CN101916301B - Three-dimensional spatial data adaptive pre-scheduling method based on spatial relationship - Google Patents
Three-dimensional spatial data adaptive pre-scheduling method based on spatial relationship Download PDFInfo
- Publication number
- CN101916301B CN101916301B CN2010102691218A CN201010269121A CN101916301B CN 101916301 B CN101916301 B CN 101916301B CN 2010102691218 A CN2010102691218 A CN 2010102691218A CN 201010269121 A CN201010269121 A CN 201010269121A CN 101916301 B CN101916301 B CN 101916301B
- Authority
- CN
- China
- Prior art keywords
- influence
- spatial
- scheduling
- range
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention relates to the technical field of geography and geoscience spatial information processing, in particular to a three-dimensional spatial data adaptive pre-scheduling method based on spatial relationship. In the method, a nine-intersection model in the spatial relationship is taken as the theoretical basis, the characteristics of cache management and spatial indexing are combined, the ideas of spatial clustering and spatial interpolation are adopted, the spatial objects in the cache are taken as the sample data, the hit rate of the objects is taken as the estimative weight, the spatial object information in the spatial index is taken as the data to be estimated and simultaneously the memory capacity of the system and the computing capability of the CPU are taken into account. The pre-scheduling method can improve the scheduling speed of the spatial data. Tree index, trigger and multithreading technology are adopted, therefore, the method effectively solves the problems of system resource plunder and use efficiency and can remarkably improve the problem of data communication blocking. The algorithm can adapt to the scheduling requirement of the three-dimensional spatial data, can be further used for multi-dimensional spatial data and can be popularized and used in various professional GIS software.
Description
Technical field
The present invention relates to geography and ground spatial information processing technology field, relate in particular to a kind of three-dimensional space data self-adaptation pre-scheduling method based on spatial relationship.
Background technology
The response time that three-dimensional space data real-time visual and the online application of multi-user etc. all need be exceedingly fast, this performance to extensive three-dimensional space data data base administration has proposed great challenge.Receive the restriction of current computer hardware, make and when three-dimensional scenic is drawn, can not disposable required whole three-dimensional space datas be called in internal memory, must dynamically call in required data according to the needs of current three-dimensional scenic.Thereby can from file or three-dimensional space data storehouse, the required data of dynamic dispatching efficiently become the key that ensures three-dimensional real-time rendering fluency to Installed System Memory.
At present; Mostly adopt certain spatial organization to set up the consistent storage index structure of memory, external memory both at home and abroad, the method for the preparatory fetch policy that use is speculated, corresponding cache management strategy and multi-threaded parallel scheduling etc. realizes the increment dynamic dispatching of mass data.Because technology such as image tile and pyramid structure are for the management of raster data and use comparative maturity in real time, the GeoRaster of many business softwares such as Google Earth, Skyline and Oracle 10/11g adopts these mature technologies to realize the management of magnanimity raster data.Oracle has abandoned the spatial character of data, merely from OO angle, has developed the pre-scheduling technology based on object relationship figure.This technology adopts object-oriented thought, for the object that has succession, derivation, associating, paradigmatic relation is set up object relationship figure, and realizes the tracking and the loading of pre-scheduling object along this figure.Funkhouser etc. are incorporated into the method that data pre-head is got loading in the visualization application of big data quantity at first.The dynamic dispatching method of existing most of landform roaming systems mostly adopts the forecasting mechanism based on planar grid.Based on the forecast model of viewpoint is the anticipation function that a viewpoint is set up in position, direction of motion, movement velocity, angular velocity etc. according to current viewpoint; According to current mutual characteristics; The several possible viewpoint positions of precomputation, the viewpoint of using precomputation to obtain carry out the observability of data and judge the prestrain of accomplishing data.The OEMM of existing canonical system such as Italian CRS group development is based on the interior external memory integral structure of Octree; The iwalk of the Correa of IBM Corporation etc. exploitation is based on the dispatching technique of the multithreading stand-alone environment of Octree, and the memory, external memory structure based on the k-D tree construction of German Saarland university exploitation is used for the dynamic data dispatching method etc. of the real-time ray tracing of large scale scene.Oracle 11g provides the pre-scheduling technology based on the complex object that concerns between the object in addition.This technology adopts OO thought to set up object relationship figure according to relation between objects, is begun by the initialization object then, reads other objects in advance through the adduction relationship between the object.In this technology, application program can be provided with some search rules, like the degree of depth of retrieval, so that in the primary network transmission, retrieve a collection of relevant object.
