Modern Data Visualization Technology

Human vision... the most direct way to accept information... or to visualize the data to make it easier for people to understand the data.

------------------------

Modern data visualization technology refers to the use of computer graphics and image processing technology to convert data into graphics or images displayed on the screen, and interactive processing theory, method and technology. It involves computer graphics, image processing, computer-aided design, computer vision and Human-Computer Interaction Technology. The concept of data visualization first comes from the concept of scientific computing visualization (Visualization in scientific computing). Scientists not only need to analyze data calculated by computers through graphical images, in addition, you need to understand the data changes in the computing process. With the development of computer technology, the concept of data visualization has been greatly expanded. It includes not only the visualization of scientific computing data, but also the visualization of engineering data and measurement data. In academia, the visualization of spatial data is often referred to as the volum visualization technology. In recent years, with the development of network technology and e-commerce, the requirements for information visualization have been put forward. We can use data visualization technology to discover hidden patterns in a large number of financial, communication, and commercial data, so as to provide a basis for decision-making. This has become a new hotspot in data visualization technology.

To adapt to the rapid development of hardware platforms, operating systems, networks and communications, visual software products have developed rapidly in recent years, it is represented by AVS/Express development, IDL (including VIP, ion) and PV-WAVE. AVS/Express development edition provides a multi-platform interactive multi-dimensional visual software development and integration environment.

Ii. Significance of developing data visualization
How can we analyze massive, complex, and multidimensional data? The answer is to provide an intuitive, interactive, and responsive visual environment like the human eye. Therefore, the main features of data visualization technology are:

(1) interaction. Users can easily manage and develop data in interactive mode.

(2) multidimensional. You can see multiple attributes or variables that represent objects or events. data can be classified, sorted, combined, and displayed based on the values in each dimension.

(3) visibility. Data can be displayed using images, curves, 2D images, 3D bodies, and animations, and its modes and relationships can be visually analyzed. History has proved that human vision has played an outstanding role in scientific discovery. In terms of visualization, the emergence of key technologies is a prelude to major scientific discoveries. The role of telescopes and microscopy in astronomy and biological development is evidenced. These tools are used to zoom in and expand the human eye. Today, this principle is still true. Human visualization allows humans to analyze a large amount of abstract data. New data development tools can greatly expand our vision. People's creativity depends not only on people's logical thinking, but also on people's image thinking. Only by visualizing massive data into an image can we stimulate people's image thinking. On the surface, it seems that the massive amount of data is disorganized, and the hidden rules are found to provide a basis for scientific discovery, engineering development, medical diagnosis, and business decision-making. Here we must also differentiate the concepts of data, information, and knowledge. Data is a collection of symbols. Information is useful data. Information is not the same as knowledge. Information cannot reflect the internal relationship between data as knowledge does. Some people advocate that knowledge can be divided into two categories. One is that it cannot be described by language or text, and it is called tacit knowledge ); the other type is that it can be described in languages and texts. It is known as explicit knowledge ). Currently, information is such explicit knowledge. Looking forward to the future, based on breakthroughs in brain science, we will develop a computer-biological computer for the human brain to create a golden age of artificial intelligence. However, even then, information cannot fully express all the hidden knowledge of mankind. Only by expressing data and information in Graphs and images can we create conditions for gaining valuable hidden knowledge. In short, data visualization can greatly speed up data processing and effectively utilize the massive amount of data generated at all times. It can implement image communication between people and data, so that people can observe the hidden phenomena in the data and provide powerful tools for discovering and understanding scientific laws. It can guide and control the computing and programming processes, change the conditions of the process through interactive means, and observe its impact.

Computers have been used for scientific computing and data processing for nearly 50 years. However, for a long time, due to the limitations of computer technology, data can only be processed in batches rather than interactive processing. You cannot intervene or guide the computing process, but can only passively wait for the output of the computing result. A large amount of output data can only be processed manually, or two-dimensional graphics can be output using a plotter. In this way, not only can we not obtain the overall concept of data intuitively and visually in a timely manner, but also may lose a large amount of information. In recent years, data from supercomputer, satellite, advanced medical imaging equipment, and geological exploration has been growing, making data visualization an urgent problem. On the other hand, in recent years, due to the rapid increase in computer computing speed, the memory capacity and disk space have been constantly expanding, and network functions have been increasingly enhanced. hardware can be used to implement many important graphics generation and image processing algorithms, this makes it possible to use data visualization technology to visually and visually display massive amounts of data and information and perform interactive processing.

