(See related pages)
The primary focus of this text has been to explore ways that geometric information is displayed graphically. Techniques in the text have ranged from orthographic multiviews to pictorials to the rendering and animation of 3D models. This chapter looks at ways that all kinds of scientific and technical information — both geometric and non-geometric — can be displayed graphically. Non-geometric information such as temperature, pressure, and torsion is generated as part of the analysis phase of the design process. This type of information is also generated as part of many other activities of engineers and technicians in the course of their jobs. The graphic display of data of all kinds is a powerful tool both as part of a personal exploration towards a solution and as a means of communicating your ideas to others.
The chapter begins by placing data visualization into the context of engineering design. The types of data are defined as are the basic methods for displaying them. Rules for the effective presentation of data are outlined emphasizing the need for applying human perception capabilities and good design principles. The data presentation methods include both hand techniques and computer-based methods. The chapter concludes with a look at a number of technical and engineering fields employing cutting edge computer technology and visualization techniques.
By presenting objects in primarily pictorial projections and using a variety of rendering and presentation techniques, products can quickly and easily be identified and their functionality understood. Traditionally, the pictorial projections were created using manual drafting methods and the rendering confined mostly to line drawing techniques. More recently, computer graphics and CAD tools have allowed technical illustration techniques to make use of 3-D modeling, rendering, and animation and software. Today technical illustrators have at their disposal a wide range of techniques to clearly and concisely communicate to professionals and lay people product functionality. This chapter introduces many of these traditional and contemporary approaches.
|
|
|
DATA VISUALIZATION IN ENGINEERING AND DESIGN |
|
|
|
| 21.1 This section establishes the context of technical data visualization in the analysis phase of engineering design. Though this section is presented in a fairly abstract way, there are numerous real world examples you can draw on to get this material across. The idea of designing a reactor vessel and exploring the relationship between temperature and pressure is used in figures 21.2, 21.3, and 21.10. This example could have just as easily been the deflection of steel beams in a roof construction under a prescribed load (civil engineering), vibrations resulting from varying the rotational speed of a compression pump (mechanical engineering), or the change in assembly line output caused by varying the cycle time on a circuit board installation (industrial engineering). |
|
|
|
DATA VISUALIZATION ELEMENTS |
|
|
|
| 21.2 This section begins the development of the taxonomy of visualization methods. Figure 21.5 lays out the structure of the taxonomy and the structure of Sections 21.2 and 21.3. |
|
|
|
| 21.2.1 Though engineering students have been working with numbers intensively for years they may, in fact, have had little or no exposure to the terms used to describe data types. Many of these terms are more likely to show up in a statistics text than an engineering text. One of the goals is merging techniques which have been used in the social sciences — and to some degree in the hard sciences — with engineering analysis. Of less interest is the methodology than the data manipulation and visualization techniques. |
|
|
|
| 21.2.2 Marks are at the heart of the perceptual basis of data visualization. What largely distinguishes one type of visualization from another is the type of marks it uses. Equally important are the subtle variations that can be applied to the marks by the user. Section 21.4 shows examples of how marks can be manipulated to more effectively communicate the information in the visualization. Good and bad usage of marks is a great way to demonstrate the power of these perceptual cues Glyphs are only mentioned in one paragraph but offer considerable potential for students to apply perceptual principles in novel and creative ways. Problems 21.1 and 21.4 are a couple of ways glyphs and compound marks can be used to expand student's knowledge in this area. . |
|
|
|
| 21.2.3 Though independent and dependent variables were alluded to in the first section of the chapter, they are formally introduced here. The design analysis problem in Figure 21.2 is also formalized with a visualization showing the relationship of an independent and dependent variable. The visualization types presented in Section 21.3 are broken down by the number of dependent and independent variables. This was done to assist students in matching visualization techniques with the type of data they are/will be working with. |
|
|
|
VISUALIZATION METHODS |
|
|
|
| 21.3 This section presents the different types of commonly used visualization techniques. By and large, the most common ones are presented in the earlier sections with the lesser well known ones presented later. Like all computer tools, proper application of these techniques means more than knowing the commands in the software package. In this case, a good grounding in perceptual and design issues is important for proper application of all types of visualizations. It is noted that graphs, charts, and plots are all common terms for types of visualizations. There is little consistency in how these terms are applied and the chapter tries to stay with the most common usage. |
|
|
|
| 21.3.1 This section covers the most common types of visualizations: line and bar graphs. This section also introduces the student to a design theme that goes throughout the chapter: less is more. Test will be — just as with any engineering graphic — what information is being conveyed and are there graphic elements to support each of these pieces of information. There are numerous parallels between the impact of CAD in engineering drawings and the effect of visualization packages on graphs and plots. The section on line graphs introduces the concept of regression lines and error bars. Though a full appreciation of these techniques requires some background in statistics, students can still be introduced to them so they know what they are seeing when they run across examples of them. Quite a few examples of bar graphs are given, in part, because of their popularity and flexibility in presenting different types of data. It is probably worth spending a few minutes making sure the students have a basic understanding of the different types of bar graphs and their appropriate usage. |