Summary of the invention
Technical matters to above-mentioned existence; The purpose of this invention is to provide a kind of three-dimensional space data self-adaptation pre-scheduling method based on spatial relationship; To Installed System Memory, ensure the fluency of three-dimensional real-time rendering with the required data of dynamic dispatching efficiently from file or three-dimensional space data storehouse.
For achieving the above object, the present invention adopts following technical scheme:
1. adopt trigger mechanism to realize real-time monitoring to the system call situation;
2. utilize the three dimensions object that has been scheduled in the buffer memory to set up the sample index tree, moving object is formed the space clustering structure by its spatial relation;
3. the factor of influence R of range of influence in the calculation sample tree calculates with recursive fashion from bottom to top, tries to achieve the factor of influence of each range of influence in the sample tree;
4. the root node from the sample tree begins, and is that bounding box is retrieved with its range of influence, retrieves the prefetching object that the root node range of influence comprises;
5. calculate the internal memory summation size K that prefetching object takies,
6. relatively the confession pre-scheduling object of K and setting remains spendable memory headroom K
ReSize, K
Re=K
Pre-K
Use, wherein, K
PreFor setting pre-scheduling spatial cache, K
UseFor taking the pre-scheduling spatial cache; If K≤K
Re, the object that retrieves is called in buffer memory as the pre-scheduling object, finish the pre-scheduling process; If K＞K
Re, 7. to step;
7. the range of influence of one deck is by the big or small descending sort brotgher of node of its factor of influence R under in sample being set;
8. by institute's alignment preface, choose first node for needing the screening range of influence;
9. the range of influence is that bounding box carries out global search, retrieves the prefetching object that the screening range of influence comprises;
10. calculate the internal memory summation size K that prefetching object takies,
K wherein
iThe memory size that the expression prefetching object takies;
Work as K
Re, finish the pre-scheduling process at=0 o'clock;
Said step is 3. middle to be adopted
The factor of influence R of range of influence in the calculation sample tree, wherein V representes the volume of range of influence, R
iThe factor of influence of representing its child node range of influence, V
iThe volume of representing its child node range of influence.
The present invention has the following advantages and good effect:
1) the present invention proposes the space clustering structure that sample is set, can come efficient and definite scientifically and rationally pre-scheduling object according to the space of moving object;
2) the present invention has adopted the mechanism of trigger and multithreading, can guarantee normally carrying out of system operation, avoids network congestion to greatest extent;
3) pre-scheduling method of the present invention has adaptive characteristic, can select the pre-scheduling object automatically, has solved the problem of system resource contention and service efficiency effectively.
Description of drawings
Fig. 1 is the process flow diagram of the three-dimensional space data self-adaptation pre-scheduling method based on spatial relationship provided by the invention.
Fig. 2 is the structural representation of the sample tree that proposes of the present invention.
Embodiment
Three-dimensional space data self-adaptation pre-scheduling method based on spatial relationship provided by the invention; Take all factors into consideration the scope and the importance of scheduled spatial object; Select optimum pre-scheduling object; Based on the pre-scheduling strategy of space cluster analysis according to hardware operating position self-adaptations such as internal memories adjustment set the pre-scheduling memory headroom, search the pre-scheduling object that satisfies condition automatically, and according to the computing power of CPU; In time start and close the pre-scheduling program, combine accompanying drawing that the present invention is described further with specific embodiment below:
It is theoretical foundation that this method is handed over model with nine in the spatial relationship, adopts the thought of space clustering and space interpolation, and the spatial object in the buffer memory is regarded as sample data, sets up the sample tree.Estimate the factor of influence value of upper level range of influence through the factor of influence of moving object or sub-range of influence in the sample tree.Take into account the real-time operating position of internal memory of system and the computing power of CPU simultaneously, adopt the related mechanism of trigger and multithreading to come realization the system computing power and the scheduling situation of monitoring CPU in real time, start the pre-scheduling process.
