Short period variable stars recognition by using MVA methods in Pi of the Sky experiment

Pi of the Sky is a system of wide field-of-view robotic telescopes to search for short timescale astrophysical phenomena, especially for prompt optical GRB emission. The system was designed for autonomous operation, monitoring a large fraction of the sky with limit magnitude in the range 12^m-13^m and time resolution on the order of 1-100 seconds. LUIZA is a dedicated framework developed for efficient off-line processing of the Pi of the Sky data, implemented in C++. The photometric algorithm based on the All Sky Automated Survey (ASAS) photometry was implemented in LUIZA and compared with the algorithm based on the pixel cluster reconstruction and simple aperture photometry algorithm. Optimized photometry algorithms were then applied to the sample of test images, which were modified to include different patterns of variability of the stars in training sample. Different statistical estimators were considered for developing the general variable star identification algorithm. The algorithm will be used to search for short-period variable stars in real data.