CN102903141B - Many seismic properties based on opacity weighting merge texture mapping object plotting method - Google Patents

Abstract

The invention discloses a kind of many seismic properties based on opacity weighting and merge texture mapping object plotting method, obtain mapped color value and the opacity of voxel respectively with the transmission function of each attribute data, then determine the proportion shared by the color value of this seismic properties component according to the opacity of each voxel；Finally synthesizing new color value through opacity weighting, then draw out by the mode of 2 d texture volume drawing, the inventive method shows while achieving multiple seismic properties on said three-dimensional body is drawn, and to seismic interpretation, personnel bring convenience.

Description

Multi-seismic attribute fusion texture mapping volume rendering method based on opacity weighting

technical field

The invention relates to a multi-seismic attribute fusion texture mapping volume rendering method based on opacity weighting.

Background technique

With the application and development of various technologies in oilfield seismic exploration, exploration workers have higher and higher requirements for seismic results. In the face of increasingly complex and hidden oil reservoir conditions, in order to improve the success rate of drilling, it is necessary for exploration workers Provide a brand-new geological structure form and its attribute characteristics. The emergence of scientific computing visualization technology provides mature technical support for the realization of this idea. Scientific computing visualization technology is applied to seismic data, resulting in seismic visualization technology. Seismic visualization can not only improve the efficiency, accuracy and completeness of seismic interpretation, but it is also easy to understand and display a large amount of data, which greatly improves the interpreter's understanding of the data volume.

The visualization of 3D data field is the core field of scientific computing visualization. The research results of this technology have been widely used in the fields of medicine, meteorology and geological exploration. The visualization of 3D data field is mainly to analyze and process various data defined in 3D space, obtain the 3D display effect of the object object by processing the 2D data sequence, and make an objective analysis of the 3D entity object according to the display effect or interactive operation. Theories, methods and techniques of qualitative or quantitative analysis.

The visualization methods of 3D data field can be roughly divided into three categories according to the different data description methods in the drawing process. One is to describe the three-dimensional structure of the object by splicing and fitting the surface of the object, which is called the surface rendering method; the other is to directly generate a two-dimensional image on the screen from the three-dimensional volume data, which is called the volume rendering method. Also known as the direct volume rendering method. In addition, there is another type of algorithm that aims to reflect the overall information of the data and uses geometric shapes as the display unit. This type of algorithm is called a hybrid rendering method. Volume rendering methods directly apply vision principles to synthesize 3D images by resampling volume data.

The direct volume rendering algorithm can be divided into image space algorithm and object space algorithm. The image space algorithm starts from the pixels on the screen, samples and mixes in the volume data field to accumulate the pixel color, and directly obtains the final image; while the object space algorithm It is to scan voxels or units in a predetermined order and project the voxels or units onto the pixels of the screen. The most representative ones in image space volume rendering algorithms are RayCasting and TextureMapping volume rendering. Shear-Warp and Splatting algorithms are representative algorithms in object space volume rendering.

The first three drawing methods are far from the requirements of real-time interactive display. Even with some improved acceleration algorithms, it is still impossible to realize real-time dynamic display of large-scale volume data on a PC. In order to achieve large-scale applications, people turn to hardware, and use the texture mapping function provided by high-end graphics cards to perform direct volume rendering of 3D data fields, thereby greatly improving the rendering speed.

Two-dimensional texture mapping is a direct volume rendering algorithm in object space, which converts the resampling process of three-dimensional space into the resampling process of two-dimensional plane. Using a simple quadratic linear interpolation operation instead of a cubic linear interpolation operation can greatly reduce the amount of calculation. Define a series of sampling polygons in the three-dimensional data field. These polygons are parallel to each other and perpendicular to the coordinate axis with the smallest angle with the normal direction of the viewing plane in the object space. These polygons are equivalent to a series of slices of volume data. Due to the polygon interval and The sampling density is different from the original data, and only by resampling can the value of each sampling point on this series of parallel planes be obtained. If the viewing direction changes by more than 90 degrees, the direction of the sampled polygon normal must also change, so multiple copies of the volume dataset need to be kept in memory, each copy must represent a different polygon normal direction. Finally, composite into the final image.

However, the above visualization techniques are only displayed for the same type of seismic data. In seismic exploration, geophysicists can extract more than a hundred attributes (such as amplitude, frequency, phase, waveform, structure, prestack attributes, and spectral decomposition attributes, etc.) from seismic data, and try to Qualitative or quantitative description of subsurface geological bodies through seismic attributes. However, due to the unavoidable multi-solutions of geophysical exploration methods, geological interpretation problems are often prone to deviations when only using a single seismic attribute or a composite single comprehensive attribute, and often lose a lot of useful information when only using the same attribute for interpretation. information. However, it is very difficult to perform fusion calculations on many different types of seismic data (such as amplitude volume and coherence volume). Not only the complexity is particularly high but also the effect is not good.

