[vsnet-chat 4061] MISAO Project with PIXY System 2 starts

[Seiichi Yoshida <comet@aerith.net>]

MISAO Project Announce Mail (February  7, 2001)

Hello. I am Seiichi Yoshida working on the MISAO project.

At the beginning of the 21st century, the MISAO Project begins to
operate all of the activities with the PIXY System 2.

The PIXY System 2 is still on the way of development. Many functions
are not yet implemented. Some functions are implemented but no
graphical user interface (GUI) is not yet provided. However, some of
the MISAO Project activities, as enumerated below, which have been
operated using the old PIXY System 1, are ready to run also using the
PIXY System 2.

o Examine the image and obtain the image information, such as the
  R.A. and Decl. of the center, the field of view, the limiting
  magnitude, etc.

o Measure the R.A. and Decl., and the magnitude of all stars in the
  image field.

o Compare overlapping images and discover new variable stars.

Unfortunately the database is not yet ready, we cannot operate the
following activity still now using the PIXY System 2. 

o Measure the magnitude of known variable stars and report to the
  VSNET (Variable Star Network) and VSOLJ (Variable Star Observers'
  League in Japan).

Starting activities with the PIXY System 2, we examine not only the
new images offered to the project after this using the new system, but 
also re-examine all of the past images using the PIXY System 2 which
have been examined using the old PIXY System 1.
There are two reasons for the decision. One is, all of the data format 
is changed in the PIXY System 2. Another one is, the examination
result by the new system is refined better than that by the old
system.

The PIXY System 2 adopts the XML (eXtensible Markup Language) as the
data storage format. In the old system, the examination result is
saved in the PXF format, the original format of the PIXY System. The
PXF format has some disadvantages, such as lack of uniformity, no
guarantee for invertibility between the data on memory and data stored
in the file, and so on. The XML solves these problems and increases
the availability. The data access from the GUI becomes simple and
easy.

Other files in the database of the known variable stars or the
database of the measured magnitude, and the batch files to operate the 
examination in batch mode, are also in the original format, different
formats one another. But in the PIXY System 2, everything is recorded
in the XML format.

The contents in the data are also reinvestigated. For example, the
catalog name of the reference stars for photometry and the CCD chip
name are also attached to the magnitude data, although only the filter 
code is attached in the old system.

The examination process of the PIXY System 2 is also refined in the
following two points.

1. Measure the true center of a luminous blooming star.

2. Restrict the too much deblending of a star.

In the old system, the position of a blooming star is terribly
shifted. Therefore all the blooming stars are listed up as "new
objects" because there are no data at the measured position in the
star catalogs. This problem was solved by applying the blooming cancel 
filter before examination. 

In the old system, a star is sometimes deblended into several parts
and detected as several stars, especially in case of out-of-focus
images. In this case, the deblended parts are listed up as "new
objects" because there are no data at the measured position in the
star catalogs. The PIXY System 2 restricts to the deblending, both in
case the whole image is out of focus and only part of the image is out 
of focus.

These two improvements enables another kind of examination for the
offered images.

Many stars not recorded in the star catalogs are captured by the CCD
images. So it is impossible to search new objects by only comparing
to star catalogs. However, when a very bright star, about 6 mag or 7
mag star for example, is found on the image and not recorded in star
catalogs, it is probably a real new object like a nova.

Using the old system, many numbers of stars are listed up as new
objects by mistake, so it was impossible to pick up real bright stars
not recorded in the star catalogs. But using the PIXY System 2, the
number of false alarms are much reduced. So now it is possible to
check all of the candidates listed up by the system.

As a result, now we can at least check whether a very bright new
objects are in the offered images or not even if there are no past
images of the same field. On the other hand, many of the offered
images had not been checked at all when there were no other
overlapping images before.

The success ratio of the image examination by the PIXY System 2 is a
bit increased. For instance, when 1119 images were examined using the
old PIXY System 1 in last autumn, it succeeded to examine 1082 images
of them, so the ratio is 96.6%. On the other hand, when 3737 images
are checked in last one month using the new system, it succeeded to
examine 3699 images, so the ratio is 98.9%.

After all of the past images are re-examined, all of the magnitude
data of variable stars reported to the VSNET and VSOLJ, about 50
thousand data, will be also re-measured and reported again. Now the
magnitude measurement of known variable stars are not ready yet. But
when it is ready, we will discard all of the old magnitude data and 
report the new data.

P.S.
The past MISAO project announce mails are available at:
  http://vsnet.aerith.net/misao/

--
Seiichi Yoshida
comet@aerith.net
http://vsnet.aerith.net/

VSNET Home Page

Return to Daisaku Nogami

---

vsnet-adm@kusastro.kyoto-u.ac.jp