of having stored is carried out multistage resampling, make up image pyramid and memory image file.The storage implementation procedure of the embodiment of the invention adopts the computer realization robotization to handle, and may further comprise the steps:
Step 1.1 when opening the cloud data file, is obtained system's free memory capacity, and determines the maximum amount of data M of each mapping according to this capacity.Can determine the M value by the certain proportion of system's free memory capacity, as can the M value being set to 1/5 of system's free memory capacity.Because the data volume of cloud data file may be far longer than the Installed System Memory capacity, therefore the present invention adopts the mode of memory file mapping when carrying out the cloud data file division and utilize the image storage, cloud data file contiguous segmentation on the disk also is mapped to the process virtual address space successively, can handles the cloud data file of any size in this way.
Step 1.2 is got a segmentation and is mapped to the process virtual address space from the cloud data file.Carry out memory file when mapping, calculate the deviation post of mapping starting point with respect to cloud data file section start according to the maximum amount of data M of each mapping.Referring to Fig. 3, the map section of the cloud data file in the magnetic disk memory in the process virtual address space shifts.Carrying out the File mapping I first time
0The time, mapping starting point P
0Be positioned at the file section start, carrying out the File mapping I second time
1The time, the starting point P of mapping
1Be positioned at P
0+ M place is promptly with respect to file section start skew M.The later I of mapping each time
nIn, the mapped data amount all is M, the file reference position P of mapping
nBe positioned at P
N-1+ M place.
During memory file mapping each time, parsing is mapped to the cloud data of process virtual address space and storing and resolving is obtained in internal memory data, calculate and update stored in the minimum outer bounding box size of the cloud data in the internal memory, obtain the minimax three-dimensional coordinate x of minimum bounding box
Max, x
Min, y
Max, y
Min, z
Max, z
MinParsing is a process of extracting useful information from the cloud data of mapping.In the contiguous segmentation mapping, the minimum outer bounding box of the cloud data in the internal memory may change, can adopt the iteration update mode to determine the minimum outer bounding box of the cloud data in the current internal memory: that the embodiment of the invention is calculated is axial outer bounding box (AxisAligned Bounding Box, AABB), computing method are after reading the information of a point from the cloud data file at every turn, the X that will put just, Y, the Z three-dimensional coordinate is with acquired x
Max, x
Min, y
Max, y
Min, z
Max, z
MinCompare, for example, if x＞x
Max, then use x to substitute x
MaxIf, x＜x
Min, then use x to substitute x
MinIf, x 〉=x
MinAnd x≤x
Max, then outer bounding box is not done renewal.
Step 1.3, after each memory file mapping finishes, the precomputation internal memory mid point cloud data volume and the summation of the cloud data amount of mapping next time.When estimating summation, exactly with the maximum amount of data M of each mapping as shine upon possible cloud data amount next time.Obtain computing machine system's free memory capacity of this moment, if the data volume summation of estimating is less than system's free memory capacity of this moment, all whether judging point cloud data file mapping, if, then enter subsequent step, suddenly the next part file content in the cloud data file is mapped to the process virtual address space otherwise return previous step; If should be worth system's free memory capacity, then enter subsequent step greater than this moment;
Step 1.4 is stored as the multi-band image file with the cloud data that is stored in the internal memory, the information of a dimension of every wave band storage.Among the embodiment, the step that cloud data in the internal memory is stored into the multi-band image file on the disk is as follows: at first create a multi-band image that can hold all cloud datas in the internal memory, the wave band number of image is identical with the information dimension, and each dimensional information is written in the corresponding wave band.For example, the x that has that includes N point, y, z is when the cloud data collection of r four dimensions information is written to multi-band image, the figure image width of creating is W, height is H, W * H 〉=N wherein, and this image packets contains four wave bands, each wave band is stored the information of a dimension respectively, as can be with all x values of first wave band memory point cloud data centralization.Since W * H 〉=N, so the number of spots of image storage may be greater than N, and unnecessary point data is filled with invalid value.