In the literature "Multicolordisplayofspectralattributes" (TheLeadingEdge, 2007, 26(3): 268-271), Liu et al. proposed a method based on the user-defined spectrum range, for different frequency data volumes after spectral decomposition, using RGBA color fusion based on cosine function transformation technology. In the literature "Principal components analysis of spectral components" (Expanded Abstracts of 76 Annual Internat SEGMtg, 2006, 988-992), Guo et al. proposed to apply principal component analysis (PCA) technology and RGBA color fusion technology to spectral decomposition data to improve the ability of spectral decomposition data to identify river channels. Ding Feng et al. proposed in the literature "Application Research of Seismic Multi-Attribute RGB Color Fusion Technology" (Petroleum Geophysical Prospecting, 2010, 5) that the key to PCA-RGBA seismic multi-attribute RGBA color fusion is to combine four different seismic attribute values ( M, i=1, 2, 3, 4) are mapped into four colors of R, G, B, and A one by one through a certain transformation.

The above methods are all applied to the two-dimensional seismic section, and there is no method for volume perspective. Moreover, the above method can only realize the mapping of seismic attributes of up to four components. Today, with the development of survey technology, there are hundreds of seismic attributes. In some respects, the above methods fail to meet the requirements. And the A (opacity) value is also used as a component, such a component can no longer adjust the transparency, and the component of interest cannot be found from the perspective of volume.

Traditional visualizations are all about one type of seismic data. However, with the development of survey technology and subsequent processing technology, a single attribute can no longer explain seismic information well.

However, for the processing methods of various seismic attributes, there is no effective algorithm for computing fusion of various seismic attributes. The fusion during rendering has only been realized in 2D section display, and there is no good algorithm in 3D seismic volume rendering.

Contents of the invention

In order to overcome the above-mentioned shortcomings of the prior art, the present invention provides a multi-seismic attribute fusion texture mapping volume rendering method based on opacity weighting. The transfer function of each attribute data is used to obtain the mapping color value and opacity of the voxel, and then Determine the proportion of the color value of the seismic attribute component according to the opacity of each voxel; finally synthesize a new color value through opacity weighting, and then draw it out in a two-dimensional texture volume rendering method. Simultaneous display of multiple seismic attributes is realized, which brings convenience to earthquake interpreters.

The technical solution adopted by the present invention to solve its technical problems is: a method for volume rendering based on multi-seismic attribute fusion texture mapping based on opacity weighting, comprising the following steps:

Step 1. Import seismic data;

Step 2. Data preprocessing:

1) The seismic data are sampled at equal intervals in three directions: the main survey line inline, the slave survey line xline perpendicular to the main survey line, and the time axis time;

2) Process the voxel in each attribute data, and convert the value of the voxel into a false color value of 0-255;

Step 3, respectively map the color value and opacity to the voxel of each seismic attribute data volume:

1) Color setting: establish the correspondence between the pseudo-color value of the voxel and the color index table through the color transfer function;

2) Setting of opacity;

Step 4, color and opacity of fusion multiple seismic data; Described fusion method is to each attribute data according to the method of opacity weighted addition to get mean value;

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$

$\mathrm{<span\; class="notranslate">\; G</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$

$\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; r</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$

Among them, R is red, G is green, B is blue, A is opacity, i is the i-th seismic attribute data, n is a total of n seismic attribute data, and ∑ represents the summation operation;

Step 5. Generate a two-dimensional texture patch:

1) The data data is sliced in the vertical direction according to the three directions of xline, inline and time to generate a texture image;

2) Create a texture object and bind the texture;

3) texture mapping;

Step 6. Draw the image:

1) Determine the direction of drawing texture according to the direction of sight;

2) According to the line of sight, the texture is blended and drawn from the back to the front.

Compared with the prior art, the positive effect of the present invention is: realize the common display of various seismic data in 3D volume rendering, solve the problem that a single attribute cannot be well interpreted, and different data types cannot be directly fused and calculated. It improves the accuracy, reliability and effectiveness of seismic data interpretation, and provides an effective method for seismic interpretation, as follows:

(1) It can make full use of the structural and lithological change information contained in seismic attributes, overcome the shortcomings of insufficient display of single seismic attributes, improve the ability to extract geological bodies from multiple attributes, and enhance the clarity of image display. The characteristics of obviousness, rich details, high information content and multi-attribute joint display. Its display effect has obvious advantages over conventional seismic attribute display methods. It enables interpreters to visually judge and analyze various earthquake information, improving efficiency and accuracy.