Nine friendship models are set of model of a cover application space inquiry of OGC (Open Geospatial Consortium, open space information association) formulation.Nine hand over model according to different dimensions, and the spatial relationship between the dissimilar geometric objects is come the basic operator of definition space inquiry.
Basic geometric object element is had a few, line, and face, they all have very clear and definite border, inside, outside.Related definition is as shown in the table:
| Geometric object | The border | Inner | Outside |
| The point | Non-boundary | Point itself | Zone beyond the some object |
| Line | The end points of line object | Line object except that end points | The zone that line object is outer |
| Face | In the face of resembling the control limit | In the face of resembling except that the zone behind the border | In the face of resembling outer zone itself |
Nine hand over model to pass through the relatively outside of two or more objects, the border, and inner, and ask friendship, consider to ask the dimension of handing over the result, so be suitable for various objects.This model is used inside, border and the outside set of I (a), B (a), E (a) expression geometric object a respectively; Friendship is asked in outside, border, inner set through to two geometric object a, b; Utilization dimension development method is expanded; And will ask and hand over the result to be filled in the nine friendship model tables, as shown in the table:
| Inner I (a) | Boundary B (a) | Outside E (a) | |
| Inner I (a) | dim(I(a)∩I(b)) | dim(I(a)∩B(b)) | dim(I(a)∩E(b)) |
| Boundary B (a) | dim(B(a)∩I(b)) | dim(B(a)∩B(b)) | dim(B(a)∩E(b)) |
| Outside E (a) | dim(E(a)∩I(b)) | dim(E(a)∩B(b)) | dim(E(a)∩E(b)) |
The rreturn value of dim (): have-1,0,1,2. represent different implications respectively, as follows:
T: occuring simultaneously exists dim=0,1 or 2;
F: occur simultaneously and do not have dim=-1;
0: occur simultaneously to exist, but its most high-dimensional must be 0;
1: occur simultaneously to exist, but it the most high-dimensionally is necessary for 1;
2: occur simultaneously to exist, but it the most high-dimensionally is necessary for 2;
Sample tree is the space clustering structure that moving object is set up by its spatial relation, has embodied spatial coherence and the locality characteristic of spatial data between the moving object.Three-dimensional space data has the characteristic of three dimensions locality, and even the data in zone are visited, and the possibility that other data are visited in this zone so is also very big.Sample tree based on spatial relationship also will be taken into account the three dimensions locality, takes all factors into consideration the scope and the importance of spatial object, comes efficient and scientific and reasonable definite pre-scheduling object according to the space of moving object.
The sample tree has embodied the correlativity between the object between the different cluster granularity ranks sky.When design sample is set, the most important thing is the cluster function between the object of implementation space.Space clustering mainly is according to the distance relation between the spatial object spatial object to be divided into several groups automatically; And make the distance between the spatial object in the same group as much as possible little, belong to the big as far as possible a kind of method of space length between the spatial object of different groups.The multicenter characteristic between the moving object can be reflected through cluster analysis, the correlativity between the spatial object can be reflected simultaneously.The most frequently used is tree-like space index structure, and for example the R tree is set up sample tree storage organization.
Sample tree is to have other cluster structures of different grain size level by what the space length of moving object was set up.This tree has the notion of the degree of depth, and the cluster object that is on the same level has identical cluster granularity, and this is the difference of it and other cluster structures.
Describe the sample tree in detail below in conjunction with Fig. 2:
Root is the root node of sample tree among Fig. 2, and moving object 3,4,5 constitutes cluster object a, and the cluster object of object 8 is himself.Factor of influence and the range of influence of supposing object 3,4,5 are respectively R
3, R
4, R
5, V
3, V
4, V
5, for cluster object a, its factor of influence can by formula 1 calculate.Calculate from bottom to top with recursive mode, can try to achieve the factor of influence of each cluster object in the sample tree, its flow process is shown in the solid line direction among Fig. 2.
R
a＝(R
3*V
3+R
4*V
4+R
5*V
5/V
a (1)
If R
a＞R
b, explain that then the spatial object in the cluster object a range of influence is dispatched by the user continually, this zone is the focus dispatcher-controlled territory.Adopt and intersect and comprise discrimination principle; Preferentially from the focus dispatcher-controlled territory, select the pre-scheduling object; Therefore the dispatching priority of the pre-scheduling candidate target in cluster object a range of influence will be higher than other pre-scheduling candidate targets, and the filtering process of pre-scheduling object is non-moving object, pre-scheduling candidate target, pre-scheduling object.