The following is an example to illustrate the significance of developing visual technology. For a long time, humans have the desire to understand their internal structures. It was not until the emergence of computer tomography (CT) and MRI (MRI) techniques and Visualization Techniques in 1970s that the desire to obtain internal data in the human body became a reality. To achieve this goal, the National Medical Library (NLM) began implementing the visual human body Program (VHP) in 1989 ). Commissioned the University of Colorado School of Medicine to establish a database of all Anatomy of a man and a woman. They scanned a male and a female from the beginning to the foot for a CT scan and MRI scan. Male spacing is 1mm, with a total of 1878 sections; female spacing is 0.33mm, with a total of 5189 sections. Then fill the body with blue latex and wrap it with gelatin and freeze it to minus 80 degrees Celsius, and then take a digital camera photograph of the body as tissue slices at the same distance. The resolution is 2048 × 1216. The total data is 56 GB (13 Gb for men and 43 GB for women ). Users around the world can obtain this huge data for teaching and scientific research after they sign a User Agreement with the National Medical Library of the United States and pay a small fee. The emergence of VHP datasets marks the entry of computer 3D reconstruction images and virtual reality technology into the medical field, which greatly promotes the development and popularization of medicine.

3. data visualization applications

Data visualization is widely used in almost all fields, such as natural science, engineering technology, finance, communication, and commerce. The following examples illustrate several successful data visualization applications. 1. Medicine
Medical data visualization has become one of the most active research fields in the field of data visualization. Due to the development of modern non-invasive diagnostic techniques such as CT, MRI, and positron tomography (PET), doctors have been able to easily obtain a set of two-dimensional tomography images of patient-related parts. CT breaks the traditional film photosensitive imaging mode, and uses computers to reconstruct images of human organs or tissues, so that medical images can go from two dimensions to three dimensions, so that people can see the interior from outside the human body. Pet combines nuclear technology with computer technology. After being injected into the human body by the labeled tracer, the positron annihilation produced during the native decay process can be obtained through electronic detection and computer reconstruction imaging, so that we can obtain human metabolism or functional images. On this basis, the visual software is used to perform image fusion for the above multi-modal images, it can accurately determine the space location, size, geometric shape of the lesion and the spatial relationship between the lesion and the biological tissue, so as to diagnose the disease in a timely and efficient manner. ADAC lab in California, USA, John. the two-dimensional tomography are used to reconstruct three-dimensional images of relevant organs and tissues by using visual software systems, such as the hopkind University, Focal Point Graphics Corporation, integrated medical image processing system Corporation, and the University of Berlin, Germany. The software they developed has been applied in many hospitals. In addition, the University of Washington in the United States uses a visual software system and cardiac ultrasound diagnostic technology to obtain three-dimensional heart images and monitor the shape, size, and motion of the heart, providing a basis for comprehensive diagnosis. The electron beam CT (EBCT) is replaced by an electron beam scan with a mechanical scan of the X-ray tube and detector. As a result, the scanning speed increases by nearly a hundred times and the organ that is examined for motion (such as the heart vascular) the film CT is a revolution in CT technology. The fuwai cardiovascular Hospital of Peking Union Medical University of China has used EBCT 3D image reconstruction for the clinical diagnosis of Aortic Lesions and the vascular display after coronary artery bypass grafting.
Due to the high time resolution of EBCT angiography images, respiratory and Motion Artifacts are eliminated, which can clearly diagnose various Aortic Lesions and display the anatomy of coronary artery bypass vessels. 3D reconstruction images help to visually display lesions and help diagnose and guide the operation. Therefore, in the diagnosis of Aortic Lesions and the vascular display after coronary artery bypass grafting, it is expected to replace invasive conventional angiography.

On the basis of visualization technology, we can further implement computer simulation and surgical planning for radiation and orthopedic procedures. For example, when performing brain tumor radiotherapy, the skull needs to be perforated, and then the radioisotope can be accurately placed in the brain lesions, both to make the best treatment effect, it is necessary to ensure that the entire operation process and isotope rays do not hurt normal tissue. Because the internal structure of the human brain is very complex, and doctors cannot observe the actual operation without the use of a skull, it is very difficult to meet the above requirements. The visualization technology can be used to simulate the location of skull perforation, isotope placement channel, placement location, and other dose lines on the basis of the reconstructed three-dimensional image of the internal structure of the human brain, and select the best solution. At the same time, the operation can be monitored on the screen to greatly improve the success rate of the operation. For example, many children have abnormal hip development. As an orthopedic operation, they need to perform operations such as cutting, moving, and fixation on the hip joint. By using visualization technology, we can first construct a three-dimensional image of the hip joint on the computer, and then simulate a variety of schemes on the computer, such as the cutting position, cutting shape, displacement, and fixation method, this greatly improves the quality of orthopedic procedures. 2. Oil and Gas Exploration