(2) The color value of each seismic attribute data is fused instead of the original data value. Seismic attribute data are different, and it is difficult to have an algorithm for fusion (such as vector and scalar). Even if fusion is possible, the algorithm is too complicated and the gain outweighs the gain. But the realization algorithm of the present invention is simple, and effect is obvious.

(3) The method of texture mapping volume rendering is adopted, and the rendering speed is fast and the interaction is good.

detailed description

Before introducing the scheme of the present invention, introduce several concepts:

SEGY: SEGY (Society of Exploration Geophysicists Y format) seismic data is one of the general format standards for many tape-type seismic data developed by the American SEG (Society of Exploration Geophysicists) Association in the 1970s to facilitate the exchange and sharing of seismic data. It is organized in units of seismic channels and stores seismic channel data channel by channel. At present, SEGY seismic data files have become the most commonly used data recording carrier in outdoor data acquisition in the oil and gas exploration industry.

Transfer function: The transfer function mainly maps the data values of the three-dimensional data field to optical properties, such as opacity, color values, etc., which determine the imaging quality of volume rendering. Therefore, the design of the transfer function is a key process in the volume rendering process.

Mathematically, the transfer function can be defined as a mapping from the Cartesian product of three-dimensional data fields to the Cartesian product of optical properties:

τ:D ₁ ×D ₂ ×……×D _n →O ₁ ×O ₂ ×……×O _n

where D represents the data attribute of the three-dimensional data field, and is the definition domain of the transfer function. The data attribute of the three-dimensional data field can be the data value of the sampling point, or the data calculation result of the local sampling point, such as the gradient modulus, the second derivative of the gradient direction, curvature, spatial coordinates, etc.; O is the value range of the transfer function , representing the optical properties for visualization, such as color, transparency, shading parameters, reflectivity, refraction, and so on. Through the definition of the transfer function, it can be seen that the sampling points with certain data attributes in the 3D data field will be displayed in the 2D image in a certain form. Designing the transfer function is to select and design appropriate data attributes and optical attributes according to the needs of visualization, and to establish the mapping relationship between them.

Opacity: Opacity is a measure of the ability of light to pass through a surface. A fully opaque surface has an opacity of 1 and blocks all light incident on the surface from passing through. On the other hand, if the surface has an opacity of 0, it is completely transparent and the surface is invisible. A semi-transparent surface with an opacity between 0 and 1 that reveals the previously stored color subtly. It is like the effect of observing a colored object through colored glass: the color of the object can be seen, and it has the color of the glass. Sometimes we express it in terms of transparency, a surface with an opacity of A has a transparency of 1-A. If we regard the texture fragment as the source pixel and the pixel in the frame buffer as the target pixel, there are many combinations between the two, which can be replaced or fused through A.

A multi-seismic attribute fusion texture mapping volume rendering method based on opacity weighting, comprising the following steps:

Step 1. Import seismic data:

Save multiple seismic data grids separately to memory. The main features of SEGY seismic data are standard data storage format, comprehensive seismic information and certain scalability. These characteristics determine that it is more suitable for seismic data exchange and sharing among different seismic exploration software systems. On the other hand, because the redundant information contained in the SEGY seismic data is too complicated, the additional overhead of software system resources is large; therefore, in the application scenarios with high interaction requirements and strict response time requirements, the SEGY seismic data can be read directly according to the data format. Data reading efficiency is not ideal. To this end, the seismic data are first gridded in three directions: the main survey line inline, the secondary survey line xline perpendicular to the main survey line, and the time axis time and saved in the memory.

Step 2. Data preprocessing:

1) The seismic data are sampled at equal intervals in three directions: the main survey line inline, the slave survey line xline perpendicular to the main survey line, and the time axis time. Due to hardware limitations, the maximum size of the sampled data is 512×512×512, and the sampling interval is determined by the size of the data. Let the data size be l×w×h, where l, w, and h are the length, width, and height of the data respectively, and correspond to the xline, inline, and time directions respectively, and the sampling intervals are l/512+1, w/ 512+1 and h/512+1. Then the sampled size is:

L=l/(l/512+1)

W=w/(w/512+1)

H=h/(h/512+1)

Among them: L, W, and H are the length, width, and height after sampling, respectively. The intersection position of the grid is regarded as a voxel, and the value of each voxel is saved in the three-dimensional array data.

