CN103914881A - Three-dimensional model typification algorithm based on minimum spanning trees - Google Patents
Three-dimensional model typification algorithm based on minimum spanning trees Download PDFInfo
- Publication number
- CN103914881A CN103914881A CN201310007034.9A CN201310007034A CN103914881A CN 103914881 A CN103914881 A CN 103914881A CN 201310007034 A CN201310007034 A CN 201310007034A CN 103914881 A CN103914881 A CN 103914881A
- Authority
- CN
- China
- Prior art keywords
- model
- dimensional
- typification
- dimensional model
- chain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
Disclosed is a three-dimensional model typification algorithm based on minimum spanning trees. According to the method, the minimum spanning tree is constructed in each three-dimensional model group based on the distance between models; each minimum spanning tree is segmented to obtain a corresponding three-dimensional model linear arranging structure, namely a model chain; typification is conducted on each model chain, wherein during typification, one node is deleted and other nodes are moved to enable the distance ratio between models to be identical with that of the original model. To minimize the visual difference, the node which enables the moving distance of other nodes to be smallest after being deleted is deleted each time. The three-dimensional model is simplified through the iterative deletion.
Description
One, technical field
The present invention, towards the visual application of online three-dimensional model, applies as the field such as three-dimensional map, three-dimensional commodity displaying for the three-dimensional model based on wireless especially, proposes a kind of comprehensive short-cut method based on typification algorithm.The method passes through to reduce the quantity of model in multiple three-dimensional models or three-dimensional model group, in maximizing vision similarity, and deletion model, thus the visual efficiency of entirety of raising three-dimensional scenic reduces model complexity.
Two, background technology
Online three-dimensional visualization technique has been widely used in the fields such as game, navigation, ecommerce at present.Along with the development of smart mobile phone and mobile computing, increasing application need to and represent three-dimensional model by wireless way for transmitting.But three-dimensional modeling data amount is large, data complexity is high (comprising geometric data, data texturing and semantic data), is therefore necessary three-dimensional model to simplify, and improves its visual efficiency.Typification algorithm is exactly one of three-dimensional model simplifying method.When typification operates in the topological similarity of reserving model entirety, reduce the quantity of model in scene by modes such as deletion, movements, improve visual efficiency.The operation of typification is at present mainly used in the comprehensive simplification of two-dimensional map, and this patent expands to three-dimensional model by typification algorithm, thereby realizes the comprehensive simplification of three-dimensional scenic.The present invention is mainly for having linearly aligned three-dimensional model, such as, commodity in the buildings in three-dimensional city and three-dimensional shop etc.
Three, summary of the invention
The distance the present invention is based between model is obtained three-dimensional model minimum spanning tree, again minimum spanning tree is carried out to cut operator, obtain linear pattern chain of model wherein, the linear pattern chain of model generating is implemented to typification operation, in reducing model quantity, keep the similarity of itself and master mould.
According to the Gestalt theory in psychology, human eye is for being arranged in a straight line or the object of near linear has more responsive understanding, if therefore can keep the linear feature of three-dimensional model in comprehensive simplification, deleting wherein individual element affects little for the similarity of block mold.The present invention, just based on this principle, utilizes minimum spanning tree to detect the linear distribution in three-dimensional model group, thereby the model generalization of realizing based on typification operation is simplified.
Three-dimensional model integrated approach based on minimum spanning tree proposed by the invention, its particular content is as follows:
1. minimum spanning tree structure
The key of structure minimum spanning tree is the distance between model in three-dimensional scenic.This patent is first by the distance between any two independent model of structure, thus the complete graph between model in structure three-dimensional scenic, just can use existing minimal spanning tree algorithm as Prim algorithm on this figure.Therefore this patent relates generally to and how to construct model spacing in three-dimensional scenic.
Described in claims 2, the model spacing in this patent calculates by following three steps: the first, and model spacing is three-dimensional distance, takes into full account that model is in three-dimensional distribution; The second, model spacing is the minimum distance between all geometric properties of model, takes into full account the size of model itself; The 3rd, model spacing is considered the semantic distance of model, if cut apart by grid (as the network of rivers, road network) between model, model spacing is infinitely great.Fig. 1 has provided a three-dimensional city model, and (a), after road network is cut apart, ((Fig. 1 c) for the minimum spanning tree that Fig. 1 b) obtains for Fig. 1.
2. linear pattern three-dimensional model chain generates
Obtain after three-dimensional model minimum spanning tree, need to extract linear structure wherein as further comprehensive basis of simplifying.The number of degrees (quantity on the limit of containing) of this patent use node, are implemented to cut apart to minimum spanning tree as evaluation criterion with corner angle.As shown in Figure 2, first tectonic model minimum spanning tree (Fig. 2 a), detect the node that wherein number of degrees are greater than 2, and it is cut apart to (Figure 21 b), then the node that turning is greater than to 45 degree cuts apart that (Fig. 2 c) again, generate linear pattern chain of model, finally chain of model is implemented to typification algorithm, reach thereby delete wherein part of nodes the object that model generalization is simplified.