Each step of describing that the present invention proposes in detail below in conjunction with Fig. 1 based on the three-dimensional space data self-adaptation pre-scheduling method of spatial relationship:
Step 1: adopt trigger mechanism to realize real-time monitoring to the system call situation.If system is in the intermittent phase of scheduling, the utilization factor of CPU satisfies the startup pre-scheduling procedure condition of setting simultaneously, then starts the pre-scheduling program.
Step 2: utilize the three dimensions object that has been scheduled in the buffer memory to set up the sample index tree, moving object is formed the space clustering structure by its spatial relation;
Because the three-dimensional R tree construction algorithm has taken into full account three-dimensional propinquity principle; Taken into full account three-dimensional propinquity principle; The object that three dimensions is adjacent is gathered in the identical node of R tree or the contiguous brotgher of node; And its intrinsic tree-shaped hierarchical structure possesses the ability that locks local space from global space rapidly, thereby quickens the three dimensions inquiry, only needs the calculated amount of only a few just can from three dimensions, obtain the three-dimensional space data that satisfies the given query request.Therefore, the present invention is the basis with the three-dimensional R tree spacial index, in conjunction with the area of space continuity visit principles of construction sample tree of three-dimensional space data.Each non-leaf node record space information and factor of influence R, leaf node is the sample object, non-leaf node is realized the multi-level space clustering of sample, shown in Fig. 2 sample tree construction.
Step 3: the factor of influence R of range of influence (being non-leaf node space encloses box) in the calculation sample tree, calculate from bottom to top with recursive fashion, can try to achieve the factor of influence of each range of influence in the sample tree;
Adopt the factor of influence R of range of influence in
calculation sample tree.Wherein V representes the volume of range of influence; R
iThe factor of influence of representing its child node range of influence; V
iThe volume of representing its child node range of influence.
If the bounding box of moving object is bigger, explain that then it extends extensivelyr in the space, that it and other spatial object have is crossing, comprise, the probability of neighbouring relations is also bigger; If the factor of influence of moving object is bigger, it then is described by the frequent space scheduling that carries out, to have a probability that spatial object crossing, relation of inclusion is scheduled also bigger with it; Spatial index only writes down the size of non-moving object bounding box, and its detailed space distribution information is unknown, and system can only judge the spatial relationship between it and the moving object roughly, and it is rational therefore the factor of influence of cluster object being handled as mean value.
Step 4: begin from the root node of sample tree, employing is intersected and is comprised discrimination principle, is that bounding box is retrieved with its range of influence, retrieves the prefetching object that the root node range of influence comprises;
Step 5: calculate the internal memory summation size K that prefetching object takies,
K wherein
iThe memory size that the expression prefetching object takies;
Step 6: relatively the confession pre-scheduling object of K and setting remains spendable memory headroom K
ReSize, K
Re=K
Pre-K
UseK
PreFor setting pre-scheduling spatial cache, K
UseFor taking the pre-scheduling spatial cache.If K≤K
Re, the object that retrieves is called in buffer memory as the pre-scheduling object, finish the pre-scheduling process; If K＞K
Re, to step 7;
Step 7: to descending the big or small descending sort brotgher of node of the range of influence of one deck in the sample tree by its factor of influence R;
Step 8:, choose first node for needing the screening range of influence by institute's alignment preface;
Step 9: employing is intersected and is comprised discrimination principle, is that bounding box carries out global search with the range of influence, retrieves the prefetching object that the screening range of influence comprises;
Step 10: calculate the internal memory summation size K that prefetching object takies,
K wherein
iThe memory size that the expression prefetching object takies;
Step 11: relatively the confession pre-scheduling object of K and setting remains spendable memory headroom K
ReSize.If K≤K
Re, the object that retrieves is called in buffer memory as the pre-scheduling object, by institute's alignment preface, choose the next brother node for needing the screening range of influence, to step 9; If K＞K
Re, to step 7;
Step 12: when retrieving the sample leaf nodes, the next node of choosing its father node is for needing the screening range of influence, to step 9;
Step 13: work as K
Re, finish the pre-scheduling process at=0 o'clock;
Step 14: when next operation dispatching program, stop the pre-scheduling process, discharge sample tree internal memory; Or when the utilization rate of CPU surpasses certain value; Then stop the pre-scheduling process, to guarantee normally carrying out of other operations, simultaneously; Keeping sample tree storage organization in internal memory, the multiplexing sample number of having set up carries out choosing of pre-scheduling object when the utilization rate of CPU is in order conditions for use.