Step 1.5 is carried out multistage resampling to the multi-band image file of having stored, and embodiment adopts neighbor point sampling method, promptly selects the value that a point is represented whole zone in each sample area, and sample rate soon and can not change the value of original point.The resampling level of image is set, as 1/4,1/16 multistage sampling such as grade, replace the multi-band image file of having stored by image being carried out the new multi-band image that multistage resampling and storage resampling result obtain to have image pyramid, what Fig. 4 showed is the multi-band image structural representation that has image pyramid among the embodiment; Its medium wave band 1 storage x coordinate information, wave band 2 storage y coordinate informations, wave band 3 storage z coordinate informations, wave band 4 storage intensity signals, wave band 5 storage color value r components, wave band 6 storage color value b components also can be established more information such as multiband storage normal.In each level of detail model level, the first level LOD0 is 1/k
1Sampling, the second level LOD1 are 1/k
2Sampling, the 3rd level LOD2 are 1/k
3Sampling, the 4th level LOD3 are 1/k
4Sampling ... follow-up can also have more multi-layered level, indicates with LODk among the figure.Then comprise minimum outer bounding box size in the header.
During concrete enforcement, resampling mode and grade can be set earlier, again multi-band image be resampled, storage resampling data and outer bounding box information after sampling is finished, obtain being with the pyramidal multi-band image file of figure, discharge the shared internal memory of cloud data then.
Step 1.6, repetitive segment mapping and generation multi-band image file all are stored as the multi-band image file until the cloud data file.A single point cloud data file is because the relation of memory size may be stored as one or more multi-band image file.Process ends when all finishing.
The principle of magnanimity cloud data real time visualized method of the present invention is to send into video card end establishment vertex cache by different LOD level (level in the image pyramid corresponds respectively to the LOD level of the cloud data) data of variation loading of judging viewpoint distance of bounding box central point outside whole cloud data minimums to draw.
The real-time visual implementation procedure of the embodiment of the invention adopts the computer realization robotization to handle, and may further comprise the steps:
Step 2.1 by reading the minimum bounding box information that writes down in whole multi-band image files, is calculated the minimum outer bounding box (promptly total minimum outer bounding box size) of these multi-band image whole cloud datas that file provides; Minimum outer bounding box according to whole cloud datas calculates the initial viewpoint position, and computing method are as follows: obtain the height value H of the outside surface S of the outer bounding box relative with sight line, the vertical angle of known view frustums is α, utilizes formula
Calculating is from the distance L of S face, with viewpoint from the central point of S face along sight line translation in the other direction L, the position that obtains is exactly the initial position of viewpoint.
Step 2.2, according to the point of actual scanning apart from, screen resolution and resampling parameter, in the computed image pyramid each tomographic image apart from reach, promptly each LOD level is apart from reach.
Referring to Fig. 5, the point of setting actual scanning is ε apart from (being the resolution of original point cloud data), the cloud data mid point that at sampling parameter is the 1/k correspondence is apart from being k* ε (Fig. 5 mid point 1 and the distance of putting between 2), viewpoint is d to the distance of sampled point, view frustums is w at the width of sample point, the horizontal angle of sight line is θ, and screen resolution is x, and the pixel distance that two adjacent sampled points project on the screen is p.Can obtain the relation between d and the sampling parameter 1/k:
This shows, when two adjacent sampled points project to pixel distance p on the screen less than the projection minimum threshold, show that then the LOD hierarchical data in the pairing image pyramid of sampling parameter 1/k is too intensive when projecting to screen under the condition of d in viewpoint to the sampled point distance, need hierarchical data more sparse in the load image pyramid to play up; As p during greater than the projection max-thresholds, show that then the LOD hierarchical data in the pairing image pyramid of sampling parameter 1/k is too sparse when projecting to screen under the condition of d in viewpoint to the sampled point distance, need hierarchical data more intensive in the load image pyramid to play up.
Step 2.3, according to the distance of viewpoint to the outer bounding box central point of cloud data minimum, suitable level is carried out data load in the selection image pyramid, and the cloud data that will load is played up; When viewpoint moves, carry out data according to the LOD level apart from reach selection correspondence of each tomographic image in the image pyramid and switch, and carry out real-time rendering.As shown in Figure 6, the concrete step of embodiment is as follows:
Step 2.31, according to the distance of initial viewpoint position to the outer bounding box central point of cloud data minimum, and the distance range of image pyramid Different L OD level effect, calculate the LOD level.
Step 2.32, the data that load corresponding LOD level in individual multi-band image enter internal memory.Data are sent to the video card end, generate a vertex cache, and corresponding data in the deletion internal memory.
Step 2.33, repeating step 2.32, LOD hierarchical data corresponding in all multi-band images all is loaded into the video card end; The cloud data that has loaded is played up.
Step 2.34 when viewpoint moves, is judged the whether change of corresponding LOD level according to the distance of viewpoint bounding box central point outside the cloud data minimum.
