A branch of science that covers the study and analysis of all extraterrestrial objects and their phenomena.
- Origin
- Evolution
- Functions
- Observational astronomy (OA)
- Human eyes
- Telescopes
- Radio
- Micrometer (e.g. double stars)
- Spectrograph (e.g. redshift)
- Photoelectric photometry using Charge-coupled Device (CCD), which can record the image nearly down to the level of individual photons.
- Neutrino astronomy
- Gravitational wave
- Virtual observatory (VO)
Astronomy is divided into many subfields, such as galactic astronomy, planetary science, extragalactic astronomy, stellar astronomy, solar astronomy, and cosmology. In general, the theoretical and the observational.
The purpose of observational study is to observe, record, and collect the data about the universe under study and theoretical scientists mainly calculate the measurable consequences of physical models.
Theoretical astronomers use the collected data to generate the simulation model, and the corresponding observations serve the purpose of evaluating the model or indicating the need for tweaking them.
With ultra-modern technology, the astronomical data collection has been very simple, and rate is very high. And in astronomy, there are “4Vs” – volume, variety, velocity, and value.
- Volume
- data size – can be PB, EB, ZB.
- Variety
- complex elements – signals, images, videos, spectra, time series, and simulations.
- Velocity
- rate of production and transmission – sizeable synoptic survey telescopes (LSST) 20 TB per night for ten years.
- Value
- high value to the astronomy of the data.
- One-dimensional information in the form of signals;
- Two-dimensional information in images;
- multispectral (8-10) and hyperspectral (100+).
- From electromagnetic (EM) emissions
- Image data
- Spectral data
- Three-dimensional information in the video?;
- Time series (GW).
- Electromagnetic Spectrum
- From $1$ Hz to $1025$ Hz, From km to atom size.
- Stellar Classification [cite:@jing-minNewStellarSpectral2020;@chiuSearchingYoungStellar2021]
- Most potential applications
- O, B, A, F, G, K, and M. (O and M represent the hottest and coolest types)
- Pulsar Detection and Recognition (Time series, intensity and time)
- Star / galaxy separation/classification and information analysis [cite:@hausenMorpheusDeepLearning2020]
- Shape and size
- Transient Analysis (暂现源, Fast ratio burst (FRB), gamma-ray burst, pulsar, gravitational wave[cite:@zhangDetectingGravitationalWaves2022], and other transient phenomena)(FAST)
- Astronomical survey analysis (e.g. Gaia survey, Active Galactic Nuclei(AGN))
- Other applications
- Detecting gravitational waves from extreme mass ratio inspirals (EMRI) using convolutional neural networks [cite:@zhangDetectingGravitationalWaves2022]
- By: Xue-Ting Zhang, Chris Messenger, Natalia Korsakova, Man Leong Chan, Yi-Ming Hu, and Jing-dong Zhang
- Double White Dwarfs (DWDs)
- Massive Binary Black Holes (MBBHs)
- Stellar-mass Binary Black Holes (sBBHs)
- Extreme mass ratio inspirals (EMRIs)
- Stochastic gravitational-wave background (mHz frequency band)
- Teukolsky-based waveform and Numerical Kludge (NK) waveform.
- Analytic Kludge (AK) model, through post-Newtonian equations (max 4.5 now)
- Augmented Analytic Kludge (AAK). Accuracy similar to NK with the generating speed of AK
Ground noise affects accuracy, and TianQin is in space and can accurately detect gravitational waves.
Two categories. One can express the data
-
$d(t) = h(t) + n(t)$ , if signal is present -
$d(t) = n(t)$ , if there is no signal.
- Input
- Simulation data for TianQin, using AK and AKK.
- 7864320 seconds (three months)
- 1/30 Hz
- 262144 size
- Morpheus: A Deep Learning Framework for the Pixel-level Analysis of Astronomical Image Data [cite:@hausenMorpheusDeepLearning2020]
- By: Ryan Hausen and Brant E. Robertson.
- Source detection,
- Source segmentation,
- Morphological classification
A U-Net.
- Input
- astronomical FITS images
- Output
- types – spheroid, disk, irregular, point source/compact, and background.
