industrial visualisation
commercial software. Most packages include the full sci-vis pipeline with varying degrees of depth on the ends; on the other hand, most data-analysis packages now incorporate visualisation tools. General- purpose numeric programs such as Matlab, IDL and LabView include user interfaces and scripting capabilities for reading and processing data, performing computations and producing visual output in a variety of ways. Indeed, Matlab is one of the most
popular data-analysis packages, and you do most of your work from a command line or by putting together programs from these commands. Realising that it takes some time to become proficient with the program, its developers have also been working on interactive tools that address common workflows – one of them being an interactive visualisation aid. Within it you define the data to work with, and if you know which of the more than 100 plot types you want, you choose it with the Plot Selector tool; if you’re not sure which type to work with, the Plot Catalog tool proves handy. After the plot is generated, a Plot Editor button allows you to tailor the plot. Furthermore, a button push can create the Matlab code for a function that you can then call again for other datasets. For sharing results, the Report Generator records all work including graphics and comments along with the Matlab code so you can distribute it in HTML format, as a PDF, Word document or other formats. In Matlab, the data to be plotted needs
to be in memory, but the software offers several ways to import a portion for analysis and plotting. You can interactively use the Import Wizard tool to read in files such as Excel and select particular columns, and there are also functions to read subsets of data (e.g. select certain rows and columns, start reading data at arbitrary points within a file, read every N points from a variable, etc.). You can also use memory mapping to index into arbitrarily large datafiles, analyse them and plot the results. IDL from ITT Visual Information
Solutions is a data visualisation and analysis programming language, similar in many ways to Matlab but addressing different markets – specifically the environmental sciences and life sciences, and earth- observation markets – according to Beau Legeer, director of product marketing. To support these growing markets,
the company created Envi, their flagship off-the-shelf product written in the IDL language and specifically designed to extract
42 SCIENTIFIC COMPUTING WORLD
The Plot Selector in Matlab allows you to review quickly the types of plots that might be suited for a given dataset
To help discover what’s in a dataset, IDL allows you to interactively rotate the data visualisations and see them from different viewpoints
information from geospatial imagery. Envi can be purchased as a standalone app or with the full IDL language to customise functionality based on specific user needs. As these and other types of datasets grow, IDL and Envi are evolving to handle them and recognise the new available computing
Support for GPUs has been implemented for some customers outside the commercial product, but this capability is planned for the commercial version. First, however, the developers are looking for the widespread adoption of a standard such as OpenCL.
Emphasis on visualisation While the above products have data analysis as their primary focus, a separate group has traditionally placed more emphasis on the visualisation end of the pipeline and added analysis over the years. Consider Origin 8.5.1, a package optimised for creating publication-quality plots and graphs (OriginPro adds extended analysis tools for statistics, 3D surface fitting, signal processing and image processing including surface fitting with multiple 3D peaks). Here, the focus is on ease of use, making it possible to perform graph customisations with a right-click and altering parameters. ‘Users can change things like the number of ticks on an axis or how many labels are
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
David Bock, Visualisation/Graphics research programmer, NCSA
power. IDL can open a dataset of an indefinite size with the ultimate goal of allowing users to get accurate results quickly, without the need for training in computer or image sciences. The IDL programming language has been
optimised for image processing and analysis with commands for ingesting data and performing wavelet-based image-domain techniques. For data import, you can apply powerful methods without looping as is needed in other languages. Although it reads common data formats from NASA and ESI, it also allows you to define a data structure.
on an axis extremely quickly,’ explains Joe Przechocki, business development manager, ‘and while these might seem mundane, many users absolutely require this exacting level of control.’ Version 8.5.1 adds some interesting
capabilities. In the past, if a curve fit was done on raw data and had the wrong fitting factor or you wanted to constrain a parameter, it was necessary to re-do the analysis. Now you can do this interactively and everything updates. It’s also possible to create customised reports, such as with a company logo or certain parameters, and reuse this tool on different datasets; in past versions if you modified the data the graphs would update but not the analysis reports, whereas now standard Origin reports and customised reports update automatically. Origin also provides a number of live
links to programs where users might find analysis more comfortable in that environment but prefer Origin for graphing/ plotting. A palette of sub-VIs for LabView is also offered, making it faster to set up plots and change them. The package also provides LabTalk scripting language plus Origin C, an Ansi C-based language with Origin- aware commands for things like graphs and
www.scientific-computing.com
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