When LOD level not during change, loaded with new data not uses loaded data to play up.When the LOD level changes, all multi-band images corresponding with this level are handled one by one, from step 2.35,
Step 2.35 is calculated the size of the data that will read in individual multi-band image, obtains system's free memory capacity, whether judge this size of data less than system's free memory capacity this moment, if less than, step 2.36 then entered, otherwise abandon loaded with new data, enter step 2.38.Embodiment is provided with this step, is beneficial to guarantee under the condition that memory size allows, and realizes that cloud data is visual.
Step 2.36, the original vertex cache data of deletion video card end.Corresponding LOD hierarchical data enters internal memory in the loading single image, and the data that are loaded in the internal memory are sent to vertex cache of video card end establishment, and the data in the deletion internal memory.
Step 2.37, repeating step 2.35 and 2.36 finishes until image traversal, and corresponding LOD hierarchical data all is loaded into the video card end or owing to system's free memory off-capacity enters step 2.38 in all multi-band images;
Step 2.38 is played up the cloud data that is loaded into the video card end.
Step 2.39, repeating step 2.34 to 2.38 withdraws from until program.
Claims (2)
1. magnanimity cloud data storage means is characterized in that: include following steps,
Step 1.1 is opened in computing machine in the cloud data file, obtains system for computer free memory capacity, and determines the maximum amount of data of each mapping according to this capacity;
Step 1.2, get a segmentation and be mapped to the process virtual address space from the cloud data file, the cloud data that parsing has been shone upon is also stored in internal memory, calculate in the internal memory all minimum outer bounding boxs of cloud datas, and estimate in the internal memory all data volume summations of cloud datas and the cloud data that next time shines upon;
Step 1.3, obtain computing machine system's free memory capacity of this moment, judge whether the data volume summation of estimating exceeds system's free memory capacity this moment, if do not exceed, all whether judging point cloud data file mapping, if then enter step 1.4, otherwise return step 1.2, continue to take off segmentation mapping from the cloud data file; If exceed system's free memory capacity this moment, enter step 1.4;
Step 1.4 is stored as the multi-band image file with the cloud data in the internal memory, the information of a dimension of wherein every wave band storage cloud data;
Step 1.5, the multi-band image file of having stored is carried out multistage resampling, for it makes up image pyramid, and the new multi-band image file that storage has an image pyramid replaces the multi-band image file of having stored, and with the minimum outer bounding box information stores of step 1.2 gained cloud data in the header of new multi-band image file; Discharge the shared internal memory of cloud data; Level in the described image pyramid corresponds respectively to the level of detail model level of cloud data;
Step 1.6, repeating step 1.2～1.5 all is stored as the multi-band image file until the cloud data file.
2. one kind generates the magnanimity cloud data real time visualized method of realizing on the multi-band image file basis according to the described method of claim 1, it is characterized in that: include following steps,
Step 2.1 by reading the minimum outer bounding box information that writes down respectively in whole multi-band image files, is calculated the minimum outer bounding box of whole cloud datas; Minimum outer bounding box according to whole cloud datas calculates the initial viewpoint position;
Step 2.2 is extracted the image pyramid in the multi-band image file, according to the point of actual scanning apart from, screen resolution and resampling parameter, each tomographic image apart from reach in the computed image pyramid;
Step 2.3, according to the distance of initial viewpoint position to the minimum outer bounding box central point of whole cloud datas, the contrast image pyramid in each tomographic image apart from reach, select the level in the image pyramid, load the cloud data of relevant details hierarchical model level, and the cloud data that will load is played up; When viewpoint moves, according to each tomographic image in the image pyramid switch the cloud data of corresponding level of detail model level apart from reach, and carry out real-time rendering.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100633728A CN101615191B (en) | 2009-07-28 | 2009-07-28 | Storage and real-time visualization implementation method of mass cloud data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009100633728A CN101615191B (en) | 2009-07-28 | 2009-07-28 | Storage and real-time visualization implementation method of mass cloud data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101615191A CN101615191A (en) | 2009-12-30 |