|
|
|
| 21.3.2 Students are also likely to encounter analytic data with two independent variables. They are also much less likely to attempt visualizing this type of data, not so much because of a lack of tools to do so as a lack of understanding as to how to do so. Multiple line graphs are a very powerful technique and allow the integration of a much higher density of data than a single line graph. One of the biggest pitfalls of this technique is choosing an appropriate encoding technique for the second independent variable representing the different lines on the graph. Remind students that they must keep in mind the different medium in which the visualization will be displayed. Will there be color? How big will the graphic be in relation to the distance it will be seen at? What is the resolution of the medium; will fine detail be lost? As mentioned in the section, time is a popular second independent variable and animation techniques are a powerful method of displaying it. Even if the students do not have the opportunity to explore animation techniques as part of a lab, try developing a demo which compares the static presentation of sequential visualizations and the same data presented as an animation. It is also worth looking at the difference in presenting the sequential visualizations in parallel (i.e. all in a single graphic image) or serially (i.e. as a slide show). Each technique has its own advantages and disadvantages which is worth imparting on the students. 3-D graphs and plots represent the first use of the third dimension in the chapter. This and other perceptual cues such as color and hidden line removal (occlusion) are all important considerations in designing such visualizations. This is an example of a visualization that was rarely used before the advent of computer-based visualization tools. It is now a standard tool in many packages and is very effective when used properly. Area rendering is not very common in engineering but has become a very popular technique in other areas such as medicine and the earth sciences. Images created by medical professionals, such as X-rays and CAT scans, lend themselves to encoding as digital renderings as do the satellite images used by geologists and meteorologists. Much more in-depth information on 2-D image generation and manipulation can be found in many of the references at the end of the chapter. Vector and flow representations are less common than some of the other areas but have considerable potential for application in areas such as aerodynamics and fluid dynamics. |
|
|
|
| 21.3.3 Use of visualizations to depict functional relationships is more common in some areas of engineering than others but is an important tool for all engineering students. For example, industrial engineers are often trying to express data concerning systems level relationships in a manufacturing industry. This type of qualitative data lends itself nicely to flowcharts, hierarchical trees, etc. In as much the current trends in the area of concurrent engineering forces all engineers to understand system level dynamics, it is worthwhile for them to know how to graphically express these relationships. |
|
|
|
OBJECT RENDERING |
|
|
|
| 21.4 Increased use of 3-D CAD systems has meant more options for displaying geometry besides as black line drawings. It is important to keep the end goal of communication in mind and make sure that whatever technique is being used clearly conveys the critical information. |
|
|
|
| 21.4.1 The rendering pipeline can be seen as a logical extension to where Chapter 7, 3-D modeling, left off. Once the geometry of the model has been created, how can it's visual characteristics be altered to better depict the features/information of interest? It is important to make the link between the coordinate systems presented in Chapter 7 and the coordinate systems used in this section. The most important transition to make is between the world coordinate system representing the object in space and the view coordinate system representing the relationship of the viewer (you) and the object. Though the section talks primarily about this view-based coordinate system, the student needs to be able to map the relationship between these two coordinate systems and understand the difference between the two. |
|
|
|
| 21.4.2 After establishing a viewpoint, the next stage in the rendering pipeline is typically hidden surface determination. This is also the most likely tool available in a 3-D modeling package. Though it is not mentioned in this section, it is worth quizzing your students on the relationship between hidden surface determination and the different types of 3-D models. For example, why can't you perform hidden surface calculations on a true wireframe model? (There are no surfaces to do any calculations on!) Probably the single most important lesson to learn from this section is that there are a considerable number of calculations to be performed to display a model with its hidden features removed. Although increased computer performance has made this a less critical factor, they need to understand the relationship between the type of modeler being used, the number of faces contained in the model, and the time it takes to do the calculations. |
|
|
|
| 21.4.3 Light definition is a topic which physical props can be of great help. Fluorescent room lighting works great at representing ambient lighting, the sun (or a single incandescent light source far away from the model) for an infinite light source, a single incandescent light bulb for a near point light source, and a flashlight for a spot light source. Make sure to emphasize key variables which distinguish one type of light source from another. Does the light fall off in intensity as it gets farther from the surface? What are the orientation of the light rays to each other?, what controls are available to the user to modify these types of light sources? Though defined light sources are usually white by default, it may be worth discussing the effects on shading if the light source is a different color. This can be discussed after the rest of Section 21.3 is covered. |
|
|
|
| 21.4.4 Though there is a little bit of math involved in this section, the principles are fairly intuitive. In explaining the cosine law, you can relate it back to the foreshortening of a surface as the angle between the surface and the line of sight deviates from 90 degrees. In this case the viewer is the light source 'seeing' the surface. The less surface the light source 'sees' the less light energy can be imparted on it. Make sure the students understand the difference between specular and diffuse reflection. Note that most surfaces behave with a combination of both types of reflection. Most 3-D rendering programs are capable of at least a couple of different types of shading algorithms (i.e. Flat, Gouraud, etc.) and it would be worth demonstrating this to the students. Depict both curved and rectilinear forms and, if possible, vary the level of refinement on the faceted curved surfaces. For those with interest in delving into more of the algorithmic bases for these different types of shaders can refer to most computer graphics programming texts, including some of the ones referenced in the back of the chapter. There are very clear distinctions as to how Flat, Gouraud, and Phong shading is implemented. |
