Depending on the scale, the time required for computer simulation is different, from computer programs that only need to run for a few minutes, to network-based computer groups that need to run for several hours, and then to large-scale simulation that needs to run for several days. The event scale of computer simulation has far exceeded any possibility (even anything imaginable) that traditional paper-and-pencil mathematical modeling can achieve: more than a decade ago, many supercomputers of the US Department of Defense's high-performance computer modernization program were used to simulate the desert battle in which one army attacked another. Among them, there are as many as 66,239 vehicles such as tanks and trucks in the simulation area around Kuwait. The research object of information visualization is the visual representation of large-scale non-digital information. This kind of information, such as a large number of files or code lines in software systems, libraries and bibliographic databases, and the relationship network on the Internet.
Information visualization focuses on establishing means and methods to convey abstract information in an intuitive way. Visual expression and interactive technology take advantage of the wide bandwidth of human eyes leading to the depths of the soul, so that users can witness, explore and even immediately understand a large amount of information. In computer graphics, rendering refers to the process of using computer programs to generate images according to models. Among them, the model is a description of three-dimensional objects with strictly defined language or data structure; This model generally contains geometry, perspective, texture, illumination, shadow and other information; The rendered image is a digital image or a bitmap (also called a raster image). The word "rendering" may be a kind of analogy for artists to render scenes. In addition, rendering is also used to describe the process of calculating the effect in the video editing file in order to generate the final video output. Surface rendering, also known as surface rendering. Stereo rendering, also known as volume rendering, volume rendering or stereo rendering, refers to the technical method of projecting and displaying three-dimensional discrete sampling data sets in two dimensions. A typical three-dimensional dataset is a set of two-dimensional slice images collected and reconstructed by CT, MRI or PET technology. Usually these images are collected and reconstructed according to some regular pattern (such as one layer every millisecond); Therefore, these images all have the same number of pixels under the same regular pattern. These are examples of regular solid grids. Wherein each entity element or voxel is represented by a single value, and the value is obtained by sampling in an adjacent area around the corresponding voxel. Important rendering techniques and methods include:
Scanning line rendering and rasterization
The advanced expression of an image must contain some elements different from pixels. These elements are called primitives. For example, in terms of schematic diagrams, line segments and curves may belong to entities. In a graphical user interface, primitives can be windows and buttons. In 3D rendering, primitives can be triangles and polygons in 3D space.
Ray projection
Ray casting is mainly used for real-time simulation, such as simulation methods used in three-dimensional computer games and cartoon animation (details are not important, or manual imitation of details is more effective in order to obtain better performance in the calculation stage). This is especially true when a large number of frames need animation. Without any extra skills, the final surface will have a typical "flat" effect, as if all the objects in the scene were painted matte.
Radiation coloring
Radiation coloring, also known as global illumination, refers to the method of simulating a directly illuminated surface as an indirect light source to illuminate other surfaces. This method will produce more realistic shadows, thus better capturing the ambient light of indoor scenes. A typical example is the shadow in the corner of the room.
Ray tracing
Ray tracing is an extension of the same technical method established in scanning line rendering and rasterization. Similar to the situation of scanning line rendering and rasterization, complex objects are dealt with, and it is possible to describe these objects mathematically. Different from scanning line and projection method, ray tracing is almost always a Monte Carlo technique, while the latter depends on the average of many samples randomly generated from the same model. Star formation: The right picture depicts the three-dimensional schematic diagram of the logarithm of gas/dust density in the simulation of Enzo and the Milky Way. Among them, the high-density area is white, and the low-density area tends to be blue, which is more transparent.
Gravitational wave: With the help of Globus toolbox and several supercomputers, the researchers simulated the gravitational effect of black hole collision.
Supernova explosion of big stars: visualization is the result of three-dimensional radiation hydrodynamics calculation about supernova explosion of big stars. Among them, the three-dimensional calculation of SN 1987A model explosion uses DJEHUTY star evolution code.
Rendering of 3D molecular structure: The pictures selected above are the rendering results of molecular structures created by using the general drawing function of VisIt. Its raw data is taken from protein database (PDB) and converted into VTK file in VisIt before rendering. Drawing surface topography by computer: Using computer to draw surface topography, mathematicians can test the theory of how various materials will change under stress conditions. This imaging task is one of the tasks of the Electronic Visualization Laboratory of the University of Illinois in Chicago, USA, which is funded by NSF.
Graph: VisIt can use data read from a file to draw a curve. With the help of corresponding operators or queries, it can also be used to extract and draw curve data from high-dimensional data sets. The curve in this selected picture is drawn by using the lineout function of VisIt, which corresponds to the elevation data distributed above the elevation line of DEM data. Lineout enables users to draw lines interactively, thus specifying the path of extracting data in detail. VisIt can then plot the final data into a curve.
Image annotation: This selected image shows the leaf area index of NetCDF dataset; This index is an indicator reflecting the global vegetation status. The main picture is the big picture at the bottom, showing the world's lai. The two small pictures in the upper part are actually annotations, which were generated by VisIt earlier. Image annotation can be used to contain materials aimed at enhancing visualization results, such as auxiliary drawings, experimental data images, project logos, etc.
Scatter chart: The access of scatter chart can be used to visualize multivariate data up to four dimensions. This scatter plot can accept multiple scalar variables and use them on different axes in phase space. Different variables are combined with each other and * * * is isomorphic to the coordinates of each point in the phase space. In addition, each point is displayed by a symbol and colored according to the value of another scalar variable. Porsche 9 1 1 model: NASTRAN model. This selected picture is the grid diagram of Porsche 9 1 1 imported from NASTRAN batch data file. Access can read a limited subset of NASTRAN batch data files. Usually, this limited subset is enough to import those geometric features suitable for visualization in the model.
Rendering of the city: Here, VisIt reads an ESRI shape file; This file contains a polygon description of the building footprint. Then, these polygons are resampled on a straight grid. Finally, this linear grid is highlighted as the selected urban landscape.