Above embodiment only supplies to explain the present invention's usefulness, but not limitation of the present invention, the technician in relevant technologies field; Under the situation that does not break away from the spirit and scope of the present invention; Can also make various conversion or modification, so all technical schemes that are equal to, all fall into protection scope of the present invention.
Claims (2)
1. the three-dimensional space data self-adaptation pre-scheduling method based on spatial relationship is characterized in that, comprising:
1. adopt trigger mechanism to realize real-time monitoring, when the utilization rate of CPU surpasses certain value, then stop the pre-scheduling process the system call situation;
2. utilize the three dimensions object that has been scheduled in the buffer memory to set up the sample index tree, moving object is formed the space clustering structure by its spatial relation;
3. the factor of influence R of range of influence in the calculation sample index tree calculates with recursive fashion from bottom to top, tries to achieve the factor of influence of each range of influence in the sample index tree;
, wherein V representes the volume of range of influence, R
iThe factor of influence of representing its child node range of influence, V
iThe volume of representing its child node range of influence;
4. the root node from the sample index tree begins, and is that bounding box is retrieved with its range of influence, retrieves the prefetching object that the root node range of influence comprises;
5. calculate the internal memory summation size K that prefetching object takies,
;
6. relatively the confession pre-scheduling object of K and setting remains spendable memory headroom K
ReSize, K
Re=K
Pre-K
Use, wherein, K
PreFor setting pre-scheduling spatial cache, K
UseFor taking the pre-scheduling spatial cache; If K≤K
Re, the object that retrieves is called in buffer memory as the pre-scheduling object, finish the pre-scheduling process; If K is>K
Re, 7. to step;
The big or small descending sort brotgher of node of 7. its factor of influence R being pressed in the range of influence of following one deck in the sample index tree;
8. by institute's alignment preface, choose first node for needing the screening range of influence;
9. the range of influence is that bounding box carries out global search, retrieves the prefetching object that the screening range of influence comprises;
10. calculate the internal memory summation size K that prefetching object takies,
, K wherein
iThe memory size that the expression prefetching object takies;
Relatively the confession pre-scheduling object of K and setting remains spendable memory headroom K
ReSize, if K≤K
Re, the object that retrieves is called in buffer memory as the pre-scheduling object, judge K then
ReWhether be 0 and the node chosen whether be the non-leaf node of last bottom, if K
Re=0 or the node chosen be the non-leaf node of last bottom, then finish the pre-scheduling process, otherwise, choose the next brother node for needing the screening range of influence, arrive step 9.; If K is>K
Re, judge then whether following one deck of the node of choosing is leaf node, if leaf node is then chosen the next brother node for needing the screening range of influence, to step 9.,, then arrive step 7. if not leaf node.
