PENGGUNAAN DEEP NEURAL NETWORKS (DNN) DALAM PENGUKURAN PERGESERAN MERAH FOTOMETRIK MENGGUNAKAN CITRA GALAKSI: STUDI AWAL
In the cosmological and extragalactic study, distance to a galaxy is an important parameter, by knowing the distance, we can find any other physical parameter such as mass, luminosity, star formation rate, and metallicity. By applying the specific cosmological model, a distance can be measured fr...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Subjects: | |
| Online Access: | https://digilib.itb.ac.id/gdl/view/35660 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | In the cosmological and extragalactic study, distance to a galaxy is an important
parameter, by knowing the distance, we can find any other physical
parameter such as mass, luminosity, star formation rate, and metallicity. By
applying the specific cosmological model, a distance can be measured from the
redshift. The accurate redshift can only be measured by using the spectroscopic
technique (Doppler effect). However, spectroscopic observation limited to
brighter objects and number of objects in a single field of view (FoV). While
photometric observation can capture fainter objects and more objects in a single
FoV. In this work, we use datasets from SDSS DR10. In the SDSS DR10
numbers of images from photometric observation are about 250 times more
than the number of spectrums from spectroscopic observation.
Measurements of photometric redshift can be done by comparing the
SED curves of the elliptical galaxy with known spectroscopic redshifts from
other elliptical galaxy which we want to find the photometric redshift. Another
method is to do linear or non-linear regression, by assuming the redshift is a
function of magnitude in each bandpass filter.
Therefore, we propose a technique that using full galaxy images in each
measured bands and machine learning method for measuring photometric
redshift. We pass entire multi-band galaxy images into the machine learning
architecture to obtain an estimated redshift. We use DenseNet, which is one of
Deep Neural Networks (DNN) architecture, to recognize unique features from
the
ux of a galaxy per pixel in the single image which represents the redshift.
We note that DNN requires more computing resources and better accuracy
than standard machine learning architecture. |
|---|