|
|
|
| 21.4.5 These advanced shading techniques such as ray tracing are less likely to be available on modelers in the schools. If anything, you may have access to a modeler that is capable of shadow casting. This is worth demonstrating. Like the shading algorithms, the basis of these rendering techniques is quite fascinating. You can create a plan view of a scene with multiple objects which are reflective, transmissive, or a combination of both, and have students trace rays of light from multiple light sources through the scene. |
|
|
|
| 21.4.6 Color is an extremely important concept for students to grasp if they are going to make effective use of rendering tools. From both an aesthetic and a human factors (ergonomics) standpoint, understanding the perceptual components of hue, saturation, and value (HSV) are critical. For understanding the computer implementation of color systems, the red, green, and blue (RGB) color system is very important. Explaining the concepts of color is often best done with a combination of theory and practice. The theoretical side is laid out in the RGB and HSV color models and related material in the text. The practical side of color can be experienced with just about any computer graphics program which supports 8-bit color or better. Have students get on the computer and use the color mixing interface of the program to manipulate color by both HSV and RGB. Give students target colors to mix; either swatches on the computer or real objects (color cards from a paint store make great swatches). You can also have them create matrices of color swatches on the computer varying hue on one axis and saturation on the other or, working with a single hue, value on one axis and saturation on the other. Tell the students to have six to eight steps along each axis. Each of these steps should be perceptually the same distance apart; the steps between swatches cannot be measured by HSV value numbers, they must be judged by the student. Most libraries will have books on color in their art or design sections. The Munsell Color System will give examples of the value/saturation matrix just described. If you don't have a computer graphics program available (and even if you do) you can also have students mix these colors using acrylic artist's paints in the subtractive primaries, white, and black. This is very challenging, but also very rewarding. |
|
|
|
| 21.4.7 More advanced computer rendering systems support texture and bump mapping. Even without these computer tools available, you can do a low-tech demonstration by wrapping photocopies or prints of different patterns at different scales around physical models. Make sure the students understand the difference between texture and bump mapping. Texture mapping manipulates the complete color of the surface whereas bump mapping only manipulates the normals (and therefore the value of the surface color). The two are often combined, making it hard to tell which is contributing what to the surface color. |
|
|
|
INFORMATION INTEGRATION |
|
|
|
| 21.5 As with most of the rest of this chapter, this section is best integrated with exercises and labs. At the same time, there may be problems with getting access to software tools which allow you to explore some of these techniques. Though Macintosh computers have long had dominance in the areas of desktop publishing and multimedia, PCs have rapidly caught up. Design an exercise where the students have to produce a user manual for some mechanical product. Students can be broken up into teams, producing the graphics and integrating them into text to explain the product's functionality. |
|
|
|
| 21.5.1 This section covers what traditionally is called desktop publishing. CAD graphics can be exported from their native package and integrated with text and graphics from other sources with the help of these software tools. Interactivity and hypermedia links (discussed in the next section) can also be integrated to extend the range of information linked and integrated with traditional CAD graphics. If your students have a keen interest in this area, there are numerous books on this topic available; some geared for specific software packages. |
|
|
|
| 21.5.2 Animation is another area that could easily be a course in and of itself. Though you can read about animation, it is best experienced directly. You can spend thousands of dollars on a professional package, but there are many simple animation packages available at relatively modest cost. Pay attention to which packages require you to import graphics from another source and which ones allow you to generate graphics from within. The most flexible ones will allow you to do both. You may also find that some visualization software packages have some basic animation capabilities built in. |
|
|
|
| 21.5.3 All that was said for animation also goes for hypermedia. Again, you may be able to arrange a demonstration to show the capabilities of some of the hardware and software tools mentioned in this section. |
|
|
|
SUMMARY |
|
|
|
Data visualization is a powerful tool for communication data resulting from the testing of a technical design model. The type of data collected from the model determines the method used for the display. Central to an effective presentation is the application of good design principles, especially in the use of color, 3-D modeling, and other techniques that make use of human perceptual capabilities. Data presentations are created both with hand and computer tools. Once completed, data presentations are either viewed by themselves, or integrated with text and other graphics. Computer rendering is used to realistically and symbolically add color, shadow, and occlusion to 3-D computer models. In addition, line rendering can be used to add realism and readability. Flow lines, callouts, and spot details all add information pertaining to parts and assemblies. Technical illustrations are typically integrated with text and numeric information as parts of manuals, instruction sheets, and reports. In addition to more traditional page layout methods, computer-based multimedia and hypermedia are being used increasingly as vehicles for integrating and distributing technical information. |