CN104199094A - Light-shade processing method for seismic data - Google Patents

Abstract

The invention discloses a light-shade processing method for seismic data. The light-shade processing method for seismic data comprises the following steps of: pre-processing the seismic data, inputting a scene, carrying out interpolation calculation and then obtaining a normal value, and carrying out illumination calculation, a model selection and image rendering to realize display for the seismic data. The light-shade processing method for the seismic data disclosed by the invention has the following beneficial effects that: according to the own characteristics of volume data, big-data calculation for the seismic data can be supported by virtue of multi-resolution voxel management and GPU-based acceleration operation; in addition, a target reliability conception is introduced to carry out interestingness distinguishing for the volume data, thus a target which cannot be accurately described can be modelled; a user semi-automatically inputs an interestingness transparency value, thus a purpose of local illumination enhancement can be realized.

Description

A kind of shading method of geological data

Technical field

The invention belongs to the field visualized field of three-dimensional data, relate to Volume Rendering Techniques, be specifically related to a kind of shading method of geological data.

Background technology

Current graphics and image processing techniques reach its maturity, and Volume Rendering Techniques is most important as the field visualized field of three-dimensional data, a development in recent years technology the most rapidly.The illumination of volume drawing is the key content in volume drawing research.Optical illumination calculates for the expression effect of drawing scene extremely important, and the quality of scene effect performance depends on the accurate application of illumination model to a great extent.1975, phong proposed an empirical model that is used for gauging surface mirror-reflection luminance brightness, i.e. phong model, and from phong illumination model first Application in 1988, since volume drawing, it is widely used in various visualization tools.

Nearly many researchers of more than ten years improve model.Within 1977, Blinn calculates and is improved the reflected light of phong illumination model, makes computational complexity be able to relative reduction.Nineteen eighty-two Cook and Torrance proposed one based on physioptial surface reflection model on cook-Torrance model, make in model catoptrical position and distribution and actual conditions very approaching, thereby there is good texture with the figure that it is drawn, but Cook-Torrance model does not contact with object foundation around, to diffuse reflection undertreatment.Calendar year 2001, Kniss etc. propose half-angle section, and this method based on section is carried out secondary rendering for each section, once from observer's angle once from light source direction, this system allows the light intensity that arrives this voxel on each voxel to have decay, has realized high-level efficiency hatching effect.A lot of models are limited to the processing power of CPU and GPU in realization, although improved effect, do not have actual application value.Along with the raising of GPU computing power, Yunpeng Zou in 2010 on phong model basis, improve reflection model and be applied in medical image visual on, the model after improvement can meet the requirement of real-time rendering.Within 2011, Michael Breu β has inquired into the contribution of specular reflectance model to the figure sense of reality on the basis of phong model.Florian Lindemann in 2011 and Timo Ropinski are to illumination model---the phong model occurring in recent years, half-angle dicing method and global illumination model are summed up, by thereby modular concept, picture quality, time loss etc. are contrasted and find out the contact developing between model, for possibility has been created in the proposition of brand-new visual illumination model.Shaohao Wang in 2013 is applied to 3D cartoon by phong model and plays up, and is generating on the basis of a cartoon outward appearance, and improved cartoon is played up lighting model and realized a unique effect.

In geological data is processed, researchist often can run into a large amount of 3d space discrete bodies data.Compare with the data of other field, geological data has the large and high feature of data discrete degree of data volume, three-dimensional geological survey data often up to tens GB to hundreds of GB, inner structure comprises sand body, river course etc., data are discontinuous, between voxel and voxel, relevance is little.When people directly use existing illumination model, often can not highlight Useful Information in data, DeGrain and specific aim are not strong.

Summary of the invention

Technical matters to be solved by this invention is be shown as the problem of aberration and propose a kind of shading method of geological data for the 3D seismic data existing in prior art.