2) Process the voxels in each attribute data. That is, the value of the voxel is converted into a pseudo-color value of 0-255. The false color value is of type unsignedchar. In the present invention, a first-order linear transformation is used. which is

${\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; no</span>}}}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; 255</span>}$

Among them, V _i is the final pseudo-color value of the i-th voxel, v _i is the attribute value of the i-th voxel, v _max and v _min are the maximum and minimum values of the data, and the user is allowed to adjust this correspondence. By adjusting the range of the data, the user can set the maximum and minimum values of a data, only the volume data within this range can be mapped to the color index table, and the data outside this range will not be drawn. This is beneficial to eliminate the impact of irrelevant data on the final image and highlight the key parts of the image.

Considering the most commonly used attributes of seismic waves, such as positive and negative reciprocating distribution characteristics of amplitude and frequency, the present invention has also made a further refinement, so that the pseudo-color values of non-negative sampling point data are distributed to the interval (0,127), and the sampling of negative numbers Pseudocolor values for point data are distributed into the interval (128, 255).

Step 3: Map the color value and opacity to the voxel of each seismic attribute volume respectively:

The char-type pseudo-color value of each voxel is mapped into a color through a transfer function. The colors here are in RGBA format, namely red (R), green (G), blue (B), and opacity (A). The current computer graphics color system generally supports 32-bit (256×256×256×256) color depth, in which red (R), green (G), blue (B), and transparency (A) each occupy a color channel. Computer display devices produce more colors by mixing the four components of RGBA, that is, I _RGBA = s(I _R , I _G , I _B , I _A ). I _RGBA represents the color of a certain point; I _R , I _G , I _B , and I _A represent four colors of red, green, blue, and transparent respectively; S represents the color mixing transformation, which is completed by the computer display device.

Reasonable and effective design of transfer function occupies a very important position in volume rendering algorithm. The volume rendering transfer function is responsible for assigning the optical properties (color, opacity, etc.) information.

1) The setting of the color, the transfer function of the color is the correspondence between the pseudo color value of the voxel and the color index table.

The color index table is provided by the user, and the user only needs to provide the index value between 0-255 and the corresponding color component RGB. According to the color index provided by the user, it is interpolated to obtain a complete gradient color index table, and the color index table corresponds to the pseudo color value of 0-255. For each voxel, look up the color index table to get the color of this voxel. After traversing each voxel, the setting of all data colors is completed.

2) The setting of opacity. Whether the voxels in the opaque volume data are visible in the drawn image depends on the opacity value assigned by the transfer function. The voxel of interest is assigned a higher opacity value, and vice versa. the opacity value. The voxel value has been mapped into a color value, so we can adjust the opacity according to the color table, so that some interesting colors have higher opacity, so that they can be seen more clearly.

Step 4. Blend the color and opacity of multiple seismic data:

For n seismic data data ₁ , data ₂ . . . data _n . Among them, data _i stores the color value and opacity of the i-th seismic data, and the size is L×W×H. For a certain position in the grid, we get a color value I _RGBAi of the i-th seismic data voxel at this position from data _i . Among them, RGB is the color value, and A is the opacity, then the fusion result of all colors and opacity at this position is:

I _RGBA ＝f(I _RGBA1 ,I _RGBA2 ,……I _RGBAn )

where f is the fusion method.

The present invention adopts a fusion method in which each RGB component is weighted and added according to the opacity to obtain an average value. The reason why opacity is chosen as the weighting factor is that the setting of opacity is closely related to our interested goal. Therefore, the data values of each attribute are:

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$

$\mathrm{<span\; class="notranslate">\; G</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$

$\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$

$\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$

It can be seen from the above fusion formula that the high opacity means that the color of the seismic attribute that is relatively important to us accounts for a relatively larger proportion, so that we can easily see the effect we want in the overall drawing. The RGBA of each voxel can be set. The color and opacity of all voxels are fused, and the fused value is saved in data.

Step 5. Generate a two-dimensional texture patch:

1) Slice the data data in the vertical direction of xline, inline, and time to generate a texture image. The reason why three directions are used is that if you only slice along one direction, you will see a line when the line of sight is perpendicular to this direction, and the effect is very bad. However, if you slice along the line of sight, the texture will be regenerated every time the direction changes after rotation, and the interactivity is not good. Here, the time direction is taken as an example: Slice sampling is performed at an interval of distance d along the Z-axis direction, and the size of the obtained slice is L×W, and the number of slices is W/d. Among them, the sampling interval d determines the quality of the drawing. Smaller d can get more slices, which can more accurately describe the data field, but at the same time it will increase the drawing operations of the graphics hardware and reduce the interaction speed. Larger d can get faster drawing speed, but at the cost of sacrificing the quality of the drawn image. Here we take d as 1, so there are H slices in the time direction, and each slice is saved as a two-dimensional array textureData. All the data in this array forms the pixel values of a picture, that is, a texture image. Similarly, the xline direction and the inline direction are also sliced separately.