3. retain the linear pattern three-dimensional model typification method of visual signature
Typification algorithm mainly refers in ensureing distortion, by reducing the quantity of model in three-dimensional scenic, reduces the complexity of three-dimensional model, thereby improves the method for visual efficiency.This patent, by the typification operation to linear pattern chain of model, when reserving model linear topology distributes, has reduced model quantity.Particularly, this patent first in preference pattern chain among the nearest node of non-other models of fringe node middle distance the model of minimum bounding box minimum deleted; Secondly move remaining intermediate node (non-fringe node) and make the ratio of distances constant between mobile rear model identical with master mould along master mould chain locus, thereby reach the object that minimizes topological variation; Last adjustment model angle makes to have identical angle with master mould chain.
As shown in Figure 3, (node (C) of Fig. 3 a), first selecting to have among the nearest non-fringe node (C, D) of other nodes of its middle distance minimum bounding box is node to be deleted to simplify former three-dimensional model chain; After deletion of node C, move all the other intermediate nodes (B, D) to reposition (B ', D ') along master mould chain.For making reposition keep master mould ratio, the first transport coefficient e=AE/ (AE-CD) of computation model, reposition is calculated as follows: AB '=AB+BB '=AB*e, B ' D '=B ' C+CD '=BC*e and D ' E=D ' D+DE=DE*e; Finally, owing to moving to new position, need to adjust angle, for example B ' need to adjust angle BAB ', and (Fig. 3 b).Repeat the process of above-mentioned selection, deletion and adjustment, just can control the degree of model simplification, concrete adjustment amount can be modified according to real needs and application.
The present invention proposes a kind of comprehensive shortcut calculation of linear pattern three-dimensional model based on minimum spanning tree, this algorithm comprises that minimum spanning tree produces, linear pattern chain of model generates and three major parts of three-dimensional model typification algorithm.This algorithm, in reserving model similarity, has reduced the quantity of model in three-dimensional scenic, reduces volume of transmitted data thereby improve display efficiency, can be used in the multiple fields such as three-dimensional city, ecommerce displaying and game, has wider adaptability.
Four, brief description of the drawings
Fig. 1 minimum spanning tree structure example
Fig. 2. linear pattern three-dimensional model chain generates and typification schematic diagram
Fig. 3. the typification operation based on three-dimensional model chain
Five, embodiment
The present invention can implement in various application occasions, and flexible configurable simplified model is provided.Concrete steps are as follows:
Step 1 is set model spacing in three-dimensional scenic, should realize the calculating that comprises three-dimensional distance and semantic distance;
Step 2, according to model spacing, generates complete graph, and utilizes on this basis Prim algorithm to calculate the minimum spanning tree of model;
Step 3 is cut apart obtained minimum spanning tree, obtains linear pattern chain of model;
Step 4 is carried out typification operation to each linear pattern chain of model, thereby reduces model quantity, generates and the rear model of output integrated simplification;
Claims (4)
1. the three-dimensional model typification algorithm based on minimum spanning tree, the feature of its method implementation procedure is specific as follows: based on the distance between three-dimensional model, minimum spanning tree between production model; By to the cutting apart of minimum spanning tree, obtain the three-dimensional model chain with linear distribution feature; On three-dimensional model chain basis, realize typification operation, reduce three-dimensional model quantity and improve visual efficiency.
2. according to the described three-dimensional model typification algorithm based on minimum spanning tree of claims 1, the generating mode feature of its model minimum spanning tree is: first model spacing is three-dimensional distance, takes into full account that model is in three-dimensional distribution; Secondly, model spacing is the minimum distance between all geometric properties of model, takes into full account the size of model itself; Finally, model spacing is considered the semantic distance of model, if cut apart by grid (as the network of rivers, road network) between model, model spacing is infinitely great.On this distance definition basis, utilize minimal spanning tree algorithm, obtain the minimum spanning tree of model.The node of this tree is three-dimensional model, and the weight on limit is model spacing, between model, by mesh segmentation, deletes limit therebetween.
3. according to the described linear pattern chain of model of claims 1 generation method, (have 3 be connected with the top) node that it is characterized in that exceeding 2 by decomposing the number of degrees, the chain type of obtaining three-dimensional model distributes, again by detecting the wide-angle flex point (exceeding 45%) in chain, chain structure is decomposed into linear pattern three-dimensional model chain, and the linear pattern in this patent is near linear.