Many years ago, many large oil and gas fields were found. At present, the oil industry is facing a severe problem: how to find a small and buried oil and gas field. In addition to finding new oil fields, the emergence of new technologies also allows us to optimize existing oil fields and prolong the lifetime of many oil fields by improving analysis and recovery methods. Scientists and engineering and technical personnel must first accurately interpret a large amount of seismic exploration data before they can determine whether the oilfield exists and determine the mining and management plan for underground resources. The main method of oil and gas exploration is to reconstruct a wide range of geological structures through the propagation of natural seismic waves or sound waves produced by artificial explosion in geological structures, the formation structure of the local region is learned through the logging data, the location and distribution of the reservoir are verified, and the reserves and exploration value are estimated. Because the volume of seismic data and log data is extremely large and unevenly distributed, it is impossible to analyze the data on paper. The visualization technology can be used to construct the isosurface and contour of interest from a large amount of geological exploration data or logging data, and display its range and direction, multiple parameters and their relationships are displayed in different colors, so that professionals can correctly interpret the raw data and obtain important information such as whether the mineral deposits exist, the location of the mineral deposits, and the size of the reserves. This not only can guide drilling operations, reduce invalid wells, and save money, but also will greatly improve the efficiency of searching for reservoirs, thus having significant economic and social benefits. British PGs Tigress Co., Ltd. has developed data visualization software, which has been widely used in many oilfield and natural gas development in the world. This software can be used for seismic data processing, multi-well logging evaluation, and simulation of the storage and production processes of oil and gas. It can not only determine the location of oil and gas storage, but also track the movement of oil and gas to determine the optimal path for oil and gas exploitation. China Daqing Exploration and Development Research Institute developed a geological data visualization system, it provides a comprehensive, convenient, and flexible three-dimensional display of faults, parts, layers, and internal data types and geological attributes in 3D data bodies, with an object-oriented development environment, it can meet various data visualization requirements of users. 3. Weather Forecast
Weather forecasts are related to the lives of hundreds of millions of people, the sustained development of the national economy, and national security. The forecast and prevention of disastrous weather will greatly reduce the loss of people's lives and property. The accuracy of weather forecast depends on the calculation of a large amount of data and the analysis of the calculation results. On the one hand, scientific computing visualization can convert a large amount of data into images, the screen displays the isothermal, vortex, cloud layer position and movement at a certain time point, the position and intensity of the rainstorm area, the size and direction of the wind force, etc.

The reporter can accurately analyze and predict the future weather. On the other hand, according to the Global meteorological monitoring data and calculation results, the global temperature distribution, air pressure distribution, rainfall distribution and wind direction in different periods can be expressed in the form of images, in this way, we can study and predict the global meteorological conditions and their changing trends. The forecasting system laboratory of the National Oceanic and Atmospheric Administration (NOAA) developed an advanced version of the WFO-advanced office for weather forecasting. Its key part is to display 3D images of weather data. To this end, the laboratory has developed a 3D visualization software system display 3D (d3d ). This system can be used to display and process a large amount of data collected from balloons, ground stations, radars, airplanes and satellites, and track and evaluate important local meteorological conditions in a timely manner, in this way, the weather forecast can be made in a timely and accurate manner. Generally, meteorological workers manually overlay two-dimensional layered data for analysis. Using three-dimensional visualization, meteorological workers can be freed from a large number of two-dimensional image calculations, let them focus on the actual values required for forecasting. With WFO-advanced and d3d, meteorological workers can establish a medium-scale (20-200 km) or regional forecast model for the next 12-18 hours within four hours. The animation module in the software can generate image sequences and Display Dynamic images. The biggest feature of this software is that the generation of clouds is extremely lifelike. Our military meteorological department recently developed the "military numerical weather forecasting system", which can process meteorological observation data of thousands of meteorological stations at a high speed, automatic Rolling Production of daily military weather forecasts, military Weather Element forecasts and 3D visualization information within 10 days.

4. Engineering

Computer Aided Engineering (CAE) includes computer aided design (CAD), computer aided manufacturing (CAM), and Computer Aided operation. Visualization Technology facilitates the integration and streamlining of the entire project process, and enables engineering leaders and technicians to see and understand the dynamic impact of parameter changes on the whole process, in this way, the development cycle is shortened and the project life-cycle cost is saved. Visualization Technology can combine various data from multiple sources (including table data, discrete sample data, body coordinate data, multiple semi-structured mesh data, and non-structured mesh data) into three-dimensional images.

