CN102999936A - Three-dimensional streamline volume rendering algorithm based on ocean flow field data - Google Patents

Three-dimensional streamline volume rendering algorithm based on ocean flow field data Download PDF

Info

Publication number
CN102999936A
CN102999936A CN2012104651943A CN201210465194A CN102999936A CN 102999936 A CN102999936 A CN 102999936A CN 2012104651943 A CN2012104651943 A CN 2012104651943A CN 201210465194 A CN201210465194 A CN 201210465194A CN 102999936 A CN102999936 A CN 102999936A
Authority
CN
China
Prior art keywords
dimensional
texture
volume rendering
streamline
rendering algorithm
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
Application number
CN2012104651943A
Other languages
Chinese (zh)
Inventor
田丰林
马纯永
于方杰
韩勇
陈戈
吴合义
李勃
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qingdao Jingwei Blue Image Information Technology Co Ltd
BEIJING ZHONGHAI XINTU TECHNOLOGY Co Ltd
Original Assignee
Qingdao Jingwei Blue Image Information Technology Co Ltd
BEIJING ZHONGHAI XINTU TECHNOLOGY Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Qingdao Jingwei Blue Image Information Technology Co Ltd, BEIJING ZHONGHAI XINTU TECHNOLOGY Co Ltd filed Critical Qingdao Jingwei Blue Image Information Technology Co Ltd
Priority to CN2012104651943A priority Critical patent/CN102999936A/en
Publication of CN102999936A publication Critical patent/CN102999936A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Image Generation (AREA)

Abstract

The invention belongs to the ocean visualization fields and particularly relates to a three-dimensional streamline volume rendering algorithm based on real ocean flow field data. According to the three-dimensional streamline volume rendering algorithm, a line integral convolution (LIC) method based on texture is expanded to three-dimensional vector field visualization, the generated three-dimensional streamline texture image is improved by designing sparse texture line noise and matching up an LIC convolution kernel, space depth feeling is remarkably strengthened, and a Volume LIC image is generated by means of a ray cast direct volume rendering method. The three-dimensional streamline volume rendering algorithm displays the full view of an ocean flow field in an image mode, and detailed changes can be shown. Due to the fact that the texture is a continuous image, the method based on the texture can solve the problem caused by density and spacing of arrow plots and vector lines. Therefore, the three-dimensional streamline volume rendering algorithm has great advantages.

