-
Efficient Adaptation of Deep Neural Networks for Semantic Segmentation in Space Applications
Authors:
Leonardo Olivi,
Edoardo Santero Mormile,
Enzo Tartaglione
Abstract:
In recent years, the application of Deep Learning techniques has shown remarkable success in various computer vision tasks, paving the way for their deployment in extraterrestrial exploration. Transfer learning has emerged as a powerful strategy for addressing the scarcity of labeled data in these novel environments. This paper represents one of the first efforts in evaluating the feasibility of e…
▽ More
In recent years, the application of Deep Learning techniques has shown remarkable success in various computer vision tasks, paving the way for their deployment in extraterrestrial exploration. Transfer learning has emerged as a powerful strategy for addressing the scarcity of labeled data in these novel environments. This paper represents one of the first efforts in evaluating the feasibility of employing adapters toward efficient transfer learning for rock segmentation in extraterrestrial landscapes, mainly focusing on lunar and martian terrains. Our work suggests that the use of adapters, strategically integrated into a pre-trained backbone model, can be successful in reducing both bandwidth and memory requirements for the target extraterrestrial device. In this study, we considered two memory-saving strategies: layer fusion (to reduce to zero the inference overhead) and an ``adapter ranking'' (to also reduce the transmission cost). Finally, we evaluate these results in terms of task performance, memory, and computation on embedded devices, evidencing trade-offs that open the road to more research in the field.
△ Less
Submitted 22 April, 2025;
originally announced April 2025.
-
Machine Learning for Mini-EUSO Telescope Data Analysis
Authors:
Mario Bertaina,
Mikhail Zotov,
Dmitry Anzhiganov,
Dario Barghini,
Carl Blaksley,
Antonio Giulio Coretti,
Aleksandr Kryazhenkov,
Antonio Montanaro,
Leonardo Olivi
Abstract:
Neural networks as well as other methods of machine learning (ML) are known to be highly efficient in different classification tasks, including classification of images and videos. Mini- EUSO is a wide-field-of-view imaging telescope that operates onboard the International Space Station since 2019 collecting data on miscellaneous processes that take place in the atmosphere of Earth in the UV range…
▽ More
Neural networks as well as other methods of machine learning (ML) are known to be highly efficient in different classification tasks, including classification of images and videos. Mini- EUSO is a wide-field-of-view imaging telescope that operates onboard the International Space Station since 2019 collecting data on miscellaneous processes that take place in the atmosphere of Earth in the UV range. Here we briefly present our results on the development of ML-based approaches for recognition and classification of track-like signals in the Mini-EUSO data, among them meteors, space debris and signals the light curves and kinematics of which are similar to those expected from extensive air showers generated by ultra-high-energy cosmic rays. We show that even simple neural networks demonstrate impressive performance in solving these tasks.
△ Less
Submitted 28 August, 2023;
originally announced August 2023.
-
Existence of inter coupled structural, electronic and magnetic states in Sm$ _{2} $NiMnO$ _{6} $ double perovskite
Authors:
S. Majumder,
M. Tripathi,
D. O. de Souza,
A. Sagdeo,
L. Olivi,
M. N. Singh,
S. Pal,
R. J. Choudhary,
D. M. Phase
Abstract:
Coupling between different interactions allows to control physical aspects in multifunctional materials by perturbing any of the degrees of freedom. Here, we aim to probe the correlation among structural, electronic and magnetic observables of Sm$ _{2} $NiMnO$ _{6} $ ferromagnetic insulator double perovskite. Our employed methodology includes thermal evolution of synchrotron X-ray diffraction, nea…
▽ More
Coupling between different interactions allows to control physical aspects in multifunctional materials by perturbing any of the degrees of freedom. Here, we aim to probe the correlation among structural, electronic and magnetic observables of Sm$ _{2} $NiMnO$ _{6} $ ferromagnetic insulator double perovskite. Our employed methodology includes thermal evolution of synchrotron X-ray diffraction, near edge and extended edge hard X-ray absorption spectroscopy and bulk magnetometry. The magnetic ordering in SNMO adopts two transitions, at T$ _{C} $=159.6K due to ferromagnetic arrangement of Ni-Mn sublattice and at T$ _{d} $=34.1K because of anti-parallel alignment of polarized Sm paramagnetic moments with respect to Ni-Mn network. The global as well as local crystal structure of SNMO undergoes isostructural transitions across T$ _{C} $ and T$ _{d} $, observed by means of temperature dependent variation in Ni/Mn-O, Ni-Mn bonding characters and super exchange angle in Ni-O-Mn linkage. Hybridization between Ni, Mn 3\textit{d}, O 2\textit{p} electronic states is also modified in the vicinity of magnetic transition. On the other hand, the signature of Ni/Mn anti-site disorders are evidenced from local structure and magnetization analysis. The change in crystal environments governs the magnetic response by imposing alteration in metal - ligand orbital overlap. Utilizing these complimentary probes we have found that structural, electronic and magnetic states are inter-coupled in SNMO which makes it a potential platform for technological usage.
△ Less
Submitted 1 May, 2022;
originally announced May 2022.
-
Microscopic insights of magnetism in Sm$ _{2} $NiMnO$ _{6} $ double perovskite
Authors:
Supriyo Majumder,
Malvika Tripathi,
H. E. Fischer,
D. O. de Souza,
L. Olivi,
A. K. Sinha,
R. J. Choudhary,
D. M. Phase
Abstract:
The functional characteristics of double perovskites with unique ferromagnetic-insulator ground state have been controversial due to the unavoidable presence of anti-site disorders (ASDs). Here, we aim to investigate the origin of magnetic ordering on local and global scales in Sm$_{2}$NiMnO$_{6}$ (SNMO) double perovskite system. Different calcination routes are exploited to generate different cat…
▽ More
The functional characteristics of double perovskites with unique ferromagnetic-insulator ground state have been controversial due to the unavoidable presence of anti-site disorders (ASDs). Here, we aim to investigate the origin of magnetic ordering on local and global scales in Sm$_{2}$NiMnO$_{6}$ (SNMO) double perovskite system. Different calcination routes are exploited to generate different cation arrangements in SNMO and the corresponding magnetic configurations are examined using the high energy (E $ \sim $0.3 eV) `hot neutrons', which has helped to overcome Sm absorption as well as to record total (Bragg's+diffuse) scattering profiles with high momentum transfer (Q$ _{max} \sim $24 angstrom$ ^{-1} $). We have observed that the Ni-Mn sublattice adopts long range collinear ferromagnetic $ F_{x}F_{z} $ structure with commensurate $k$=(0, 0, 0) propagation vector, below ordering temperature T $ \lesssim $ 160 K, irrespective of variable ASD concentrations. In addition, the signatures indicating the antiparallel polarization of Sm paramagnetic moments with respect to Ni-Mn network, are noticed in the vicinity of anomalous magnetic transitions at T $ \lesssim $ 35 K. The real space pair distribution function calculations have provided a direct visualization of ASDs by means of broadening in Ni/Mn-Mn/Ni linkage. Employing the Reverse Monte Carlo approach on diffuse magnetic scattering profiles, we have observed the negative spin-spin correlation function which suggests the Ni-Ni antiferromagnetic exchange interactions ranging up to first nearest neighbor distance. These results confirm that the existence of ASDs in cation ordered host matrix leads to competing ferromagnetic-antiferromagnetic phases in a broad temperature range, which quantitatively governs the temperature dependent bulk magnetic observables of SNMO system.
△ Less
Submitted 7 March, 2022;
originally announced March 2022.
-
A sensitivity study of VBS and diboson WW to dimension-6 EFT operators at the LHC
Authors:
Riccardo Bellan,
Giacomo Boldrini,
Daniele Brambilla,
Ilaria Brivio,
Riccardo Brusa,
Flavia Cetorelli,
Marco Chiusi,
Roberto Covarelli,
Vittorio Del Tatto,
Pietro Govoni,
Andrea Massironi,
Leonardo Olivi,
Giacomo Ortona,
Giorgio Pizzati,
Alessandro Tarabini,
Antonio Vagnerini,
Elena Vernazza,
Jie Xiao
Abstract:
We present a parton-level study of electro-weak production of vector-boson pairs at the Large Hadron Collider, establishing the sensitivity to a set of dimension-six operators in the Standard Model Effective Field Theory (SMEFT). Different final states are statistically combined, and we discuss how the orthogonality and interdependence of different analyses must be considered to obtain the most st…
▽ More
We present a parton-level study of electro-weak production of vector-boson pairs at the Large Hadron Collider, establishing the sensitivity to a set of dimension-six operators in the Standard Model Effective Field Theory (SMEFT). Different final states are statistically combined, and we discuss how the orthogonality and interdependence of different analyses must be considered to obtain the most stringent constraints. The main novelties of our study are the inclusion of SMEFT effects in non-resonant diagrams and in irreducible QCD backgrounds, and an exhaustive template analysis of optimal observables for each operator and process considered. We also assess for the first time the sensitivity of vector-boson-scattering searches in semileptonic final states.
△ Less
Submitted 11 April, 2022; v1 submitted 6 August, 2021;
originally announced August 2021.
-
Mapping the magnetic state as a function of anti-site disorder in Sm$ _{2} $NiMnO$ _{6} $ double perovskite thin films
Authors:
S. Majumder,
M. Tripathi,
R. Raghunathan,
P. Rajput,
S. N. Jha,
D. O. de Souza,
L. Olivi,
S. Chowdhury,
R. J. Choudhary,
D. M. Phase
Abstract:
The predictability of any characteristic functional aspect in a double perovskite system has always been compromised by its strong dependence over the inevitably present anti-site disorders (ASD). Here, we aim to precisely map the quantitative and qualitative nature of ASD with the corresponding modifications in observables describing the magnetic and electronic state in epitaxial Sm$ _{2} $NiMnO…
▽ More
The predictability of any characteristic functional aspect in a double perovskite system has always been compromised by its strong dependence over the inevitably present anti-site disorders (ASD). Here, we aim to precisely map the quantitative and qualitative nature of ASD with the corresponding modifications in observables describing the magnetic and electronic state in epitaxial Sm$ _{2} $NiMnO$ _{6} $ (SNMO) double perovskite thin films. The concentration and distribution patterns of ASD are effectively controlled by optimizing growth conditions and estimated on both local and global scales utilizing extended X-ray absorption fine structure and bulk magnetometry. Depending upon the defect densities, the nature of disorder distribution can vary from homogeneous to partially segregated patches. Primarily, the effect of varying B-site cationic arrangement in SNMO is reflected as the competition of long range ferromagnetic (FM) and short scale antiferromagnetic (AFM) interactions originated from ordered Ni-O-Mn and disordered Ni-O-Ni or Mn-O-Mn bonds, respectively, which leads to systematic shift in magnetic transition temperature and drastic drop in saturation magnetization. In addition, we have observed that the gradual increment in density of ASD leads to significant deviation from uniaxial anisotropy character, reduction in anisotropy energy and enhancement of moment pinning efficiency. However, the observed signatures of $ Ni^{2+}+Mn^{4+} \longrightarrow Ni^{3+}+Mn^{3+} $ charge disproportionation is found to be independent of cation disorder densities. This work serves as a basic route-map to tune the characteristic magnetic anisotropy, magnetic phase transitions, and magnetization reversal mechanism by controlling ASD in a general double perovskite system.
△ Less
Submitted 13 January, 2022; v1 submitted 13 September, 2020;
originally announced September 2020.
-
Chemical induced delithiation on LixMnPO4: an investigation about the phase structure
Authors:
Melanie Köntje,
Giorgia Greco,
Giuliana Aquilanti,
Luca Olivi,
Peter Axmann,
Margret Wohlfahrt-Mehrens
Abstract:
Understanding the LiMnPO4/MnPO4 phase transition is of great interest in order to further improve the electrochemical performance of this cathode material. Since most of the previously published literature deals with characterization of chemically delithiated Lix MnPO4, the aim of this study is to compare and study the composition and structure of the different phases that are generated upon chemi…
▽ More
Understanding the LiMnPO4/MnPO4 phase transition is of great interest in order to further improve the electrochemical performance of this cathode material. Since most of the previously published literature deals with characterization of chemically delithiated Lix MnPO4, the aim of this study is to compare and study the composition and structure of the different phases that are generated upon chemical delithiation of LixMnPO4. Bare and carboncoated lithium manganese phos-phates are prepared via a combined coprecipitation-calcination method. Partial delithiation to two different degrees of delithiation Lix MnPO4 (x = 0.24/0.23 and 0.45) for carbon-coated and/or bare materials is achieved using an excess of nitro-nium tetrafluoroborate in acetonitrile. The effect of carboncoating has been also considered. Standard materials characterization with XRD (X-Ray Diffraction) and ICPOES (Inductive Coupled Plasma spectrometry and Optical Emission Spectroscopy) analysis are in accordance with literature data, but further cerimetric analysis revealed serious deviations, showing differences in the degree of delithiation to the average degree of oxidation. A structural characterization of the atomic and electronic local structure of the materials is also ob-tained using XAS (X-ray Absorption Spectroscopy) technique.
△ Less
Submitted 22 June, 2019;
originally announced June 2019.
-
Development and tests of a new prototype detector for the XAFS beamline at Elettra Synchrotron in Trieste
Authors:
S Fabiani,
M Ahangarianabhari,
G Baldazzi,
P Bellutti,
G Bertuccio,
M Bruschi,
J Bufon,
S Carrato,
A Castoldi,
G Cautero,
S Ciano,
A Cicuttin,
M L Crespo,
M Dos Santos,
M Gandola,
G Giacomini,
D Giuressi,
C Guazzoni,
R H Menk,
J Niemela,
L Olivi,
A Picciotto,
C Piemonte,
I Rashevskaya,
A Rachevski
, et al. (8 additional authors not shown)
Abstract:
The XAFS beamline at Elettra Synchrotron in Trieste combines X-ray absorption spectroscopy and X-ray diffraction to provide chemically specific structural information of materials. It operates in the energy range 2.4-27 keV by using a silicon double reflection Bragg monochromator. The fluorescence measurement is performed in place of the absorption spectroscopy when the sample transparency is too…
▽ More
The XAFS beamline at Elettra Synchrotron in Trieste combines X-ray absorption spectroscopy and X-ray diffraction to provide chemically specific structural information of materials. It operates in the energy range 2.4-27 keV by using a silicon double reflection Bragg monochromator. The fluorescence measurement is performed in place of the absorption spectroscopy when the sample transparency is too low for transmission measurements or the element to study is too diluted in the sample. We report on the development and on the preliminary tests of a new prototype detector based on Silicon Drift Detectors technology and the SIRIO ultra low noise front-end ASIC. The new system will be able to reduce drastically the time needed to perform fluorescence measurements, while keeping a short dead time and maintaining an adequate energy resolution to perform spectroscopy. The custom-made silicon sensor and the electronics are designed specifically for the beamline requirements.
△ Less
Submitted 12 January, 2016;
originally announced January 2016.
