Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userVariable stars identification in digitized photographic data
135
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We identify 339 known and 316 new variable stars of various types among 250000 lightcurves obtained by digitizing 167 30x30cm photographic plates of the Moscow collection. We use these data to conduct a comprehensive test of 18 statistical characteristics (variability indices) in search for the best general-purpose variability detection statistic. We find that the highest peak on the DFT periodogram, interquartile range, median absolute deviation, and Stetsons L index are the most efficient in recovering variable objects from the set of photographic lightcurves used in our test.
rate research
Read More
We present the results of determination of Plutos positions derived from photographic plates taken in 1930 - 1960. Observations were made with Normal Astrograph at Pulkovo Observatory. Digitization of these plates was performed with high precision scanner at Royal Observatory of Belgium (ROB Digitizer). Mean values of standard errors of plate positions (x,y) lie between 12 and 18 mas. The UCAC4 catalogue was used as an astrometric calibrator. Standard errors of equatorial coordinates obtained are within 85 to 100 mas. Final table contains 63 positions of Pluto referred to the HCRF/UCAC4 frame.
Photographic plate archives contain a wealth of information about positions and brightness celestial objects had decades ago. Plate digitization is necessary to make this information accessible, but extracting it is a technical challenge. We develop algorithms used to extract photometry with the accuracy of better than ~0.1m in the magnitude range 13<B<17 from photographic images obtained in 1948-1996 with the 40cm Sternberg institutes astrograph (30x30cm plate size, 10x10deg field of view) and digitized using a flatbed scanner. The extracted photographic lightcurves are used to identify thousands of new high-amplitude (>0.2m) variable stars. The algorithms are implemented in the free software VaST available at http://scan.sai.msu.ru/vast/
We report the discovery of three new variable stars in Indus: USNO-B1.0 0311-0760061, USNO-B1.0 0309-0771315, and USNO-B1.0 0315-0775167. Light curves of 3712 stars in a 87 x 58 field centered on the asynchronous polar CD Ind were obtained using a remotely controlled 150 mm telescope of Tzec Maun Observatory (Pingelly, Western Australia). The VaST software based on SExtractor package was used for semi-automatic search for variable stars. We suggest the following classification for the newly discovered variable stars: USNO-B1.0 0311-0760061 - RR Lyr-type, USNO-B1.0 0309-0771315 - W UMa-type, and USNO-B1.0 0315-0775167 - W UMa-type.
We present a comparison of the Gaia DR1 samples of pulsating variable stars - Cepheids and RR Lyrae type - with the OGLE Collection of Variable Stars aiming at the characterization of the Gaia mission performance in the stellar variability domain. Out of 575 Cepheids and 2322 RR Lyrae candidates from the Gaia DR1 samples located in the OGLE footprint in the sky, 559 Cepheids and 2302 RR Lyrae stars are genuine pulsators of these types. The number of misclassified stars is low indicating reliable performance of the Gaia data pipeline. The completeness of the Gaia DR1 samples of Cepheids and RR Lyrae stars is at the level of 60-75% as compared to the OGLE Collection dataset. This level of completeness is moderate and may limit the applicability of the Gaia data in many projects.
Recently, machine learning methods presented a viable solution for automated classification of image-based data in various research fields and business applications. Scientists require a fast and reliable solution to be able to handle the always growing enormous amount of data in astronomy. However, so far astronomers have been mainly classifying variable star light curves based on various pre-computed statistics and light curve parameters. In this work we use an image-based Convolutional Neural Network to classify the different types of variable stars. We used images of phase-folded light curves from the OGLE-III survey for training, validating and testing and used OGLE-IV survey as an independent data set for testing. After the training phase, our neural network was able to classify the different types between 80 and 99%, and 77-98% accuracy for OGLE-III and OGLE-IV, respectively.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