Description

A kind of three-dimensional streamline volume rendering algorithm of Sea Current data
Technical field
The invention belongs to the visual field of Yu Haiyang, be specifically related to a kind of three-dimensional streamline volume rendering algorithm of Sea Current data.
Background technology
Along with the continuous progress of hydrospace detection means, the ability that the mankind obtain oceanographic data improves constantly, and the oceanographic data amount of accumulation is also more and more huger.The visual efficient that helps to improve scientific research of seas of oceanographic data accurately.The geographical environment that the ocean is special is so that people are difficult to directly observe all sidedly ocean every aspect phenomenon and internal feature thereof.By the flow field data visualization technique, the Sea Current data can be expressed as far as possible comprehensively, intuitively the interpretation work of assisting the ocean scientific research personnel.
For two-dimentional three-dimensional flow field data visualization, have at present the visual method for drafting of multiple vector.Method based on texture has extremely important status in vector field visualization, the most frequently used in the method based on texture is exactly the method for line integral convolution, because the huge advantage of line integral convolution method, it has obtained significant progress.The people such as Stalling have introduced the redundant computation that a kind of fast line integral convolution method reduces classic method, so that Line integral convolution can be used in practice.The people such as Shen have proposed in 1997 a kind of visualization method to be carried out in astable field, astable field also can with the people such as the same Cao of displaying of stationary field on the cardiac motion that on the ICBBE in 2009 the line integral convolution method has been used in ultrasound medicine is visual.These methods have realized that the LIC of two-dimensional vector field draws, and not for the Sea Current volume data, can not comprehensively reflect the three dimensional fluid motion characteristics.Because the Sea Current data volume is large and time to time change, can not satisfy efficiently real-time characteristics, visual inefficiency based on the algorithm of CPU generating three-dimensional streamline.
Summary of the invention
The present invention proposes a kind of algorithm that accelerates based on GPU, namely based on three-dimensional line integral convolution (VolumeLIC) algorithm of Sea Current data.It is field visualized that this algorithm will be extended to trivector based on the line integral convolution method of texture, by designing sparse line Noise texture and cooperating convolution kernel to improve generating three-dimensional Streamline Texture image, significantly strengthen the spatial depth sense, and adopt ray cast direct volume drawing (Ray-Casting) method to generate the VolumeLIC image.
For achieving the above object, the present invention adopts following technical scheme, and concrete steps are:
(1) three-dimensional Sea Current data pre-service;
(2) three-dimensional LIC algorithm generating three-dimensional body texture;
(3) the Ray-Casting method is drawn the body texture.
Above-mentioned steps (1) is processed the ocean raw data format, generates a volume data that comprises Sea Current information.Step (2) selects white noise as the input texture, and each pixel value of output texture all obtains by line integral convolution.Step (3) each pixel from screen, send a ray, this ray passes the voxel matrix of 3 d data field, select several equidistant sampled points along this ray, opacity value and color value to all sampled points, adopt from back to front or method from front to back makes up color and the opacity of each sampled point, thereby calculate the color value at this pixel place on the screen.
Beneficial effect of the present invention is: this algorithm shows the overall picture of Sea Current with graphic form, and can show variations in detail, because texture is continuous image, can overcome the problem that arrow is marked on a map and line of vector density causes based on the method for texture, have larger superiority.And this algorithm realizes in GPU, and flow line is extracted and to carry out visual efficient higher.
Description of drawings
Fig. 1 LIC algorithmic descriptions;
Fig. 2 South Sea data VolumeLIC visualization result.
Embodiment
The present invention is treated to four-dimensional volume data (attribute is longitude, dimension, the degree of depth and flow field value) according to the magnanimity Dynamic and Multi dimensional feature of Sea Current data with it; And the LIC streamline follow the tracks of generated expand to three dimensions from two dimensional surface, use at last the Ray-Casting method to draw, be presented on the two dimensional surface.May further comprise the steps:
1. three-dimensional Sea Current data pre-service
The oceanographic data organizational form is equally distributed four-dimensional data volume DataSet on the longitude and latitude plane, and the form of expression is:
Value=DataSet(Lat,Lon,Dep,Time);Value∈{temp,salt,uvw};
Storage format is science data form NC file.The dimension of this data volume comprises longitude, latitude, the degree of depth and time, the tripartite upwards scalar value (UVW) in storage flow field on each data point.The rule of vertical data acquisition determine the closer to the sea level or the degree of depth more shallow, the data of collection are closeer, the degree of depth is larger, the data of collection are less.And submarine topography rises and falls and to have determined that each depth capacity on vertical of data volume is also different, so be uneven distribution on should four-dimension data volume vertical.The degree of depth on each collection point is stored among the lookup file HSet.H=HSet (Lat, Lon, Num) can search the depth value of corresponding different order on the even distribution longitude and latitude.Adopt the method for linear interpolation, go out flow speed value on each depth layer according to the consecutive point interpolation.Finally according to depth information and three-dimensional flow field volume data of NC file generated.
