After experiencing with many free and some commercial software, the author found that there is a general lack of support or flexibility for geophysical data visualization, in particular in the field of solid earth geophysics. This is probably due to the fact there are many different kinds of data as well as diverse data formats. Currently many people are using the popular GMT software (Wessel and Smith, 1995), since it is well developed and suited to plot various geology and geophysics related figures with high quality outputs in Postscript format. However GMT software has no built-in visual support so its users have to rely on other software (such as Ghostscript) to view the plotting results. Besides if you want to use large scale datasets such as GTOPO30, USGS DEM, satellite images, and continental scale potential field data released by NGDC, you would have to have either some kind of commercial software (if you could afford one) or public domain software to extract and visualize the data. The public domain software are generally released as they are without any warranty and much technical support, and often require users to have some knowledge or experience of the relevant computing system or language to make them work. Although this is usually a drawback, it offers a great potential to further develop into a useful and effective software, as demonstrated by the free software presented in this paper.
Although there are many image view tools available both as free and commercial software, none of those is intended specifically for dealing with a variety of geophysical data formats. The need for the development of a geophysics oriented software is thus derived from the desire of being able to read many kinds of geophysical data formats in a way that could match software industry standard. That is the software should at lease enable users to have a user-friendly tool to visualize data through a mouse-click operation, and to have an option to generate hard copies or image files in commonly available formats so that the files created could be easily exported for the purpose of publication or exchange of data in a visualized form.
The aim of this paper is to notify people who are interested or involved with geophysical data handling and visualization, that a free software has been released recently to the public domain, especially offered to the general geophysics community. The software, named as the extended VISU, is designed for the visualization and interactive manipulation of 2D gridded scientific data. This software is developed on top of the previously released free software Visu-2.0 by Pierre-Louis Bossart (1996), who wrote the original program that performs all the data formatting, color allocation and displaying work. The extended VISU is essentially a Tk extension of the scripting language Tcl/Tk (Ousterhout, 1994). Tcl/Tk is one of the most powerful contemporary software developing languages for writing GUI (graphical user interface) programs, and it is widely used in nearly every domain of graphical and engineering applications.
Currently the software has built-in supports for a number of major geophysical data formats including GMT (or netCDF), NCSA HDF, USGS DEM, GTOPO30, NGDC CD-ROM, Landsat MSS/TM, ASCII text and raw binary. Although the released software does not claim to cover all available data formats, it provides a general reading mechanism to deal with any binary or ASCII text files (see more detailed description in the next section).
The software can be used in a wide application spectrum of geophysical studies. For instance, it can be used to view and process potential field data, and it is well suited to view large scale topographic data with additional capability to extract drainage patterns. It can also be utilized to view satellite images for investigating various landscapes or subtle changes of the Earth's surface.
The important improvements of the extended Visu are: (1) it lets you read many different kinds of geophysical data formats, (2) imaging results can be saved as image files in common formats (GIF, TIF, JPEG, PS, PPM, RAS, TGA, HDF, BMP and PCX) as well as the option to print directly to a printer, and (3) it has many useful built-in processing capabilities including:
It should be mentioned that GMT also has many built-in filters including most ones listed above. The extended Visu has additional filters such as fractal filtering, 3-D analytical signal and reduction to pole.
Several examples of the above mentioned processing results can be found at this web link.
When you start the extended Visu, a main control panel will be created, which looks like the following figure:
The Main Control Panel will show the dimensions and the pixel values of the loaded data when the mouse is entered and moved within the canvas window that displays the data image. On the left side pixel location and input value (either color code or data value) are shown. On the right side the physical location (e.g. longitude x latitude) and data value are shown. Both of these entry values are shown interactively. The botton-right corner box indicates the status of running the extended Visu.
When you start the Visu-3.0, the program will first print out your display visual type and after you load in a data file, it will also print out how many colors available to use in the program. Now there are two new options that control the allocation of colors in the Color menu under Colormap option. You can now specify either having maximum 128 or 64 colors to use so that other programs can use the rest of your X server color resources. This is very useful if you also want to use Tk Photo image widget to statically view 24-bits image, such as the relief shading function in the program described in the user's guide. By default the Visu-3.0 will try to grab as many colors as possible (with a maximum number of 240) since other programs such as Openwin also needs some colors one really can't have 256 colors ! So if you want this option to work, you need to change the colormap definition before going on loading any data file.
In what follows each button/menu function will be described briefly.
The File function has a number of entries as described below:
When you select the Read entry in the File Menu, there are eight types of data that you can choose, i.e. GMT grd /netCDF, NCSA HDF data, USGS DEM, GTOPO30, NGDC CD-ROM, Landsat, ASCII text and raw binary. The first two kinds of data can be loaded in automatically. The others require the file structure information specified through the input data info window as shown below.
The extended Visu provides this mechanism to allow you load any binary data by clicking on one of the xxx (file type) buttons, so long as you first enter the correct file info including the required extract width / extract height (number of points in x / y direction), the x / y begin location, header length in bytes, data type (float, int, short or char), file width (NCOLS), file height (NROWS), data byte order (either MSBF meaning most significant byte first and often called "big endian", or LSBF meaning least significant byte first and often called "little endian" which is generally associated with data storage in DOS machines), and other self-explaining parameters as shown in the figure below.