4. according to the described three-dimensional model chain of claims 1 typification algorithm, it is characterized in that first two nearest models of preference pattern chain middle distance, wherein the model of minimum bounding box (Minimum Bounding Box) is deleted, and mobile residue model makes distance ratio therebetween identical with master mould.By the typification operation of iteration, reduce model quantity from realizing, improve the three-dimensional model of visual efficiency and comprehensively simplify.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310007034.9A CN103914881A (en) | 2013-01-09 | 2013-01-09 | Three-dimensional model typification algorithm based on minimum spanning trees |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310007034.9A CN103914881A (en) | 2013-01-09 | 2013-01-09 | Three-dimensional model typification algorithm based on minimum spanning trees |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103914881A true CN103914881A (en) | 2014-07-09 |
Family
ID=51040534
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310007034.9A Pending CN103914881A (en) | 2013-01-09 | 2013-01-09 | Three-dimensional model typification algorithm based on minimum spanning trees |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103914881A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101158966A (en) * | 2007-11-01 | 2008-04-09 | 北京航空航天大学 | City environment quantized data organization method based on vector and lattice mixed representing |
| GB2457215A (en) * | 2007-03-07 | 2009-08-12 | Nikolaos Kokkas | Automatic 3D Modelling |
| CN102308320A (en) * | 2009-02-06 | 2012-01-04 | 香港科技大学 | Generating three-dimensional models from images |
-
2013
- 2013-01-09 CN CN201310007034.9A patent/CN103914881A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2457215A (en) * | 2007-03-07 | 2009-08-12 | Nikolaos Kokkas | Automatic 3D Modelling |
| CN101158966A (en) * | 2007-11-01 | 2008-04-09 | 北京航空航天大学 | City environment quantized data organization method based on vector and lattice mixed representing |
| CN102308320A (en) * | 2009-02-06 | 2012-01-04 | 香港科技大学 | Generating three-dimensional models from images |
Non-Patent Citations (3)
| Title |
|---|
| BO MAO 等: "A Dynamic Typification Method of 3D City Models using Minimum Spanning Tree", 《6TH INTERNATIONAL CONFERENCE ON GEOGRAPHIC INFORMATION SCIENCE》 * |
| BO MAO 等: "Detection and typification of linear structures for dynamic visualization of 3D city models", 《COMPUTERS, ENVIRONMENT AND URBAN SYSTEMS》 * |
| BO MAO 等: "Detection and typification of linear structures for dynamic visualization of 3D city models", 《COMPUTERS, ENVIRONMENT AND URBAN SYSTEMS》, vol. 36, no. 3, 31 May 2012 (2012-05-31), XP028909486, DOI: doi:10.1016/j.compenvurbsys.2011.10.001 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107862738B (en) | One kind carrying out doors structure three-dimensional rebuilding method based on mobile laser measurement point cloud | |
| Génevaux et al. | Terrain generation using procedural models based on hydrology | |
| CN102629391A (en) | Three-dimensional space structure graph cutting and slicing method based on digital graph medium | |
| CN103714577B (en) | Three-dimensional model simplification method suitable for model with textures | |
| CN109377561A (en) | A kind of digital-to-analogue surface mesh generation method based on conformal geometry | |
| CN106934111A (en) | A kind of engineering tri-dimensional entity modelling method and its model building device based on terrain data | |
| CN104751505A (en) | Three-dimensional scene rendering algorithm based on LOD (Levels of Detail) model and quadtree level structure | |
| CN101303774B (en) | Method for generating quadrangle finite element grid based on three-dimensional solid model | |
| CN105225272B (en) | A kind of tri-dimensional entity modelling method based on the reconstruct of more contour line triangulation networks | |
| CN105279793A (en) | Modeling method and system based on DEM real three-dimensional map and greenway | |
| CN104778744A (en) | Establishing technique of visualized scenes of large-scale three-dimension forests based on Lidar data | |
| CN103606191A (en) | Method for establishing complex geologic model quickly | |
| CN107358640A (en) | A kind of landform of hill shading target area and the method and device of atural object | |
| CN104317886A (en) | Method for retrieving and selecting neighbor conditional data points in grid node interpolation under fault constraint | |
| CN112925865A (en) | 3D WebGIS prediction analysis method and system for mining area ground surface movement deformation | |
| CN104268941A (en) | Hot spot forming method and device for simulated three-dimensional map | |
| Zhang et al. | A new kmeans clustering algorithm for point cloud | |
| CN104392489A (en) | Topology-considered three-dimensional geospatial entity group hidden line removable method | |
| CN104715507B (en) | A kind of three-dimensional geographical entity method for auto constructing based on patch | |
| CN102646286A (en) | Digital graph medium simulation method with three-dimensional space structure | |
| CN115983162B (en) | Drainage basin modeling method based on digital twinning | |
| CN107767458B (en) | Irregular triangulation network curved surface geometric topology consistency analysis method and system | |
| CN103793937A (en) | Density clustering-based three-dimensional geometrical model simplification method and device | |
| CN103914881A (en) | Three-dimensional model typification algorithm based on minimum spanning trees | |
| CN105678840A (en) | Rapid generation custom road implement method on the basis of three-dimensional terrain software system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140709 |
|
| WD01 | Invention patent application deemed withdrawn after publication |