2. the three-dimensional space data self-adaptation pre-scheduling method based on spatial relationship according to claim 1 is characterized in that:
Said step 2. in the sample index tree be to have other cluster structures of different grain size level by what the space length of moving object was set up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102691218A CN101916301B (en) | 2010-09-01 | 2010-09-01 | Three-dimensional spatial data adaptive pre-scheduling method based on spatial relationship |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102691218A CN101916301B (en) | 2010-09-01 | 2010-09-01 | Three-dimensional spatial data adaptive pre-scheduling method based on spatial relationship |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101916301A CN101916301A (en) | 2010-12-15 |
| CN101916301B true CN101916301B (en) | 2012-07-18 |
Family
ID=43323813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010102691218A Active CN101916301B (en) | 2010-09-01 | 2010-09-01 | Three-dimensional spatial data adaptive pre-scheduling method based on spatial relationship |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101916301B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446208B (en) * | 2011-09-02 | 2013-08-28 | 华东师范大学 | Distributed algorithm for quickly establishing massive remote sensing image pyramid |
| CN103136214B (en) * | 2011-11-24 | 2016-03-30 | 中国移动通信集团公司 | A kind of dispatching method of spatial data, system and equipment |
| CN103092696A (en) * | 2012-12-29 | 2013-05-08 | 深圳先进技术研究院 | Three-dimensional space data pre-dispatching method and system thereof |
| CN103345505B (en) * | 2013-07-03 | 2016-05-11 | 武汉大学 | A kind of spatial object topological relation determination methods based on Global Scale subdivision dough sheet |
| CN108629351B (en) * | 2017-03-15 | 2022-05-13 | 腾讯科技（北京）有限公司 | Data model processing method and device |
| GB2566280B (en) * | 2017-09-06 | 2020-07-29 | Fovo Tech Limited | A method of modifying an image on a computational device |
| CN110109917A (en) * | 2018-02-01 | 2019-08-09 | 董福田 | A kind of processing method and processing device of data |
Citations (2)
| 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 |
| CN101527053A (en) * | 2008-12-09 | 2009-09-09 | 南京大学 | Three-dimensional entity model multi-resolution representation method |
-
2010
- 2010-09-01 CN CN2010102691218A patent/CN101916301B/en active Active
Patent Citations (2)
| 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 |
| CN101527053A (en) * | 2008-12-09 | 2009-09-09 | 南京大学 | Three-dimensional entity model multi-resolution representation method |
Non-Patent Citations (2)
| Title |
|---|
| 唐桂文等.基于三维GIS的海量地形数据存储和调度的研究.《测绘科学》.2008,第33卷(第3期),110-112页,120页. * |
| 孟放等.基于LOD控制与内外存调度的大型三维点云数据绘制.《计算机辅助设计与图形学学报》.2006,第18卷(第1期),1-8页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101916301A (en) | 2010-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101916301B (en) | Three-dimensional spatial data adaptive pre-scheduling method based on spatial relationship | |
| Patel et al. | STRIPES: an efficient index for predicted trajectories | |
| Šidlauskas et al. | Trees or grids? Indexing moving objects in main memory | |
| CN102289466B (en) | K-nearest neighbor searching method based on regional coverage | |
| Chen et al. | A benchmark for evaluating moving object indexes | |
| CN103714145A (en) | Relational and Key-Value type database spatial data index method | |
| Dittrich et al. | Indexing moving objects using short-lived throwaway indexes | |
| Lian et al. | Probabilistic top-k dominating queries in uncertain databases | |
| CN104408163B (en) | A kind of data classification storage and device | |
| CN103544249B (en) | A kind of method for indexing scattered point cloud space of historic building | |
| CN110070121A (en) | A kind of quick approximate k nearest neighbor method based on tree strategy with balance K mean cluster | |
| CN101692230A (en) | Three-dimensional R tree spacial index method considering levels of detail | |
| CN103294912B (en) | A kind of facing mobile apparatus is based on the cache optimization method of prediction | |
| CN102902590B (en) | Parallel digital terrain analysis-oriented massive DEM (Digital Elevation Model) deploying and scheduling method | |
| CN104346444A (en) | Optimum site selection method based on road network reverse spatial keyword query | |
| Lettich et al. | Processing streams of spatial k-NN queries and position updates on manycore GPUs | |
| CN103136214B (en) | A kind of dispatching method of spatial data, system and equipment | |
| Qiang et al. | The KD-Tree-based nearest-neighbor search algorithm in GRID interpolation | |
| Lewis et al. | G-PICS: A framework for GPU-based spatial indexing and query processing | |
| Li et al. | A Survey of Multi-Dimensional Indexes: Past and Future Trends | |
| CN101984433B (en) | Convexity based multiple spots far neighbor querying method | |
| Chen et al. | GAM: A gpu-accelerated algorithm for maxrs queries in road networks | |
| Wu et al. | Indexing continual range queries for location-aware mobile services | |
| Carniel et al. | Spatial index structures for modern storage devices: A survey | |
| Li et al. | A new K-NN query algorithm based on the clustering and sorting of minimum bounding rectangle |
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 |