2) Create a texture object and bind the texture. Use the glGenTextures function to create a texture and the glBindTexture function to bind a texture object. The glTexImage2D function transfers the pixel values in the textureData array to the currently bound texture object, thus creating a texture.

3) Texture mapping. For any texture, regardless of the actual size of the texture, the texture coordinates of the top (upper left corner) are always (0,0), and the texture coordinates of the lower right corner are always (1,1). That is, texture coordinates should be a decimal number between 0 and 1. Mapping is to match the texture coordinates with the actual three-dimensional coordinates. Mapping should be done in a unified order, otherwise the picture will be displayed incorrectly if it is pasted backwards.

Step 6. Draw the image:

1) Determine which direction to draw the texture according to the view direction. We have prepared texture maps xline, inline and time in three directions, but it is not necessary to draw all of them after each operation such as rotation, but only need to draw the coordinate axis closest to the line of sight. Assuming that the line of sight direction is unitized as d(x, y, z), the selected texture direction is the direction represented by the largest value among the three values of x, y, and z. For example, d(0, 0, 1) is to draw the texture in the time direction. This way we can always see the texture close to our line of sight, which works well. And there is no need to draw textures in all three directions, and the interactivity is good.

2) According to the line of sight, the texture is blended and drawn from the back to the front. The fusion of textures is done by hardware, but it must be blended from far to near to achieve the correct effect. The main reason is that there may be overlapping areas between different objects, and the drawing order will affect the color of the overlapping areas.

Claims (3)

1. A multi-seismic attribute fusion texture mapping volume rendering method based on opacity weighting, is characterized in that: comprise the steps: Step 1. Import seismic data; first grid the seismic data according to the main survey line inline, the slave survey line xline perpendicular to the main survey line, and the time axis time and save them in the memory; Step 2. Data preprocessing: 1) The seismic data are sampled at equal intervals in three directions: the main survey line inline, the slave survey line xline perpendicular to the main survey line, and the time axis time; 2) Process the voxel in each attribute data, and convert the value of the voxel into a false color value of 0-255; Step 3, respectively map the color value and opacity to the voxel of each seismic attribute data volume: 1) Color setting: establish the correspondence between the pseudo-color value of the voxel and the color index table through the color transfer function; 2) Setting of opacity; Step 4, color and opacity of fusion multiple seismic data; Described fusion method is to each attribute data according to the method of opacity weighted addition to get mean value; $\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$ $\mathrm{<span\; class="notranslate">\; G</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$ $\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&divide;</span>\mathrm{<span\; class="notranslate">\; no</span>}$ Among them, R is red, G is green, B is blue, A is opacity, i is the i-th seismic attribute data, n is a total of n seismic attribute data, and ∑ represents the summation operation; Step 5. Generate a two-dimensional texture patch: 1) The data data is sliced in the vertical direction according to the three directions of xline, inline and time to generate a texture image; 2) Create a texture object and bind the texture; 3) texture mapping; Step 6. Draw the image: 1) Determine the direction of drawing texture according to the direction of sight; 2) Fusion drawing from back to front texture according to line of sight. 2. the multi-seismic attribute fusion texture mapping volume rendering method based on opacity weighting according to claim 1, is characterized in that: when the voxel in each data is processed, make the false of non-negative sample point data The color value is distributed to the interval (0,127), and the pseudo-color value of the negative sampling point data is distributed to the interval (128, 255); For n seismic attribute data data ₁ , data ₂ ...data _n , where data _i stores the color value and opacity of the i-th seismic attribute data, the size is L×W×H, for a certain position in the grid, Obtain a color I _RGBAi of the i-th seismic attribute data voxel at this location from data _i , where RGB is the color value and A is the opacity, then the fusion result of all colors and opacity at this location is: I _RGBA ＝f(I _RGBA1 ,I _RGBA2 ,...I _RGBAn ) where f is the fusion method. 3. The multi-seismic attribute fusion texture mapping volume rendering method based on opacity weighting according to claim 1, characterized in that: the color setting method is: according to the index value between 0-255 provided by the user and the corresponding The color component RGB is interpolated to obtain a complete pseudo-color value gradient color index table corresponding to 0-255; for each voxel, look up the color index table to obtain the color of the voxel.