2. three-dimensional LIC algorithm generating three-dimensional body texture
LIC based on the correlativity of direction vector to the Noise texture low-pass filtering, final performance (demonstration) correlativity.Particularly, (be generally white noise: White Noise) as the input texture, each pixel value of output texture all obtains by line integral convolution LIC selection noise.At first obtain streamline based on this pixel along vector positive and negative direction symmetric integral, input noise value corresponding to all pixels pressed convolution kernel (Convolution Kernel) and participated in convolution on the streamline, and the result is as the pixel value (Fig. 1) of output texture.
Certain pixel P for output texture (image) obtains streamline ρ (τ) along positive and negative direction line of symmetry integration centered by P, wherein-L≤τ≤L, L for just (instead) to streamline length.Positive and negative streamline equal in length can keep symmetrical correlativity; Especially, respective pixel P when τ gets 0.The pixel value T ' of P (ρ (0)) is the convolution of streamline each point Noise texture value:
T ′ ( ρ ( 0 ) ) = ∫ - L L k ( τ ) T ( ρ ( τ ) ) dτ ∫ - L L k ( τ ) dτ
Discrete form is:
T ′ ( P ) = Σ i = 0 N T ( p i ) W i Σ i = 0 N W i
Wherein T is the input noise texture, and T (ρ (τ)) is each point Noise texture value on the streamline.K (τ) is convolution kernel, the correlativity of expression streamline each point and P.Pi is N discrete point on the streamline (comprising positive and negative two halves), Wi be pi to the contribution of P, i.e. the area that between a pi-1 and pi, covers of convolution kernel K (τ).Symmetric integral obtains streamline in the other direction, and input noise value corresponding to all pixels pressed convolution kernel (Convolution Kernel) and participated in convolution on the streamline, and the result is as the pixel value of output texture.
The LIC method can be extended to trivector field LIC3D.LIC3D specifically along 3D grain direction low-pass filtering, exports body texture (Volumetric Texture) for 3D noise Alongvector at last.Because this step is that the convection cell data extract on the space three-dimensional direction, can comprehensively reflect the three dimensional fluid motion characteristics.
3.Ray-Casting method is drawn the body texture
Ray cast volume rendering algorithm based on GPU mainly is comprised of three parts:
(1) determining of light terminal point: the texture coordinate component on the coordinate components on each summit of bounding box and this summit and color component correspondent equal, behind the graphic hardware rasterisation, the coordinate figure of any point equates with color value on the bounding box surface.Reject the bounding box front surface and can draw the rear surface.Because bounding box is protruding, each pixel place can corresponding to two or more bounding box front surfaces, therefore not need depth test on the viewport.Draw the rear surface of volume data bounding box, and preserve drawing result to 2D texture backMap, thereby obtain the light terminal point;
(2) calculating of radiation direction: again draw this bounding box, during drafting, make the color value on four summits of each face of bounding box equate with the coordinate figure on summit, reject the rear surface of bounding box, do not need equally depth test.Drawing result is the coordinate information of bounding box front surface, namely the starting point of throw light (start).Enable sheet unit handling procedure, according to the light incidence point be kept among the backMap light endpoint calculation light to.
(3) ray cast: try to achieve the length (being used for judging whether light stops) of the radiation direction vector dir of sheet unit, and with dir normalization, each component multiply by step-length step and just obtains sampling step length δ.Obtain thus equations of light ray:
r(n)=start+nδ;
Wherein r (n) is the coordinate of sampled point, and n is the sampling step number.
Light is from incidence point, whenever take a step forward, just obtain a new sampled point r (n), with the three-dimensional location coordinates sampling volume data texture of this sampled point, obtain the density value of this point, search color value and the opacity that transition function just obtains this sampled point with this density value again.Then adopt formula according to from front to back synthesis mode color and opacity to be accumulated.And calculate the coordinate of next sampled point r (n+1), the length of ray vectors also correspondingly reduces step.The condition that light stops: when the length of ray vectors less than or equal to 0 or opacity when running up to certain value (such as 0.999) and satisfying this condition, light stops.Because this step is accelerated to realize in GPU, can be satisfied the real-time visual of ocean current volume data.
4. experimental result
Intel PIV3.0GHz, NVIDIA GeForce GTS450 video card, the PC of 1.00GB RAM are adopted in experiment.Operating system is Windows XP, and developing software is Visual Studio2008 and OpenGL.The raw data that experiment is used is that the NC file is stored three-dimensional ocean current data, and data evenly distribute on longitude and latitude, and grid number is 473*409*100 (longitude, latitude and the degree of depth).Fig. 2 is rendering effect figure.