The technical scheme that the present invention solves its technical matters employing is: a kind of shading method of geological data, specifically comprises the steps:

S1. geological data is carried out to pre-service;

S2. input scene;

S3. based on GPU, carry out interpolation calculation, obtain following formula:

$\stackrel{\→}{N}\left(x_i\right)\≈\frac{\▿f\left(\stackrel{\→}{x_i}\right)}{\left|\right|\▿f\left(\stackrel{\→}{x_i}\right)\left|\right|}---\left(1\right)$

$\begin{array}{c}\▿f\left(\stackrel{\→}{x_i}\right)\≈[f(x_{i-1},y_j,z_k)-{\mathrm{fix}}_{i+k}{y}_j,z_k),\\ f(x_i,y_{j-1},z_k)-f(x_i,y_{j+1},z_k),\\ f(x_j,y_j,z_{k-1})-f(x_i,y_j,z_{k+1})]\end{array}---\left(2\right)$

Wherein for the gradient vector of volume data, be this unit normal vector, i, j, k are three direction vector of unit length of x, y, z, calculate the gradient of all voxels according to (2) formula, and with the normal direction value of this approximate representation voxel place face;

S4. carry out illumination calculation;

S5. increase α preference pattern and distinguish demonstration, carry out final rendering image and realize geological data 3-D display, wherein α is illumination transparency parameter.

Further, in described step S3, calculating voxel gradient adopts central difference algorithm to realize.

Further, in described step S3, based on GPU, carrying out interpolation calculation is, based on GPU, interpolation calculation is carried out in six orientation, front and back up and down of current process points.

Further, carry out illumination calculation and adopt phong illumination model to realize in described step S4, specific algorithm is:

$I_p=K_a{i}_a+{\mathrm{\Σ}}_{m\∈\mathrm{lights}}(K_d(\stackrel{\→}{L_m}\·\stackrel{\→}{N})i_{m,d}+K_s{(\stackrel{\→}{R_m}\·\stackrel{\→}{V})}^{\mathrm{\α}}{i}_{m,s})---\left(3\right)$

I wherein _pfor the intensity of illumination of the upper point in each surface, i _sand i _drespectively mirror-reflection light intensity and the diffuse reflection light intensity of each light source, i _arepresent environmental light intensity, m represents m light source, i _{m, d}represent the diffuse reflection light intensity of m light source, i _{m, s}the mirror-reflection light intensity that represents m light source, Ks is specularity factor, the reflectivity of reflection incident light, Kd is the Diffusive intensity that diffuse-reflection factor represents incident light, Ka represents diffuse reflection parameter, represent that point is to the direction of surface point, representation line vector, the reflected light of m light source, the direction vector from point to observer, when using outside direction light source, vector be a normal vector, with point location-independent, in addition about normal vector there is equation:

${\stackrel{\→}{R}}_m=2(\stackrel{\→}{L_m}\·\stackrel{\→}{N})\stackrel{\→}{N}-\stackrel{\→}{L_m}---\left(4\right)$

Beneficial effect of the present invention: the present invention is according to the feature of volume data self, utilize the management of multiresolution voxel and the acceleration computing based on GPU can support the computing of the large data of geological data, introduce in addition target confidence level concept volume data is carried out to interest-degree differentiation, can carry out modeling raising 3D seismic data to the target that cannot accurately describe and be shown as image quality, user is by semi-automatic input interest-degree transparence value in addition, can realize local light according to the object strengthening, further research and analysis data characteristics, without a large amount of manual interventions, and the present invention is to improving the efficiency of volume drawing, help user to understand rapidly and accurately earthquake visualized data and also there is vital role.

Figure of description

Fig. 1 is the algorithm block diagram of shading method of a kind of geological data of the embodiment of the present invention.

Fig. 2 is for adopting the phong modelling effect figure of classic method.

Fig. 3 is the color table of shading method of a kind of geological data of the embodiment of the present invention.

Fig. 4 is that some river course, α=0.1 effect data of shading method for a kind of geological data of the embodiment of the present invention shows.

Fig. 5 is that some river course, α=0.9 effect data of shading method for a kind of geological data of the embodiment of the present invention shows.

Fig. 6 is that some river course, the α=0.1 part effect of shading method for a kind of geological data of the embodiment of the present invention shows.