| CN101615191B true CN101615191B (en) | 2011-05-11 |
Family
ID=41494833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009100633728A Expired - Fee Related CN101615191B (en) | 2009-07-28 | 2009-07-28 | Storage and real-time visualization implementation method of mass cloud data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101615191B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110189396A (en) * | 2019-06-10 | 2019-08-30 | 浙江科澜信息技术有限公司 | A kind of scene figure construction method, system and the relevant apparatus of view-based access control model |
Families Citing this family (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8712984B2 (en) * | 2010-03-04 | 2014-04-29 | Microsoft Corporation | Buffer pool extension for database server |
| CN102096943B (en) * | 2011-01-14 | 2012-07-18 | 天津大学 | Method for extracting and rendering characteristic lines in three-dimensional (3D) real-time landscape painting |
| CN103970781A (en) * | 2013-01-30 | 2014-08-06 | 深圳市同洲电子股份有限公司 | Embedded browser picture decoding device and method |
| CN103914871B (en) * | 2014-03-06 | 2016-06-08 | 河南农业大学 | The method of body surface coordinate point is chosen based on the interactive mode of cloud data |
| CN104392480B (en) * | 2014-11-12 | 2015-09-02 | 山东地纬数码科技有限公司 | Within external memory exchanged form realize based on point the rendering intent of global illumination effect |
| CN106096016A (en) * | 2016-06-24 | 2016-11-09 | 北京建筑大学 | A kind of network three-dimensional point cloud method for visualizing and device |
| CN106445329A (en) * | 2016-08-31 | 2017-02-22 | 浙江科澜信息技术有限公司 | Image capture method for dynamic balance scene loading hierarchy |
| CN106780319A (en) * | 2016-11-11 | 2017-05-31 | 航天恒星科技有限公司 | The processing method and processing device of super large image |
| CN106874409B (en) * | 2017-01-19 | 2019-11-19 | 苏州中科图新网络科技有限公司 | The storage method and device of point 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 |
| CN108470374B (en) * | 2018-04-08 | 2022-03-25 | 中煤航测遥感集团有限公司 | Mass point cloud data processing method and device |
| CN109410118B (en) * | 2018-09-12 | 2023-06-13 | 广州南方卫星导航仪器有限公司 | Rapid loading and displaying method for massive point cloud data |
| CN111476872B (en) * | 2019-01-23 | 2022-10-11 | 华为技术有限公司 | Image drawing method and image drawing device |
| CN110021072B (en) * | 2019-04-03 | 2020-08-07 | 武汉大学 | Holographic mapping-oriented multi-platform point cloud intelligent processing method |
| CN111684494A (en) * | 2019-05-23 | 2020-09-18 | 深圳市大疆创新科技有限公司 | Method and device for displaying point cloud in real time and computer storage medium |
| CN110457499A (en) * | 2019-07-19 | 2019-11-15 | 广州启量信息科技有限公司 | Indexing means, device, terminal device and the medium of a kind of pair large-scale point cloud data |
| CN110458931B (en) * | 2019-08-16 | 2020-11-24 | 北京天马时空网络技术有限公司 | LOD (level of detail) grade-based three-dimensional model data processing method and related equipment |
| CN110659440B (en) * | 2019-09-25 | 2023-04-18 | 云南电网有限责任公司曲靖供电局 | Method for rapidly and dynamically displaying different detail levels of point cloud data large scene |
| CN111738903B (en) * | 2020-05-18 | 2024-05-10 | 完美世界(北京)软件科技发展有限公司 | Method, device and equipment for optimizing layered material of object |
| CN111552555B (en) * | 2020-05-19 | 2023-07-28 | 北京数字绿土科技股份有限公司 | Point cloud visualization method and equipment |
| CN111968211A (en) * | 2020-08-28 | 2020-11-20 | 北京睿呈时代信息科技有限公司 | Memory, and drawing method, system and equipment based on point cloud data |
| CN112395101B (en) * | 2020-10-12 | 2021-07-20 | 天津航天中为数据系统科技有限公司 | Big data fast rendering method based on bidirectional data processing mechanism |
| CN112215951B (en) * | 2020-10-23 | 2022-04-29 | 华中科技大学 | Out-of-core multi-resolution point cloud representation method and point cloud display method |
| CN112308974B (en) * | 2020-10-30 | 2022-02-18 | 南京航空航天大学 | Large-scale point cloud visualization method for improving octree and adaptive reading |
| CN113190515B (en) * | 2021-05-14 | 2022-11-29 | 重庆市勘测院 | Heterogeneous parallel computing-based urban mass point cloud coordinate transformation method |
| CN114494623A (en) * | 2022-01-24 | 2022-05-13 | 北京中航双兴科技有限公司 | LOD-based terrain rendering method and device |
| CN116109470B (en) * | 2023-04-13 | 2023-06-20 | 深圳市其域创新科技有限公司 | Real-time point cloud data rendering method, device, terminal and storage medium |
| CN117807154B (en) * | 2024-02-28 | 2024-04-30 | 成都菲宇科技有限公司 | Time sequence data visualization method, device and medium for display system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704939A (en) * | 2004-05-26 | 2005-12-07 | 鸿富锦精密工业（深圳）有限公司 | Point cloud three-dimensional handling system and method |