Computational mechanics is often used in engineering design. Computational mechanics cannot be separated from visualization technology. FEA (finite element analysis) is a numerical method for structural analysis proposed in 1950s. Finite element analysis has been widely used in aircraft design, dam construction, mechanical product design, structural stress analysis, and other fields. From the mathematical point of view, the finite element analysis divides the research object into several sub-units, and then obtains the approximate solution of partial differential equations. In finite element analysis, visualization technology can be applied to achieve grid division of the shape and graphic display of the results of finite element analysis, that is, the so-called before and after processing of finite element analysis, and according to the analysis results, optimize grid division to make the calculation results more reliable and accurate.

Aircraft, automobiles, and ships must be designed to achieve excellent performance and normal operation in gas and liquid high-speed motion environments. In the past, we put the designed aircraft model in a large wind tunnel or water hole for a physical simulation experiment of fluid dynamics, and then modify the design according to the experiment results. This approach not only consumes money but also extends the design cycle. At present, the computational fluid dynamics (CFD) has been simulated on a computer ). The core of this solution is to solve the partial differential equation that represents fluid flow. At present, supercomputer can be used to solve the equations of the complex geometric model by the web-stochastics. Finally, we can calculate the values of various parameters in the flow field at each time point, but the data volume is very large. To understand and analyze the simulation results of fluid flow, visualization technology must be used to dynamically display data on the screen. For example, the velocity, pressure, temperature, and composition of each point are displayed in different ways, and eddy current, shock wave, shear layer, wake and turbulence are displayed. In the visualization of the flow field, it is necessary to increase the display speed, and display the subtle structure of the flow field and the isosurface of various parameters in a realistic manner. Of course, computational fluid dynamics are the same as finite element analysis. The speed and accuracy of computing are greatly affected by grid division. Through visualization technology, you can find the most suitable grid division method for different objects. NASA's ams' aerospace digital simulation device (NAF) not only uses Visualization Technology for CFD computing, it is also used to reconstruct the three-dimensional flow field from the two-dimensional image obtained from the wind tunnel test, and compare the calculation results with the test results. In particular, they built a "virtual wind tunnel" by using virtual reality technology based on high Three-dimensional Interaction Characteristics, providing an intuitive research environment for analyzing the complex structures in various non-Constant flows.

Iv. Development of Information Visualization Technology

In recent years, information visualization has been raised internationally. In general, scientific computing visualization refers to the visualization of spatial data fields, while information visualization refers to the visualization of non-spatial data. With the advancement of social information and the increasingly wide application of network, the information source is becoming increasingly large. In addition to the need to store, transmit, retrieve, and classify massive data, it is imperative to understand the relationships and development trends between data. In fact, behind the surge in data, there is a lot of important information hidden. People want to analyze it at a higher level to make better use of the data. The current database system can efficiently implement data input, query, statistics, and other functions, but it cannot discover the relationships and rules in the data and predict future development trends based on the existing data. On the other hand, artificial intelligence has made significant progress since its birth in 1956. The current research focus is machine learning. Machine Learning is a science that uses computers to simulate human learning. mature algorithms include Neural Networks and genetic algorithms. The database management system is used to store data, and machine learning is used to analyze and mine the knowledge behind a large amount of data. The combination of the two facilitates "Knowledge Discovery in databases (KDD: knodge DGE discovery in databases. In fact, KDD is a cross-disciplinary discipline that involves machine learning, pattern recognition, statistics, intelligent databases, knowledge acquisition, data visualization, high-performance computing, expert systems, and other fields. KDD can be used in information management, process control, query optimization, scientific research, decision support, and data maintenance.