Fig. 7 is that some river course, the α=0.9 part effect of shading method for a kind of geological data of the embodiment of the present invention shows.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.

The algorithm block diagram of shading method that is illustrated in figure 1 a kind of geological data of the embodiment of the present invention, specifically comprises the following steps:

S1. geological data is carried out to pre-service;

S2. input scene;

S3. because phong illumination model is to be based upon on the light reflection principle of body surface, if therefore directly phong illumination model is applied to volume drawing, need to determine the visible face of object and the normal direction of face, for fear of this complicated intermediate steps, accelerate the speed of volume drawing, conventionally adopt the gradient vector of volume data carry out approximate representation x _ithe normal vector of the voxel place face of point, adopts central difference algorithm to calculate voxel gradient, and interpolation is carried out in six orientation, front and back up and down to current process points in GPU, obtains following formula:

$\stackrel{\→}{N}\left(x_i\right)\≈\frac{\▿f\left(\stackrel{\→}{x_i}\right)}{\left|\right|\▿f\left(\stackrel{\→}{x_i}\right)\left|\right|}---\left(5\right)$

$\begin{array}{c}\▿f\left(\stackrel{\→}{x_i}\right)\≈[f(x_{i-1},y_j,z_k)-{\mathrm{fix}}_{i+k}{y}_j,z_k),\\ f(x_i,y_{j-1},z_k)-f(x_i,y_{j+1},z_k),\\ f(x_j,y_j,z_{k-1})-f(x_i,y_j,z_{k+1})]\end{array}---\left(6\right)$

According to (6) formula, calculate the gradient of all voxels, and with the normal direction value of this approximate representation voxel place face;

S4. adopt phong illumination model to carry out illumination calculation:

$I_p=K_a{i}_a+{\mathrm{\Σ}}_{m\∈\mathrm{lights}}(K_d(\stackrel{\→}{L_m}\·\stackrel{\→}{N})i_{m,d}+K_s{(\stackrel{\→}{R_m}\·\stackrel{\→}{V})}^{\mathrm{\α}}{i}_{m,s})---\left(7\right)$

Determine the illumination coefficient of all voxels;

Figure 2 shows that the phong modelling effect figure that adopts classic method, by revising anomalous body outer casing thickness, be that transparency threshold testing carrys out simulated effect, color table is set as shown in Figure 3, color gradient is from white to black gray expandable, the place that transparency is low is black gray expandable, and the part that transparency is high presents white, and the geological data of choosing is river course data, while only adopting phong model to carry out emulation, obtain result shown in Fig. 2.

As shown in Figure 2, while only adopting classic method to solve Seismic Problems, river course data division and sand body data division can not well make a distinction.In geological data is processed, in space structure, anomalous body is reflected as the external envelope of segment space structure.And in data plane, by it, the connectedness in space is divided into group to anomalous body, and the triangulation network that each group is comprised of a plurality of triangles represents.Every layer of transparence value of anomalous body is all not identical.Conventionally in application is processed, need to show that the shell shape of different aspects is is further researched and analysed.When only wanting the river course intermediate level to process with effect demonstration, master mould can not meet user's needs.

So to the correct of phong model, select interest-degree to distinguish according to ROI, can be river course data outer casing thickness according to user's selective discrimination anomalous body, facilitate user further to observe inner structure, carry out S5.

S5. increase α preference pattern and distinguish demonstration, carry out final rendering image and realize geological data 3-D display, wherein α is illumination transparency parameter, and α is larger, the more smooth minute surface that is more near the mark in surface, when this parameter is very large, the part area of DE Specular Lighting is very little.α is set to respectively to 0.1,0.9, the river course effect data of shading method that obtains a kind of geological data of the embodiment of the present invention shown in Fig. 4, Fig. 5 shows, river course data division is carried out to local light according to strengthening, and other sand bodies are partly carried out to Fuzzy processing, by user, manually input interest level and to after data degree of attention assignment, effectively interested data division given to high attention degree and given low attention degree by unwanted data.The river course part effect that carefully contrast is 0.1 and 0.9 as α shows, as shown in Figure 6, Figure 7, contrast, with a part, is selected constant in the situation that at color table, and grey is outermost layer partial data, color is redder is exactly the data at center, more inclined to one side river course, after threshold value strengthens, some grey color parts are fallen by refinement, and the part in river course more highlights, and have fuzzyly like this fallen the skin in river course, thereby can carry out modeling and demonstration to the target that cannot accurately describe, and improve the efficiency of volume drawing.