Claims (2)

1. the three-dimensional streamline volume rendering algorithm of Sea Current data specifically comprises following basic step:
(1) three-dimensional ocean current flow field data pre-service;
(2) by three-dimensional L1C algorithm generating three-dimensional body texture;
(3) adopt Ray Casting method to draw to the body texture.
2. the three-dimensional streamline volume rendering algorithm of Sea Current data according to claim 1 is characterized in that: according to the magnanimity Dynamic and Multi dimensional feature of Sea Current data, it is treated to four-dimensional volume data (attribute is longitude, dimension, the degree of depth and flow field value); And the LIC streamline follow the tracks of generated expand to three dimensions from two dimensional surface, use at last Ray Casting method to draw, be presented on the two dimensional surface.
CN2012104651943A 2012-11-19 2012-11-19 Three-dimensional streamline volume rendering algorithm based on ocean flow field data Pending CN102999936A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2012104651943A CN102999936A (en) 2012-11-19 2012-11-19 Three-dimensional streamline volume rendering algorithm based on ocean flow field data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2012104651943A CN102999936A (en) 2012-11-19 2012-11-19 Three-dimensional streamline volume rendering algorithm based on ocean flow field data

Publications (1)

Publication Number Publication Date
CN102999936A true CN102999936A (en) 2013-03-27

Family

ID=47928465

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2012104651943A Pending CN102999936A (en) 2012-11-19 2012-11-19 Three-dimensional streamline volume rendering algorithm based on ocean flow field data

Country Status (1)

Country Link
CN (1) CN102999936A (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103606176A (en) * 2013-11-25 2014-02-26 北京理工大学 Two-dimension flow field directionality sparse texture synthesis method with reinforced feature
CN103761761A (en) * 2014-01-21 2014-04-30 中国科学院遥感与数字地球研究所 Marine scalar field volume rendering method based on earth sphere model
CN104658027A (en) * 2015-02-11 2015-05-27 中国海洋大学 Three-dimensional streamline dynamic visualization algorithm facing irregular ocean flow field data
CN104867186A (en) * 2015-04-29 2015-08-26 中国海洋大学 GPU-based interactive ocean three-dimensional flow field dynamic visual algorithm
CN107871337A (en) * 2016-09-26 2018-04-03 中国空气动力研究与发展中心高速空气动力研究所 A kind of method for visualizing of supersonic speed two-dimensional flow field data
CN108133504A (en) * 2018-01-16 2018-06-08 北京理工大学 A kind of three-dimensional flow field multivariate data method for visualizing based on polyhedron pipeline
CN108665524A (en) * 2018-04-28 2018-10-16 武汉大学 A kind of a wide range of discrete flow field volume rendering method based on GPU
CN110211207A (en) * 2019-06-14 2019-09-06 北京理工大学 A kind of three-dimensional flow field method for visualizing to be added up based on streamline length
CN110232735A (en) * 2019-06-14 2019-09-13 北京理工大学 A kind of three-dimensional flow field method for visualizing based on streamline counter mechanism
CN113436305A (en) * 2021-06-23 2021-09-24 北京理工大学 High-quality geometric modeling method guided by direct volume rendering
CN113487661A (en) * 2021-06-15 2021-10-08 北京道达天际科技有限公司 Visualization method of vector field data

