arXiv subjects

Sergey Mirzoyan

Publications and source records attributed to Sergey Mirzoyan.

5 recordsLinked to original sources

arXiv2025

Enhancing Galaxy Classification with U-Net Variational Autoencoders. II. JWST High Redshift Galaxy Sample

Building on our previous work, we apply a U-Net Variational Autoencoder (VAE) framework to denoise galaxy images from the James Webb Space Telescope (JWST) and enhance morphological classification. This study focuses on galaxies observed up to redshift approximately at 8, capturing them at early evolutionary stages where their faintness and structural complexity pose challenges for the traditional classification methods. By mitigating observational noise, our approach enables the identification of morphological features, particularly in distinguishing between disk and non-disk galaxy types. We evaluate the denoising performance using standard image quality metrics and demonstrate that the enhanced images lead to improved classification accuracy across multiple deep learning models. Our analysis of a sample of 292 galaxies up to z=7.69 shows 83 galaxies classified as disk-like by the GCNN model with high confidence, of those approximately 70-80 % are of redshifts greater than 3. These findings suggest that disk-like structures can be prevalent in the early universe. The results highlight the potential of VAE-based denoising as a robust pre-processing step for analyzing high-redshift galaxy populations in ongoing astronomical surveys.

astro-ph.IM↗

arXiv2025

Enhancing Galaxy Classification with U-Net Variational Autoencoders for Image Denoising

AI-enhanced approaches are becoming common in astronomical data analysis, including in the galaxy morphological classification. In this study we develop an approach that enhances galaxy classification by incorporating an image denoising pre-processing step, utilizing the U-Net Variational Autoencoder (VAE) architecture and effectively mitigating noise in galaxy images and leading to improved classification performance. Our methodology involves training U-Net VAEs on the EFIGI dataset. To simulate realistic observational conditions, we introduce artifacts such as projected stars, satellite trails, and diffraction patterns into clean galaxy images. The denoised images generated are evaluated using Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), to quantify the quality improvements. We utilize the denoised images for galaxy classification tasks using models such as DenseNet-201, ResNet50, VGG16 and GCNN. Simulations do reveal that, the models trained on denoised images consistently outperform those trained on noisy images, thus demonstrating the efficiency of the used denoising procedure. The developed approach can be used for other astronomical datasets, via refining the VAE architecture and integrating additional pre-processing strategies, e.g. in revealing of gravitational lenses, cosmic web structures.

astro-ph.IM↗

arXiv2017

Compton Edge probing basic physics at Jefferson Laboratory: light speed isotropy and Lorentz invariance

We propose to study of the light speed isotropy and Lorentz invariance at Jefferson Laboratory by means of the measurements of the Compton Edge using of the Hall A/C existing experimental setup. Methodologically the same experiment has already been successfully elaborated at GRAAL experiment at the European Synchrotron Radiation Facility in Grenoble with 6 GeV electron beam. This Proposal states two goals expected to be reached at Jefferson Laboratory, both on Lorentz invariance: (a) the one-way light speed isotropy testing accuracy, following from conservative evaluations at numerical simulations, to about an order of magnitude better than was GRAAL's; (b) the dependence of the light speed on the velocity of the apparatus (Kennedy-Thorndike measurement) will be traced to an accuracy about 3 orders of magnitudes better than the available limits.

physics.acc-ph↗

arXiv2016

LARES Satellite Thermal Forces and a Test of General Relativity

We summarize a laser-ranged satellite test of frame-dragging, a prediction of General Relativity, and then concentrate on the estimate of thermal thrust, an important perturbation affecting the accuracy of the test. The frame dragging study analysed 3.5 years of data from the LARES satellite and a longer period of time for the two LAGEOS satellites. Using the gravity field GGM05S obtained via the Grace mission, which measures the Earth's gravitational field, the prediction of General Relativity is confirmed with a 1-$σ$ formal error of 0.002, and a systematic error of 0.05. The result for the value of the frame dragging around the Earth is $μ$ = 0.994, compared to $μ$ = 1 predicted by General Relativity. The thermal force model assumes heat flow from the sun (visual) and from Earth (IR) to the satellite core and to the fused silica reflectors on the satellite, and reradiation into space. For a roughly current epoch (days 1460 - 1580 after launch) we calculate an average along-track drag of -0.50 $pm/s^{2}$.

gr-qc↗

arXiv2016

A Test of General Relativity Using the LARES and LAGEOS Satellites and a GRACE Earth's Gravity Model

We present a test of General Relativity, the measurement of the Earth's dragging of inertial frames. Our result is obtained using about 3.5 years of laser-ranged observations of the LARES, LAGEOS and LAGEOS 2 laser-ranged satellites together with the Earth's gravity field model GGM05S produced by the space geodesy mission GRACE. We measure $μ= (0.994 \pm 0.002) \pm 0.05$, where $μ$ is the Earth's dragging of inertial frames normalized to its General Relativity value, 0.002 is the 1-sigma formal error and 0.05 is the estimated systematic error mainly due to the uncertainties in the Earth's gravity model GGM05S. Our result is in agreement with the prediction of General Relativity.

gr-qc↗