This page contains a Flash digital edition of a book.
industrial visualisation


data analysis is visualisation: creating plots, charts or other visual aids to uncover trends and important data that would be virtually impossible to spot by looking at page after page of raw numbers. Scientific visualisation software continues to make it easier to create and analyse data, with the emphasis being on improving ease of use and efficiency, and the ability to handle today’s increasingly large datasets.


The scientific visualisation pipeline There are a number of steps between raw data and a finished visualisation, explains Erik Brisson, associate director of the Scientific Computing and Visualization Group at Boston University, USA. In some cases you may be able to use one tool for the full data-to-pictures process – many people do all of their work, including graphical display, using a single package. In other cases, you may use multiple tools – for instance, you might use different software for each of these tasks.


Produce Input Data Input Data Analyze, Filter, Reformat Prepared Data Apply Sci Vis Techniques Sci Vis Model Data Map to Geometry Computer Graphics Data Render, Postprocess Image Data View Results


The steps in any visualisation project (courtesy of Boston University’s Scientific Computing and Visualization Group)


www.scientific-computing.com


Visual T


he ever-growing amount of data we gather has a downside: finding the real information we want. One tool that has existed since the very beginning of


insights The focus here, he adds, is the application


of scientific visualisation techniques to create a renderable geometric model. In the middle stages of this flow of data through the pipeline lies a conversion from specific models in the science domain to those of the graphics domain. The representations that lie between the science-specific model and the graphics data structure are common to many domains – this collection of techniques/representations/models makes up the heart of scientific visualisation. Many other organisations besides


Boston University provide specialised visualisation services. ‘Many scientists don’t realise that such services exist in their universities or institutions, so they should check into these resources,’ advises David Bock, a visualisation/graphics research programmer at the US National Center for Supercomputing Applications (NCSA). He and his colleague Mark Van Moer, a visualisation programmer, have the sole task of helping researchers at NCSA – and any other organisation belonging to the TeraGrid infrastructure of the National Science Foundation – produce better, more informative images.


A collaborative effort ‘I’ve achieved the most success in a collaborative effort,’ adds Bock. ‘Creating a good image is a design process, and we work with scientists in the same way a graphic designer works with a client to create a poster. In some cases, after an initial representation the scientist discovers that the computations were wrong, or we change colours and textures to look at certain features. We can also show things they didn’t see before, so we’re helping to add insight to the project. Additionally, we have training in graphical representation; we’re educated in colours, textures and camera position. We help determine what’s best in a situation – a surface? Volume? Do we show movement with ribbons or streamlines?’


Scientific visualisation software not only helps clarify what we’ve discovered; a good image can also uncover new insights, as Paul Schreier discovers


Rendering of swirling strength (image courtesy of the Visualization Group within Advanced Applications Support at NCSA)


Bock points to an example where


scientists were studying research on gravity flow with detail that’s impossible to measure in experiments. He visualised the swirling strength using volume-rendering software he’s been working on for many years; software that can handle the 1TB of raw data and the 18TB of processed data from the study. The results (see nearby image) helped the team better understand the dyamics of the vortex interaction in gravity currents. The response he got from the scientist was that ‘we could not see these structures with any off-the-shelf software.’ ‘Many people don’t need the cinematic


quality of Dave’s rendering software for PowerPoints or publications,’ adds his colleague Van Moer, ‘but they do come to us when they have very large files sizes that many programs simply can’t load. They also come to us when they have very complex simulation results in an unstructured format, and here we find tools such as the open source package ParaView very helpful because it can split up large datasets and allow interactive visualisations with its GUI. It’s also interesting that a lot of the work I do is outside visualisation proper. Sometimes I have to write file loaders just to get the data into a standard file format.’


Commercial offerings In many cases, scientists can get all the visualisation power they need through


JUNE/JULY 2011 41


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52