Data mining is the core technology of KDD ). It is from a large number of incomplete, noisy, fuzzy, and random data, the process of extracting potentially useful information and knowledge hidden in it that people do not know beforehand. People regard raw data as the source of knowledge, just like mining from ore. Raw data can be structured, such as data in relational databases, or semi-structured, such as text, graphics, image data, or even different configuration data distributed on the network. The data mining method can be mathematical or non-mathematical. It can be deductive or inductive. Data mining can be used to discover various types of knowledge, including generalized knowledge that reflects the common nature of similar things and feature knowledge that reflects various aspects of things; differential knowledge that reflects the property differences between different things; dependent or associated knowledge that reflects the dependency between one thing and other things; Predictive knowledge that predicts future data based on current history and current data; reveal the deviation knowledge of things from conventional anomalies. To discover these different types of knowledge. A variety of knowledge discovery tools are required. In order to make the process and results of knowledge discovery easy to understand and perform human-computer interaction in the process of knowledge discovery, a visual method of knowledge discovery should be developed. To understand the relationship between data and its development trend, people can turn to visualization technology. Information Visualization not only uses images to display multi-dimensional non-spatial data, but also enables users to better understand the meaning of the data. It also uses visual images to guide the retrieval process and accelerate the retrieval speed. In scientific computing visualization, the objects displayed involve space data of different categories such as scalar, vector, and tensor. The research focuses on how to display 3D data fields in a real and fast manner. In information visualization, the displayed objects are mainly multi-dimensional scalar data. Currently, the focus of research is, the design and selection of display methods can help users understand the massive multi-dimensional data and their relationships, which are more related to psychology, human-computer interaction technology, and other issues.
Information Visualization has broad application prospects in business, finance, communication, and other fields. In the communication field, on the one hand, more sophisticated and advanced network models are being developed to assist in future planning. On the other hand, more complex transmitting and switching devices provide greater freedom and flexibility for current network reconstruction, but the raw data running on a single network unit is constantly increasing. To optimize the operation of all networks, it is necessary to use all these signal sources effectively, and between different traditional fields such as market, network planning and daily management, exchange information and ideas dynamically. Covering physical networks is a broad field that includes sound, data, and image services, each of which has its own data and management requirements. In addition, modern networks are not restricted by national borders. They are an international structure covering many countries and carriers. Therefore, their potential data volume and complexity increase by a larger order of magnitude. Information visualization technology has been fully applied to BT networks. The Network has more than six thousand switching devices and more than 25 million customer lines, resulting in several megabytes of network status and control data per minute. In the BT network, about 60 thousand local paths connected to the digital switch are reported to the central operating unit (ceutral Operations Unit) every five minutes ), the central operating unit then uses these numbers for real-time network monitoring and control. By measuring a large number of running parameters, data of over two thousand MB is generated every day. The Graphic Output depicts the geographic distribution of the selected running parameters and the animation in the interval of interest to you. The minimum, maximum, and average values of parameters in each region can be expressed in a color bar chart. Visualization has aroused wide interest in non-spatial data, such as its application in financial indicators or liquidity statistics. Many visualization tools and technologies for engineering and scientific applications can be quickly transferred to Finance and Statistics. The key to successful visualization applications is its ability to provide users with interactive research data and reveal trends, loops, and patterns that are hard to reveal using other methods. A typical example of application in a non-spatial data range is network statistics, which includes recording the characteristics, switches, large areas, or geographical groups of a single network unit. In addition, cityscape visualization is also a potential useful technology .. The city scene is an extended 3D bar chart, in which the scalar value in the 2D domain is represented as a block or building on an even grid. Visualizes the call failure rate statistical data that is divided into ten geographic bands for each month of the year. BT has used the city scene application to investigate the service statistics by area on a monthly basis and the operating performance of the transmission system. These applications can be easily used for financial information, such as stock gains for each region and period, or visualized mining by geography and revenue ), shows the total amount of money consumed by each branch in terms of currency circulation, total income, and cash operation. For example, the Central Bank of Italy has used the total amount of money developed by Italian artificial intelligence software companies to detect illegal activities through the banking system from abnormal phenomena. Information Visualization provides near-real-time solutions to the increasingly significant "data overload" problem, which will have an important impact on information management in the business, financial and communication fields. It can be seen that the increasing amount of data and information is useful, and the key is to give us useful knowledge as soon as possible.

5. Vigorously Promote the Development of Visualization Technology in China

The Research of scientific computing visualization technology in China started in early 1990s S. Because data visualization processes a large amount of data and the algorithms used to generate images are complex, mega computers and high-end graphics workstations are often used in the past. Therefore, data visualization has been applied to national research centers, high-level universities, and research and development centers of large companies. In recent years, with the improvement of PC functions, the development of various graphics and graphics cards and visualization software, visualization technology has been extended to scientific research, engineering, military, medical, economic and other fields. With the rise of internetr, information visualization technology is in the ascendant. In 1980s, China began to research and apply scientific computing visualization technology. So far, China has achieved a large number of gratifying achievements in algorithm, oil and gas exploration, meteorology, computational mechanics, medicine, and other fields. However, in general, there is still a considerable gap with foreign advanced levels, especially in commercial software. Therefore, it is imperative that organizations develop visual business software and gradually mature it through market competition.