CN104732582A - Three-dimensional ocean vector field dynamic pulse surface visualization algorithm based on GPU - Google Patents
Three-dimensional ocean vector field dynamic pulse surface visualization algorithm based on GPU Download PDFInfo
- Publication number
- CN104732582A CN104732582A CN201510066037.9A CN201510066037A CN104732582A CN 104732582 A CN104732582 A CN 104732582A CN 201510066037 A CN201510066037 A CN 201510066037A CN 104732582 A CN104732582 A CN 104732582A
- Authority
- CN
- China
- Prior art keywords
- data
- gpu
- ocean
- vector field
- visual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Processing Or Creating Images (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a three-dimensional ocean vector field dynamic pulse surface visualization algorithm based on a GPU. By utilization of the ocean data characteristic, ocean three-dimensional vector field data are stored in the GPU in a three-dimensional texture mode, particles in the vector field are driven by ping-pong fashion to flow, and coarsening and refining of the particles of the three-dimensional pulse surface are carried out in combination with a geometry shader of the GPU. The even isobathic three-dimensional vector field data are sampled in the GPU for the ocean coordinate-Z data, visualization of the three-dimensional time-varying vector field data is carried out, and the algorithm can assist ocean researchers in scientific research and analysis.
Description
Technical field
The invention belongs to infotech and marine technology field, particularly relate to the three-dimensional ocean vector field dynamic arteries and veins face visualized algorithm based on GPU.
Background technology
Flow-field visualized is an important directions of visual research, and in occupation of very important status in scientific algorithm and engineering analysis, it also has very important meaning and value to the scientific research of ocean with analyzing.
In interactivity is flow-field visualized, the integration of flow line and visual from just become a kind of tool master.In time-varying flow field, because trace and arteries and veins line can reflect that the key character in flow field has become the focus of research.Trace can show the movement locus of fluid particle, and arteries and veins line can show the curve that certain fluid particle of specifying moment all that once to cross certain fixed point forms.
On the basis of streamline visualization, the visual of integral surface has also confirmed its validity.Stream interface and track plane structure be understandable, main thought is exactly that the most front line in opposite carries out integration, and can carry out the adaptive meticulous or alligatoring to line in due course.Along with the growth of line, the face of generation can't change.And the visual complexity of wanting in arteries and veins face is many, each time frame wants all data of stream field to carry out recalculating drafting, and this has had very large requirement for algorithm and computing power.
In computer assisted flow-field visualized field, several different method for visualizing also also exists analogy, but arteries and veins face visual also because its calculate complicacy, seldom by Sea Current visual apply.Arteries and veins face is visual may change shape anywhere, and each spatial point, and each moment is carrying out integral operation, each part in arteries and veins face in each time in the execution carrying out alligatoring or segmentation Strategy.This by the restriction of Computing efficiency, may may carry out a definite limitation to data on the one hand on the other hand.
Both at home and abroad arteries and veins face visual in have also been made much relevant research, and carry out the visual realization in arteries and veins face on CPU or on GPU respectively, but actually use for three-dimensional flow field visual application in ocean is not relevant and realize, and also not do not carry out the experiment of being correlated with and progress for the feature of oceanographic data, and arteries and veins face is visual has the process carrying out oceanographic data to ocean scientific research personnel and the understanding more goed deep into have huge help.
Summary of the invention
The invention provides a kind of three-dimensional ocean vector field dynamic arteries and veins face visualized algorithm based on GPU.It carries out arteries and veins face visual in trivector field, ocean be the structure carrying out visual framework completely based on GPU, and carry out the data processing in GPU for ocean z coordinate data characteristics, and this programme finally forms verification system.
For achieving the above object, the present invention adopts following technical scheme, and it comprises the following steps:
(1) the visual framework in dynamic arteries and veins face, trivector field, ocean based on GPU builds;
(2) carry out directly visual for ocean vector field Z coordinate data.
The present invention is intended to study the three-dimensional ocean vector field dynamic arteries and veins face visualized algorithm based on GPU.
Mentality of designing of the present invention is: set up the visual framework in dynamic arteries and veins face, trivector field, ocean based on GPU, three-dimensional flow field data are stored in GPU in the mode of texture, and the Geometry Shader(geometry unit tinter in GPU) in carry out the execution of refinement or alligatoring strategy, in Vertex Shader(vertex shader) in carry out the integration of particle, carried out the continuous driving of particle by the ping-pong Fashion characteristic of GPU.
In described step (1), for the feature of oceanographic data, ocean trivector field data is stored in GPU in the mode of three-D grain, and store t1 in advance, the data of t2 two time points are at GPU, the preservation being used for carrying out multiple time point vector field data in CPU also can carry out the exchange of data between GPU according to time t, meanwhile, the mode that also can be read by file in CPU carries out the exchange of data in CPU.
In described step (2), for the feature of ocean Z coordinate data, ocean variable field data is stored in three-D grain, and by repeatedly the data of different depth being carried out interpolation to same texture sampling in GPU, be sampled into the dark trivector field data such as uniform, when carrying out three-dimensional with this, become the visual of vector field data.