Those of ordinary skill in the art will appreciate that, embodiment described here is in order to help reader understanding's principle of the present invention, should be understood to that protection scope of the present invention is not limited to such special statement and embodiment.Those of ordinary skill in the art can make various other various concrete distortion and combinations that do not depart from essence of the present invention according to these technology enlightenments disclosed by the invention, and these distortion and combination are still in protection scope of the present invention.

Claims (4)

1. a shading method for geological data, is characterized in that, comprises the following steps:

S1. geological data is carried out to pre-service;

S2. input scene;

S3. based on GPU, carry out interpolation calculation, obtain following formula:

$\stackrel{\→}{N}\left(x_i\right)\≈\frac{\▿f\left(\stackrel{\→}{x_i}\right)}{\left|\right|\▿f\left(\stackrel{\→}{x_i}\right)\left|\right|}---\left(1\right)$

$\begin{array}{c}\▿f\left(\stackrel{\→}{x_i}\right)\≈[f(x_{i-1},y_j,z_k)-{\mathrm{fix}}_{i+k}{y}_j,z_k),\\ f(x_i,y_{j-1},z_k)-f(x_i,y_{j+1},z_k),\\ f(x_j,y_j,z_{k-1})-f(x_i,y_j,z_{k+1})]\end{array}---\left(2\right)$

Wherein for the gradient vector of volume data, for this unit normal vector, i, j, k are three direction vector of unit length of x, y, z, calculate the gradient of all voxels according to (2) formula, and seemingly represent the normal direction value of voxel place face with this;

S4. carry out illumination calculation;

S5. increase α preference pattern and distinguish demonstration, carry out final rendering image and realize geological data 3-D display, wherein α is illumination transparency parameter.

2. the shading method of a kind of geological data as claimed in claim 1, is characterized in that: in described step S3, calculate voxel gradient and adopt central difference algorithm to realize.

3. the shading method of a kind of geological data as claimed in claim 1, is characterized in that: in described step S3, based on GPU, carrying out interpolation calculation is, based on GPU, interpolation calculation is carried out in six orientation, front and back up and down of current process points.

4. the shading method of a kind of geological data as claimed in claim 1, is characterized in that: in described step S4, carry out illumination calculation and adopt Phong illumination model to realize, specific algorithm is:

$I_p=K_a{i}_a+{\mathrm{\Σ}}_{m\∈\mathrm{lights}}(K_d(\stackrel{\→}{L_m}\·\stackrel{\→}{N})i_{m,d}+K_s{(\stackrel{\→}{R_m}\·\stackrel{\→}{V})}^{\mathrm{\α}}{i}_{m,s})---\left(3\right)$

I wherein _pfor the intensity of illumination of the upper point in each surface, i _sand i _drespectively mirror-reflection light intensity and the diffuse reflection light intensity of each light source, i _arepresent environmental light intensity, m represents m light source, i _{m, d}represent the diffuse reflection light intensity of m light source, i _{m, s}the mirror-reflection light intensity that represents m light source, Ks is specularity factor, the reflectivity of reflection incident light, Kd is the Diffusive intensity that diffuse-reflection factor represents incident light, Ka represents diffuse reflection parameter, represent that point is to the direction of surface point, representation line vector, the reflected light of m light source, the direction vector from point to observer, when using outside direction light source, vector be a normal vector, with point location-independent, in addition about normal vector there is equation:

${\stackrel{\→}{R}}_m=2(\stackrel{\→}{L_m}\·\stackrel{\→}{N})\stackrel{\→}{N}-\stackrel{\→}{L_m}---\left(4\right)$