The purposes of visualization are to analyze, understand and communicate information contained within computed data. Over the years a number of computer graphics and imaging standards have been developed, but it is important to realize that there is no single visualization tool or package that can ever meet everyone's needs. Even within a given discipline people view objects and concepts in a very personal way. Furthermore, especially in research activities how an object is viewed or represented will evolve as more knowledge and understanding is accumulated.
Therefore, it is important to have visualization tools that are flexible and interoperable with other tools as well as a variety of changing applications. These tools must be able to provide methods to accept a wide variety of data types and file formats, and to provide most efficient conversion services. Since many visualization techniques and processes are comprised of multiple pipe-lined steps, automatic data typing and conversions are highly desirable.
The Amira visualization package performs well or excels in the above areas. It is a toolbox with some similarities to earlier packages like AVS, Khoros, IRIS Explorer and IBM Data Explorer. Each of these packages are based on a data flow execution model with visual programming interfaces. The user constructs a pipe line or network of pipelines by selecting and dragging various computational and display modules to a special screen where their inputs and outputs are connected to specify the flow of both data and parameters. Execution flow is implicit based on the flow of data through the network. This allows for incremental execution. If data or parameters are altered in the middle of the module network, only the downstream modules need to re-execute.
While there are many advantages to such systems, and especially so for visually oriented applications, there can also be significant complexity and resource overhead. User interaction with the downstream modules is particularly difficult to communicate to the upstream modules (e.g. mouse picks in the display and retrieving the input unprocessed data)
Perhaps the greatest complexity for the user is coming to terms with
all of the various modules, each with their own data type requirements
and restrictions, their own parameters and their own algorithms.
For example AVS has more than 700 user contributed modules in addition
to the several hundred that are part of the product. Often modules
perform only very primitive actions, so constructing a complicated visual
application can get to be very complex. And higher level modules
often have a large number of parameters that may be less than intuitive
even for experienced users.
Modules in Amira are generally higher level than in other systems. Most modules 'know" how to operate on a variety of data types automatically. They also "know" which display, compute and measurement methods are appropriate for which kinds of data objects. And, data objects "know" which kinds of editors can operate on them. Thus it is possible to load a data object, perform several different computations on it and display it in several different ways simultaneously without the user ever having to decide whether or not a specific module will work in a given case ,or whether there is a data type issue, or whether the right connections are made. Indeed, disallowed options are not even presented.
Having said this, it is possible to go farther still and select other modules from the general suite (not just the methods the data objects "know" about) and attempt to make incompatible connections, but this generally happens only after the user has some experience with the system and is attempting to do more complex work, which is the opposite of other packages where new user has to learn many concepts and procedures just to get a very simple start. Overall, the users are most often able to very quickly begin using Amira and get reasonable result with very minimal start up time.
This information is available in alternative formats upon request by individuals with disabilities. Please send email to alt-format@msi.umn.edu or call 612-624-0528. |
||||||||||||
|
HOME |
QUESTIONS |
FEEDBACK |
||||||||||||
|
||||||||||||