Innovation of the present invention is mainly reflected in:
(1) carried out for ocean z coordinate data characteristics, based on the building of the visual framework in arteries and veins face of GPU;
(2) for ocean vector field Z coordinate data, by transferring data in GPU, in GPU, carry out the process of data, process data into evenly etc. dark data, when carrying out three-dimensional with this on the visual framework basis built, become the visual of vector field data.
Beneficial effect of the present invention is: use arteries and veins line develops out in mechanics arteries and veins face to carry out becoming the visual of vector field data when ocean is three-dimensional, helps with a kind of new Sea Current method for visualizing the scientific research activity that ocean researcher carries out being correlated with.
Embodiment
Three-dimensional ocean vector field dynamic arteries and veins face visualized algorithm based on GPU of the present invention, comprises the following steps:
(1) the visual framework in dynamic arteries and veins face, trivector field, ocean based on GPU builds.
Each surperficial block is made up of four summits, and stores two values, the value of a calculated product substep, and a value calculating the refinement degree of depth represents.These data all exist in a continuous print data block in vertical array buffer memory.Because in GPU be not the size of the change data stored therein of dynamic realtime, the data therefore for particle need to apply in advance, and this has just had certain restriction to the size of representation of data and scope.Selected the data n of surperficial block by user, maximum refinement degree of depth n, also have maximum integration step m, the number of the storage surface block of buffer memory application is n × 2d × (m-d+1).Timeline is the particle in synchronization release.Concrete algorithm flow is as follows.
1. timeline release, every bar timeline is set (xi, 0 of particle ...., xi, n).
2. every bar timeline is calculated to be surperficial block, each surperficial block is made up of four summits, and separates repeated storage.
3. first store the surperficial block p0j (j=0,1,2 of n sky ... n-1).
4. each time step repeats the surperficial block of release new in streamline.
5., before integration, refinement strategy can be carried out.Can judge whether to be divided into two surperficial blocks.Set the distance that threshold value s2, a s are two adjacent particles, when the area of a surperficial block is greater than α s2, just the longest edge of surperficial block and most minor face is connected from mid point and segments.Wherein α is that a number being greater than 1 is used for controlling segmentation degree.Such two new surperficial blocks store, and delete original surperficial block.
6. each integration step, all buffer memory elements to be delivered in Geometry Shader then by method process below: n/2 element p0j (j=0 before transmitting, 1,2 ... n/2-1), p0j * 2 and p0 (j*2)+1 is put in output buffer memory simultaneously.Every n/2 remaining element shader can add 2 block elements in buffer memory, carrys out the reposition after their integration of store storage.
7. the integration of particle adopts fourth order Runge-Kutta integration, and is carried out the renewal of particle by the ping-pong Fashion of GPU.
(2) carry out directly visual for ocean vector field Z coordinate data.
Ocean Z coordinate data is the dark but interlayer uneven data spaced apart such as every layer, in this algorithm, for these type of data, data are stored in three-D grain, and pass through to same texture sampling repeatedly in GPU, each sampling is sampled the depth value of adjacent two layers of its place degree of depth respectively, and carries out linear difference according to the depth interval of desired data, becomes the visual of vector field data when carrying out three-dimensional with this.
Claims (3)
1., based on arteries and veins face, trivector field, the ocean dynamic and visual algorithm of GPU, it is characterized in that, comprise following mentality of designing:
(1) arteries and veins face, trivector field, the ocean dynamic and visual framework based on GPU builds;
(2) carry out directly visual for ocean vector field z coordinate data.
2. arteries and veins face, trivector field, the ocean dynamic and visual algorithm based on GPU according to claim 1, it is characterized in that, in described step (1), for the feature of oceanographic data, ocean trivector field data is stored in GPU in the mode of three-D grain, and store t1 in advance, the data of t2 two time points are at GPU, the preservation being used for carrying out multiple time point vector field data in CPU also can carry out the exchange of data between GPU according to time t, meanwhile, the mode that also can be read by file in CPU carries out the exchange of data in CPU.