You can also use the Input Data Info window (shown below) to check the physical size of the input data file structure, since x_min, x_max, x_inc, y_min, y_max and y_inc are automatically updated once a data file is loaded in successfully. The dx and dy (i.e. physical increment distance in metres or km) entries are not required if you only want to view the data.
Once you click on one of the xxx (file type) buttons, a file selector window will show up. You can either highlight the input file or type-in the file name in the space under Section heading, then click on OK button will start the actual file reading process. If the file is ok, a canvas window will show up displaying the data image. The example file information shown above lets you read a GTOPO30 file (a global 30 arc seconds topographic dataset). Here is the image display from the selected data.
The Color function will let you change the colors interactively. This function contains two entries - Colormap and look-up table. Colormap Level changes the setting of the colormap (four options: default, default_128, default_64 and private). Look-up table (LUT) will let you choose one of the pre-defined 12 lookup color tables including rainbow, spectrum, 4, 6, 8 and 12 band HIS (hue/intensity/saturation model) and a couple of gray-scale tables.
The Overlay function in the main control panel will let you do basically a simple paint job as well as put coastline, lake or national boundary onto the image.
The Extract function in the main control panel contains a few graph applications. The extended VISU can extract a profile or a histogram and grab an area power spectrum or fractal parameters.
The Roi function will let you extract a rectangular region of interest (ROI) of the active image. You can also use the Cut/Display command in the Roi window, which will let you cut out a rectangular region of the image without a zoom factor (i.e., zoom factor = 1). This function is useful to save a subregion of the data in a separate image or data file.
The Image pulldown button in the main control panel contains three entries - Pict relief shade, Photo relief shade (both of which generate color shaded relief images) and Topo tool which allows user to view topographic slop or drainage patterns as well as generate 3-D images.
The Filter function lets you perform many applied geophysical filtering processes for a given set of controlling parameters. These filters include low, high and band pass, upward or downward continuation, fractal filter, 2nd vertical derivates, horizontal gradient, 3-D analytical signal, pseudo gravity and reduction to the pole.
The Info function will create a pop up window displaying licence terms first, and if you click on the Next button within the first info window, a second window will pop up displaying the general info about Visu-3.0 (i.e. the extended Visu).
The Help function will fire up html_view with the tutorial. At the moment, this tutorial is pointed at the author's web page. The html_view runs separately from the extended Visu, and it is a simple program based on Tk4.2 wish assumed to be installed in the default directory /usr/local/bin.
This software has so far been successfully tested and installed on a SUN SPARC 10 running Solaris 2.5. Since the core library of the extended Visu is written in C mostly independent of operating systems and its GUI is built with the popular Tcl/Tk which has been successfully ported to all kinds of operating systems, it may be possible to port the extended Visu to other computer platforms. However the author has no such experience yet in doing so due to a limited time available, it is highly recommended if anyone is intended to port the extended Visu to systems other than Sun Solaris it should be worthwhile to check out Tcl/Tk WWW sites (http://www.sco.com/Technology/tcl/Tcl.html, http://www.tcltk.com, http://cuiwww.unige.ch/eao/www/TclTk.html) that provide information on how to compile and install the Tcl/Tk on their preferred systems.
To compile the extended Visu, you need the following additional libraries Tcl7.6/Tk4.2, BLT2.1, GMT3.0, HDF4.1r1, JPEG and TIFF, installed in your system as well as the standard Xlib.
To run the extended Visu, you only need the executable visu and its Tcl lib scripts normally installed under the directory /usr/local/visu/lib/visu. The Tcl scripts control the look and interface of the program, so if you are not familiar with Tcl/Tk, it is highly recommended NOT to edit these script files.
Since this software is built on Tcl/Tk, it can be easily extended if you are familiar with the scripting language Tcl/Tk and being able to write C code. For instance, if you want to add a new process code to the library or use the extended Visu to view your processed data, all you need to do is to write a simple code or add a few lines in your processing code that acts as an interface for the image display of your data. A detailed example can be found at this web link.
For those who want to try out this software, more information can be found at the following internet address:
http://www.geol.ku.dk/zhou/visu-2.1
From this address anyone can download the software free of charge !
Finally by releasing this software to the public domain, the author wishes, it would help users simplify their process of data visualization and facilitate their studies involved with geophysical data handling. At the same time the author hopes to get some responses such as comments or bug reports, so that the software can be further improved or developed in the near future.
The author wishes to ackowledge the helpful comments made by Mike Bevis (EOS Geodesy Editor) and an anonymous reviewer, from which this article has been improved. Stanislaw Wybraniec is thanked for generously providing his code to generate color shaded relief images.
Pierre-Louis Bossart, 1996. Hypertools in image and volume visualization, TCL Workshop 1996, pp.10.
J.K. Ousterhout, 1994. Tcl and Tk toolkit, Addison-Wesley.
P. Wessel and W.H.F. Smith, 1995. New version of the Generic Mapping Tools released, EOS Trans. AGU, vol. 76, p.329.
Return to EES Contents