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
BERNHARD KAINZ等: "In vivo interactive visualization of four-dimensional blood flow patterns—Realtime assessment of volumetric phase contrast MRI", 《THE VISUAL COMPUTER 》 *
CHEN GE等: "Design and implementation of a 3D ocean virtual reality and visualization engine", 《JOURNAL OF OCEAN UNIVERSITY OF CHINA》 *
颜廷华: "基于物理特征的海洋流场可视化技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103606176A (en) * 2013-11-25 2014-02-26 北京理工大学 Two-dimension flow field directionality sparse texture synthesis method with reinforced feature
CN103606176B (en) * 2013-11-25 2016-03-30 北京理工大学 A kind of two-dimensional flow field directivity sparse texture synthetic method of Enhanced feature
CN103761761A (en) * 2014-01-21 2014-04-30 中国科学院遥感与数字地球研究所 Marine scalar field volume rendering method based on earth sphere model
CN104658027A (en) * 2015-02-11 2015-05-27 中国海洋大学 Three-dimensional streamline dynamic visualization algorithm facing irregular ocean flow field data
CN104867186A (en) * 2015-04-29 2015-08-26 中国海洋大学 GPU-based interactive ocean three-dimensional flow field dynamic visual algorithm
CN104867186B (en) * 2015-04-29 2018-01-19 中国海洋大学 Ocean three-dimensional flow field dynamic and visual algorithm is interacted based on GPU
CN107871337B (en) * 2016-09-26 2020-12-08 中国空气动力研究与发展中心高速空气动力研究所 Visualization method of supersonic two-dimensional flow field data
CN107871337A (en) * 2016-09-26 2018-04-03 中国空气动力研究与发展中心高速空气动力研究所 A kind of method for visualizing of supersonic speed two-dimensional flow field data
CN108133504A (en) * 2018-01-16 2018-06-08 北京理工大学 A kind of three-dimensional flow field multivariate data method for visualizing based on polyhedron pipeline
CN108133504B (en) * 2018-01-16 2021-03-23 北京理工大学 Three-dimensional flow field multivariable data visualization method based on polyhedral pipeline
CN108665524A (en) * 2018-04-28 2018-10-16 武汉大学 A kind of a wide range of discrete flow field volume rendering method based on GPU
CN108665524B (en) * 2018-04-28 2021-09-24 武汉大学 Large-range discrete flow field volume rendering method based on GPU
CN110211207A (en) * 2019-06-14 2019-09-06 北京理工大学 A kind of three-dimensional flow field method for visualizing to be added up based on streamline length
CN110232735A (en) * 2019-06-14 2019-09-13 北京理工大学 A kind of three-dimensional flow field method for visualizing based on streamline counter mechanism
CN110232735B (en) * 2019-06-14 2020-10-02 北京理工大学 Three-dimensional flow field visualization method based on streamline counting mechanism
CN113487661A (en) * 2021-06-15 2021-10-08 北京道达天际科技有限公司 Visualization method of vector field data
CN113487661B (en) * 2021-06-15 2022-12-23 北京道达天际科技股份有限公司 Visualization method, device and equipment of vector field data and readable storage medium
CN113436305A (en) * 2021-06-23 2021-09-24 北京理工大学 High-quality geometric modeling method guided by direct volume rendering

Similar Documents

Publication Publication Date Title
CN102999936A (en) Three-dimensional streamline volume rendering algorithm based on ocean flow field data
CN104574263A (en) Quick three-dimensional ultrasonic reconstruction and display method on basis of GPU (graphics processing unit)
CN102332179B (en) Three-dimensional model data simplification and progressive transmission methods and devices
Hong et al. A local stereo matching algorithm based on weighted guided image filtering for improving the generation of depth range images
CN103530907B (en) Complicated three-dimensional model drawing method based on images
CN106355552B (en) A kind of depth map top sampling method based on virtual viewpoint rendering quality
CN105825471A (en) Unity-3D-based three-dimensional surface reconstruction and rendering method
Tatarchuk et al. Real-time isosurface extraction using the GPU programmable geometry pipeline
Litvinov et al. Incremental solid modeling from sparse and omnidirectional structure-from-motion data
CN108874932B (en) Ocean water sound field three-dimensional visualization method based on improved ray projection algorithm
Bartz et al. Voxels versus polygons: A comparative approach for volume graphics
CN113281779B (en) 3D object rapid detection method, device, equipment and medium
Lu et al. Real-time stereo matching: A cross-based local approach
Gu et al. Ue4-nerf: Neural radiance field for real-time rendering of large-scale scene
CN103945206B (en) A kind of stereo-picture synthesis system compared based on similar frame
EP4287134A1 (en) Method and system for generating polygon meshes approximating surfaces using root-finding and iteration for mesh vertex positions
CN103745495A (en) Medical volume data based volume rendering method
CN110738719A (en) Web3D model rendering method based on visual range hierarchical optimization
CN105791798A (en) Method and device for converting 4K multi-viewpoint 3D video in real time based on GPU (Graphics Processing Unit)
Kolhatkar et al. Real-time virtual viewpoint generation on the GPU for scene navigation
Çalışkan et al. Overview of Computer Graphics and algorithms
Varadarajan et al. Surface reconstruction for RGB-D data using real-time depth propagation
Lim et al. Reliable space leaping using distance template
Corrigan et al. Computing and rendering implicit surfaces composed of radial basis functions on the GPU
Geetha Kiran et al. Automatic 3D view generation from a Single 2D Image for both Indoor and Outdoor Scenes

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20130327