The message below from Kirill Sokolovsky is to informs folks that a new version of VaST has been released (http://scan.sai.msu.ru/vast/). "VaST is a program aimed to find objects of variable brightness on a series of astronomical images." I recommend the software, and used it to find numerous variables in images taken at the Abbey Ridge Observatory (AAVSO member Dave Lane's observatory). Unfortunately, as is often the case, we haven't found the time to assemble those findings into a short study. Nevertheless, I think AAVSO members will find the (free) software rather useful.
Dear VaST users,
I am happy to announce that vast-1.0rc65 has just been released.
The changes include:
* Added linear magnitude correction as a function of star position on
a frame. This is useful to correct extinction on very wide field (a few
degrees) images such as those obtained in the NMW survey
( http://scan.sai.msu.ru/nmw/ ). By default, the program will turn on this
correction if the reference image has more than 10000 stars on it. You may
manually turn this correction on (off) by starting VaST with "-j" ("-J")
flag. If you have wide field images, try to experiment with this option
and see what works best for your data.
* The star matching algorithm is now allowed to use saturated and blended
stars for image identification. This slightly degrades the quality of the
initial plate solution, but that is corrected for at a later stage. The
good thing is that this change makes the algorithm much more stable and it
typically requires less reference stars to match images.
The image matching process becomes faster.
* Fixed broken links to WISE-Atlas in the transient detection mode.
* Fixed the lightcurve viewer (./lc) crash when one tries to delete the
first point from a non-VaST-formatted lightcurve.
The new VaST version may be obtained with the following commands:
# Download the archive file containing VaST source code
wget -c "http://scan.sai.msu.ru/vast/vast-1.0rc65.tar.bz2"
# Unpack the archive
tar -xvjf vast-1.0rc65.tar.bz2
# Change directory to the VaST root directory
# Compile VaST
More information may be found at the VaST homepage:
Enjoy the new version and, as usual, please send me your bug reports,
feature requests and notes about your overall impression of the software.
With best wishes,
Sounds like a great automation, sort of a "blink comparator" in software! Before I spend too much time on this, could you tell me if it will work properly in Cygwin (Unix emulator for Windows)? Has anyone successfully used it there?
I played with it a few years ago and intend to do so again. I don't remember if I used Cygwin or Ubuntu (I've used both off and on). On the other hand, it is SO easy to set up a Linux flavor in a virtual machine these days (see http://www.virtualbox.org) that I would say, why bother with Cygwin? VirtualBox is free and there are free Linux 'guest" builds - see, for example, http://virtualboxes.org. I don't have anything against Cygwin per se but, IMHO, perhaps its time has passed?
Cygwin has worked well for me, especially on "light" systems, such as what I may operate remotely at observing siites off of solar power. I think VM's require a more robust system, a lot of RAM especially.
You are going to run VaST on a remote "light" system?
Well, yes thats what I was thinking since I saw this thread. It says its compiled in GNU/GCC which is my normal working dev environment, so it fits well into Cygwin, which will output a single Windows .EXE executable image that will run directly on Windows without any VM added on. My remote system is a low end Netbook, 1 Ghz processor, 2 GB RAM, 300GB HD. I am limited in network bandwidth, so I thought it could do the image analysis on site, and upload the results to me at home, rather than the huge FITS files, would be too time consuming at the low bandwidth.
The new vast-1.0rc68 has just been released. It features a number of bug fixes, mostly related to communication with VizieR and a speed-up of the image matching algorithm. It is also possible now to specify input images in a text file instead of the command line. A time correction can be specified in the input file providing an easy solution to the situation when a computer clock offset is discovered after imaging. (The proper solution for this would be, of course, to edit imaging time recorded in FITS headers, which is not always easy.) More details on the new VaST release may be found here http://scan.sai.msu.ru/vast/#news
Best wishes, Kirill
The updated version of the software is now available at http://scan.sai.msu.ru/vast/ It includes a command-line period search and lightcurve-shuffling (to estimate significance of a detected period) tool as well as a large number of bug fixes (comapred ot the rc68 version reported here).
May I advertise a paper http://arxiv.org/abs/1609.01716 comapring various statistical methods to identify variable objects in a set of lightcurves. VaST implements all the variability indexes discussed in that paper, but only the best ones (sigma of a clipped lightcurve, weighted sigma, MAD, IQR, RoMS, Stetson's J and L, S_B, and 1/eta) are displayed in the interactive tool. Use 'M/N' key on the keboard to switch between the indexes. See the source code if you (for whatever reason) need one of the remaining indexes.
Have fun! :)