3. arteries and veins face, trivector field, the ocean dynamic and visual algorithm based on GPU according to claim 1, it is characterized in that, in described step (2), for the feature of ocean Z coordinate data, ocean variable field data is stored in three-D grain, and by sampling to texture in GPU, and data are carried out linear interpolation, be processed into the dark trivector field data such as uniform, when carrying out three-dimensional with this, become the visual of vector field data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510066037.9A CN104732582B (en) | 2015-02-07 | 2015-02-07 | Three-dimensional ocean vector field dynamic arteries and veins face method for visualizing based on GPU |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510066037.9A CN104732582B (en) | 2015-02-07 | 2015-02-07 | Three-dimensional ocean vector field dynamic arteries and veins face method for visualizing based on GPU |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104732582A true CN104732582A (en) | 2015-06-24 |
| CN104732582B CN104732582B (en) | 2019-01-04 |
Family
ID=53456450
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510066037.9A Active CN104732582B (en) | 2015-02-07 | 2015-02-07 | Three-dimensional ocean vector field dynamic arteries and veins face method for visualizing based on GPU |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104732582B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109299343A (en) * | 2018-09-05 | 2019-02-01 | 上海彩虹鱼海洋科技股份有限公司 | A kind of dynamic and visual method and system towards multi-source global ocean big data |
| CN114638912A (en) * | 2022-03-16 | 2022-06-17 | 广州博进信息技术有限公司 | Space-time visualization and interaction method, medium and equipment for marine science vector data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103700143A (en) * | 2013-12-30 | 2014-04-02 | 四川九洲电器集团有限责任公司 | Three-dimensional dynamic marine simulation method based on GPU (Graphics Processing Unit) multi-pass drawing |
| CN103714574A (en) * | 2013-12-19 | 2014-04-09 | 浙江大学 | GPU acceleration-based sea scene modeling and real-time interactive rendering method |
| WO2014130044A1 (en) * | 2013-02-23 | 2014-08-28 | Hewlett-Packard Development Company, Lp | Three dimensional data visualization |
-
2015
- 2015-02-07 CN CN201510066037.9A patent/CN104732582B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014130044A1 (en) * | 2013-02-23 | 2014-08-28 | Hewlett-Packard Development Company, Lp | Three dimensional data visualization |
| CN103714574A (en) * | 2013-12-19 | 2014-04-09 | 浙江大学 | GPU acceleration-based sea scene modeling and real-time interactive rendering method |
| CN103700143A (en) * | 2013-12-30 | 2014-04-02 | 四川九洲电器集团有限责任公司 | Three-dimensional dynamic marine simulation method based on GPU (Graphics Processing Unit) multi-pass drawing |
Non-Patent Citations (3)
| Title |
|---|
| KAI B¨URGER 等: "Interactive Streak Surface Visualization on the GPU", 《IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS》 * |
| MATT EDMUNDS 等: "Surface-Based Flow Visualization", 《COMPUTERS & GRAPHICS》 * |
| 李久松 等: "海洋和大气数据多模式动态可视化系统的设计和实现", 《海洋科学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109299343A (en) * | 2018-09-05 | 2019-02-01 | 上海彩虹鱼海洋科技股份有限公司 | A kind of dynamic and visual method and system towards multi-source global ocean big data |
| CN114638912A (en) * | 2022-03-16 | 2022-06-17 | 广州博进信息技术有限公司 | Space-time visualization and interaction method, medium and equipment for marine science vector data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104732582B (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101937425B (en) | Matrix parallel transposition method based on GPU multi-core platform | |
| CN104317768B (en) | Matrix multiplication accelerating method for CPU+DSP (Central Processing Unit + Digital Signal Processor) heterogeneous system | |
| Hsieh et al. | Particle swarm optimisation (PSO)-based tool path planning for 5-axis flank milling accelerated by graphics processing unit (GPU) | |
| CN103310481A (en) | Point cloud reduction method based on fuzzy entropy iteration | |
| CN104851126A (en) | Three-dimensional model decomposition method and three-dimensional model decomposition device based on generalized cylinder | |
| CN104732582A (en) | Three-dimensional ocean vector field dynamic pulse surface visualization algorithm based on GPU | |
| WO2014032008A3 (en) | Method and system for generating mesh from images | |
| CN104281490B (en) | A kind of method of the supercomputing hologram based on many GPU | |
| EP3182299A3 (en) | Methods and systems for estimating the number of points in two-dimensional data | |
| WO2015143985A1 (en) | Result vector determining method and apparatus | |
| CN104657333A (en) | Dynamic two-dimensional vector field streamline visualization algorithm based on GPU (Graphics Processing Unit) | |
| CN102096940B (en) | Pre-sorting method and system for removing hidden surface of image object | |
| CN105205206B (en) | A kind of acquiring method of complex plane piece intersection section | |
| JP7410961B2 (en) | arithmetic processing unit | |
| Rahim et al. | Evaluation of adaptive subdivision method on mobile device | |
| CN202736078U (en) | Feature extraction module used for digital image processing | |
| CN102447898B (en) | Method for realizing KLT (Karhunen-Loeve Transform) by means of FPGA (Field Program Gate Array) | |
| Ang et al. | Evolutionary Computation Automated Design of Ship Hull Forms for the Industry 4.0 Era | |
| Sun et al. | Research on robot target recognition based on deep learning | |
| Barina et al. | Accelerating discrete wavelet transforms on GPUs | |
| Hsu | Single-machine scheduling with aging effects and optional maintenance activity considerations | |
| CN103338373B (en) | A kind of adjacent boundary length deriving method and device | |
| WO2015009294A3 (en) | System and method for histogram computation using a graphics processing unit | |
| CN203366317U (en) | Image wave beam forming device | |
| CN110825803B (en) | Ocean current visualization method based on parallel search set and parallel computation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |