-
Automatic location detection based on deep learning
Authors:
Anjali Karangiya,
Anirudh Sharma,
Divax Shah,
Kartavya Badgujar,
Dr. Chintan Thacker,
Dainik Dave
Abstract:
The proliferation of digital images and the advancements in deep learning have paved the way for innovative solutions in various domains, especially in the field of image classification. Our project presents an in-depth study and implementation of an image classification system specifically tailored to identify and classify images of Indian cities. Drawing from an extensive dataset, our model clas…
▽ More
The proliferation of digital images and the advancements in deep learning have paved the way for innovative solutions in various domains, especially in the field of image classification. Our project presents an in-depth study and implementation of an image classification system specifically tailored to identify and classify images of Indian cities. Drawing from an extensive dataset, our model classifies images into five major Indian cities: Ahmedabad, Delhi, Kerala, Kolkata, and Mumbai to recognize the distinct features and characteristics of each city/state. To achieve high precision and recall rates, we adopted two approaches. The first, a vanilla Convolutional Neural Network (CNN) and then we explored the power of transfer learning by leveraging the VGG16 model. The vanilla CNN achieved commendable accuracy and the VGG16 model achieved a test accuracy of 63.6%. Evaluations highlighted the strengths and potential areas of improvement, positioning our model as not only competitive but also scalable for broader applications. With an emphasis on open-source ethos, our work aims to contribute to the community, encouraging further development and diverse applications. Our findings demonstrate the potential applications in tourism, urban planning, and even real-time location identification systems, among others.
△ Less
Submitted 16 March, 2024;
originally announced March 2024.
-
Cross-Correlation of Near and Far-Infrared Background Anisotropies as Traced by Spitzer and Herschel
Authors:
Cameron Thacker,
Yan Gong,
Asantha Cooray,
Francesco De Bernardis,
Joseph Smidt,
Ketron Mitchell-Wynne
Abstract:
We present the cross-correlation between the far-infrared background fluctuations as measured with the Herschel Space Observatory at 250, 350, and 500 μm and the near-infrared background fluctuations with Spitzer Space Telescope at 3.6 μm. The cross-correlation between far and near-IR background anisotropies are detected such that the correlation coefficient at a few to ten arcminute angular scale…
▽ More
We present the cross-correlation between the far-infrared background fluctuations as measured with the Herschel Space Observatory at 250, 350, and 500 μm and the near-infrared background fluctuations with Spitzer Space Telescope at 3.6 μm. The cross-correlation between far and near-IR background anisotropies are detected such that the correlation coefficient at a few to ten arcminute angular scales decreases from 0.3 to 0.1 when the far-IR wavelength increases from 250 μm to 500 μm. We model the cross-correlation using a halo model with three components: (a) far-IR bright or dusty star-forming galaxies below the masking depth in Herschel maps, (b) near-IR faint galaxies below the masking depth at 3.6 μm, and (c) intra-halo light, or diffuse stars in dark matter halos, that likely dominates fluctuations at 3.6 μm. The model is able to reasonably reproduce the auto correlations at each of the far-IR wavelengths and at 3.6 μm and their corresponding cross-correlations. While the far and near-IR auto-correlations are dominated by faint dusty, star-forming galaxies and intra-halo light, respectively, we find that roughly half of the cross-correlation between near and far-IR backgrounds is due to the same galaxies that remain unmasked at 3.6 μm. The remaining signal in the cross-correlation is due to intra-halo light present in the same dark matter halos as those hosting the same faint and unmasked galaxies. In this model, the decrease in the cross-correlation signal from 250 μm to 500 μm comes from the fact that the galaxies that are primarily contributing to 500 μm fluctuations peak at a higher redshift than those at 250 μm.
△ Less
Submitted 16 December, 2014; v1 submitted 12 December, 2014;
originally announced December 2014.
-
Lens Models of Herschel-Selected Galaxies From High-Resolution Near-IR Observations
Authors:
J. A. Calanog,
Hai Fu,
A. Cooray,
J. Wardlow,
B. Ma,
S. Amber,
M. Baes,
J. Bock,
N. Bourne,
R. S. Bussmann,
C. M. Casey,
S. C. Chapman,
D. L. Clements,
A. Conley,
H. Dannerbauer,
G. DeZotti,
L. Dunne,
S. Dye,
S. Eales,
D. Farrah,
C. Furlanetto,
A. I. Harris,
R. J. Ivison,
S. J. Maddox,
G. Magdis
, et al. (14 additional authors not shown)
Abstract:
We present Keck-Adaptive Optics and Hubble Space Telescope high resolution near-infrared (IR) imaging for 500 um-bright candidate lensing systems identified by the Herschel Multi-tiered Extra-galactic Survey (HerMES) and Herschel Astrophysical Terahertz Survey (H-ATLAS). Out of 87 candidates with near-IR imaging, 15 (~17%) display clear near-IR lensing morphologies. We present near-IR lens models…
▽ More
We present Keck-Adaptive Optics and Hubble Space Telescope high resolution near-infrared (IR) imaging for 500 um-bright candidate lensing systems identified by the Herschel Multi-tiered Extra-galactic Survey (HerMES) and Herschel Astrophysical Terahertz Survey (H-ATLAS). Out of 87 candidates with near-IR imaging, 15 (~17%) display clear near-IR lensing morphologies. We present near-IR lens models to reconstruct and recover basic rest-frame optical morphological properties of the background galaxies from 12 new systems. Sources with the largest near-IR magnification factors also tend to be the most compact, consistent with the size bias predicted from simulations and pre- vious lensing models for sub-millimeter galaxies. For four new sources that also have high-resolution sub-mm maps, we test for differential lensing between the stellar and dust components and find that the 880 um magnification factor (u_880) is ~1.5 times higher than the near-IR magnification factor (u_NIR), on average. We also find that the stellar emission is ~2 times more extended in size than dust. The rest-frame optical properties of our sample of Herschel-selected lensed SMGs are consistent with those of unlensed SMGs, which suggests that the two populations are similar.
△ Less
Submitted 22 October, 2014; v1 submitted 5 June, 2014;
originally announced June 2014.
-
H-ATLAS: The cosmic abundance of dust from the far-infrared background power spectrum
Authors:
Cameron Thacker,
Asantha Cooray,
Joseph Smidt,
Francesco de Bernardis,
K. Mitchell-Wynne,
A. Amblard,
R. Auld,
M. Baes,
D. L. Clements,
A. Dariush,
G. De Zotti,
L. Dunne,
S. Eales,
R. Hopwood,
C. Hoyos,
E. Ibar,
M. Jarvis,
S. Maddox,
M. J. Michalowski,
E. Pascale,
D. Scott,
S. Serjeant,
M. W. L. Smith,
E. Valiante,
P. Van der Werf
Abstract:
We present a measurement of the angular power spectrum of the cosmic far-infrared background (CFIRB) anisotropies in one of the extragalactic fields of the Herschel Astrophysical Terahertz Large Area Survey (H-ATLAS) at 250, 350 and 500 μm bands. Consistent with recent measurements of the CFIRB power spectrum in Herschel-SPIRE maps, we confirm the existence of a clear one-halo term of galaxy clust…
▽ More
We present a measurement of the angular power spectrum of the cosmic far-infrared background (CFIRB) anisotropies in one of the extragalactic fields of the Herschel Astrophysical Terahertz Large Area Survey (H-ATLAS) at 250, 350 and 500 μm bands. Consistent with recent measurements of the CFIRB power spectrum in Herschel-SPIRE maps, we confirm the existence of a clear one-halo term of galaxy clustering on arcminute angular scales with large-scale two-halo term of clustering at 30 arcminutes to angular scales of a few degrees. The power spectrum at the largest angular scales, especially at 250 μm, is contaminated by the Galactic cirrus. The angular power spectrum is modeled using a conditional luminosity function approach to describe the spatial distribution of unresolved galaxies that make up the bulk of the CFIRB. Integrating over the dusty galaxy population responsible for the background anisotropies, we find that the cosmic abundance of dust, relative to the critical density, to be between Ω_dust=10^{-6} and 8 x 10^{-6} in the redshift range z ~ 0-3. This dust abundance is consistent with estimates of the dust content in the Universe using quasar reddening and magnification measurements in the SDSS.
△ Less
Submitted 10 December, 2012;
originally announced December 2012.