| CN101377851A (en) * | 2007-08-29 | 2009-03-04 | 鸿富锦精密工业（深圳）有限公司 | System and method for computing a minimum distance from point cloud to point cloud |
-
2009
- 2009-07-28 CN CN2009100633728A patent/CN101615191B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704939A (en) * | 2004-05-26 | 2005-12-07 | 鸿富锦精密工业（深圳）有限公司 | Point cloud three-dimensional handling system and method |
| CN101377851A (en) * | 2007-08-29 | 2009-03-04 | 鸿富锦精密工业（深圳）有限公司 | System and method for computing a minimum distance from point cloud to point cloud |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110189396A (en) * | 2019-06-10 | 2019-08-30 | 浙江科澜信息技术有限公司 | A kind of scene figure construction method, system and the relevant apparatus of view-based access control model |
| CN110189396B (en) * | 2019-06-10 | 2023-03-28 | 浙江科澜信息技术有限公司 | Scene graph construction method, system and related device based on vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101615191A (en) | 2009-12-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101615191B (en) | Storage and real-time visualization implementation method of mass cloud data | |
| CN110738721B (en) | Three-dimensional scene rendering acceleration method and system based on video geometric analysis | |
| CN105378796B (en) | Scalable volume 3D reconstruct | |
| US8570322B2 (en) | Method, system, and computer program product for efficient ray tracing of micropolygon geometry | |
| KR101546703B1 (en) | System for processing massive bim data of building | |
| US9754405B1 (en) | System, method and computer-readable medium for organizing and rendering 3D voxel models in a tree structure | |
| CN101441779B (en) | Rapid dynamic drafting method of magnanimity digital terrain | |
| US20100179788A1 (en) | System and method for hybrid solid and surface modeling for computer-aided design environments | |
| US20120268465A1 (en) | Image file generation device, image processing device, image file generation method, image processing method, and data structure for image files | |
| KR101546705B1 (en) | Method for visualizing building-inside bim data by bim data process terminal | |
| US10460510B2 (en) | Methods and systems for viewing a three-dimensional (3D) virtual object | |
| Strugar | Continuous distance-dependent level of detail for rendering heightmaps | |
| Wand et al. | Interactive Editing of Large Point Clouds. | |
| CN113034657B (en) | Rendering method, device and equipment for illumination information in game scene | |
| CN112906125B (en) | Light-weight loading method for BIM model of railway fixed facility | |
| He et al. | All range and heterogeneous multi-scale 3D city models | |
| US10891780B2 (en) | Methods and systems for viewing a three-dimensional (3D) virtual object | |
| JPH05266213A (en) | High-speed display system for three-dimensional space data | |
| CN111275806A (en) | Parallelization real-time rendering system and method based on points | |
| US11423618B2 (en) | Image generation system and method | |
| CN116958457A (en) | OSGEarth-based war misting effect drawing method | |
| Chao et al. | The computation of delaunay triangulation of LiDAR point cloud based on GPU | |
| Schollmeyer et al. | Direct isosurface ray casting of NURBS-based isogeometric analysis | |
| KR100392516B1 (en) | real-time rendering method for noninterpolated volume data | |
| Levinski et al. | Interactive function-based artistic shape modeling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: CHANGSHU ZIJIN INTELLECTUAL PROPERTY SERVICE CO., Free format text: FORMER OWNER: WUHAN UNIVERSITY Effective date: 20121211 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 430072 WUHAN, HUBEI PROVINCE TO: 215500 SUZHOU, JIANGSU PROVINCE |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20121211 Address after: 215500 Changshou City South East Economic Development Zone, Jiangsu, Jin Road, No. 8 Patentee after: Changshu Zijin Intellectual Property Service Co., Ltd. Address before: 430072 Hubei city of Wuhan province Wuchang Luojiashan Patentee before: Wuhan University |
|
| ASS | Succession or assignment of patent right |
Owner name: CHANGSHU BROADCASTIONG COOPERATION Free format text: FORMER OWNER: CHANGSHU ZIJIN INTELLECTUAL PROPERTY SERVICE CO., LTD. Effective date: 20130604 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20130604 Address after: 215500 No. 29, Yu Bei Road, Suzhou, Jiangsu, Changshou City Patentee after: Changshu Municipal Broadcast & TV Station Address before: 215500 Changshou City South East Economic Development Zone, Jiangsu, Jin Road, No. 8 Patentee before: Changshu Zijin Intellectual Property Service Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110511 Termination date: 20180728 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |