Search

Abstract Deep neural networks (DNNs) have achieved outstanding results in domains such as image processing, computer vision, natural language processing and bioinformatics. In recent years, many methods have been proposed that can provide a visual explanation of decision made by such classifiers. Saliency maps are probably the most popular. However, it is still unclear how to properly interpret saliency maps for a given image and which techniques perform most accurately. This paper presents a methodology to practically evaluate the real effectiveness of saliency map generation methods. We used three state-of-the-art network architectures along with specially prepared benchmark datasets, and we proposed a novel metric to provide a quantitative comparison of the methods. The comparison identified the most reliable techniques and the solutions which usually failed in our tests.

JWST has ushered in an era of unprecedented ability to characterize exoplanetary atmospheres. While there are over 5000 confirmed planets, more than 4000 Transiting Exoplanet Survey Satellite (TESS) planet candidates are still unconfirmed and many of the best planets for atmospheric characterization may remain to be identified. We present a sample of TESS planets and planet candidates that we identify as “best-in-class” for transmission and emission spectroscopy with JWST. These targets are sorted into bins across equilibrium temperature T _eq and planetary radius R _p and are ranked by a transmission and an emission spectroscopy metric (TSM and ESM, respectively) within each bin. We perform cuts for expected signal size and stellar brightness to remove suboptimal targets for JWST. Of the 194 targets in the resulting sample, 103 are unconfirmed TESS planet candidates, also known as TESS Objects of Interest (TOIs). We perform vetting and statistical validation analyses on these 103 targets to determine which are likely planets and which are likely false positives, incorporating ground-based follow-up from the TESS Follow-up Observation Program to aid the vetting and validation process. We statistically validate 18 TOIs, marginally validate 31 TOIs to varying levels of confidence, deem 29 TOIs likely false positives, and leave the dispositions for four TOIs as inconclusive. Twenty-one of the 103 TOIs were confirmed independently over the course of our analysis. We intend for this work to serve as a community resource and motivate formal confirmation and mass measurements of each validated planet. We encourage more detailed analysis of individual targets by the community.

New results on the behavior of the double-mode RR Lyrae V338 Boo are presented. The Transiting Exoplanet Survey Satellite (TESS) observed this star again in 2022, and an observing campaign of the American Association of Variable Star Observers (AAVSO) was completed after the TESS observations as a follow-up. We find that the first overtone pulsation mode in this star completely disappears during the TESS observing window. This mode reappears at the end of the TESS observations, and the AAVSO observing campaign shows that in the months that followed, the first overtone mode was not only present but was the dominant mode of pulsation. This star, and potentially others like it, could hold the key to finally solving the mystery of the Blazhko effect in RR Lyrae.

Abstract There has recently been intensive development of methods for identification and personal verification using the human finger vascular system (FVS). The primary focus of these efforts has been the increasingly sophisticated methods of image processing, and frequently employing machine learning. In this article, we present a new concept of imaging in which the finger vasculature is illuminated using different wavelengths of light, generating multiple FVS images. We hypothesised that the analysis of these image sets, instead of individual images, could increase the effectiveness of identification. Analyses of data from over 100 volunteers, using five different deterministic methods for feature extraction, consistently demonstrated improved identification efficiency with the addition of data obtained from another wavelength. The best results were seen for combinations of diodes between 800 and 900 nm. Finger vascular system observations outside this range were of marginal utility. The knowledge gained from this experiment can be utilised by designers of biometric recognition devices leveraging FVS technology. Our results confirm that developments in this field are not restricted to image processing algorithms, and that hardware innovations remain relevant.

Monochromatic images are used mainly in cases where the intensity of the received signal is examined. The identification of the observed objects as well as the estimation of intensity emitted by them depends largely on the precision of light measurement in image pixels. Unfortunately, this type of imaging is often affected by noise, which significantly degrades the quality of the results. In order to reduce it, numerous deterministic algorithms are used, with Non-Local-Means and Block-Matching-3D being the most widespread and treated as the reference point of the current state-of-the-art. Our article focuses on the utilization of machine learning (ML) for the denoising of monochromatic images in multiple data availability scenarios, including those with no access to noise-free data. For this purpose, a simple autoencoder architecture was chosen and checked for various training approaches on two large and widely used image datasets: MNIST and CIFAR-10. The results show that the method of training as well as architecture and the similarity of images within the image dataset significantly affect the ML-based denoising. However, even without access to any clear data, the performance of such algorithms is frequently well above the current state-of-the-art; therefore, they should be considered for monochromatic image denoising.

Personal identification using analysis of the internal and external characteristics of the human finger is currently an intensively developed topic. The work in this field concerns new methods of feature extraction and image analysis, mainly using modern artificial intelligence algorithms. However, the quality of the data and the way in which it is obtained determines equally the effectiveness of identification. In this article, we present a novel device for extracting vision data from the internal as well as external structures of the human finger. We use spatially selective backlight consisting of NIR diodes of three wavelengths. The fast image acquisition allows for insight into the pulse waveform. Thanks to the external illuminator, images of the skin folds of the finger are acquired as well. This rich collection of images is expected to significantly enhance identification capabilities using existing and future classic and AI-based computer vision techniques. Sample data from our device, before and after data processing, have been shared in a publicly available database.

NASA’s Transiting Exoplanet Survey Satellite (TESS) mission promises to improve our understanding of hot Jupiters by providing an all-sky, magnitude-limited sample of transiting hot Jupiters suitable for population studies. Assembling such a sample requires confirming hundreds of planet candidates with additional follow-up observations. Here we present 20 hot Jupiters that were detected using TESS data and confirmed to be planets through photometric, spectroscopic, and imaging observations coordinated by the TESS Follow-up Observing Program. These 20 planets have orbital periods shorter than 7 days and orbit relatively bright FGK stars (10.9 < G < 13.0). Most of the planets are comparable in mass to Jupiter, although there are four planets with masses less than that of Saturn. TOI-3976b, the longest-period planet in our sample ( P = 6.6 days), may be on a moderately eccentric orbit ( e = 0.18 ± 0.06), while observations of the other targets are consistent with them being on circular orbits. We measured the projected stellar obliquity of TOI-1937A b, a hot Jupiter on a 22.4 hr orbit with the Rossiter–McLaughlin effect, finding the planet’s orbit to be well aligned with the stellar spin axis (∣ λ ∣ = 4.°0 ± 3.°5). We also investigated the possibility that TOI-1937 is a member of the NGC 2516 open cluster but ultimately found the evidence for cluster membership to be ambiguous. These objects are part of a larger effort to build a complete sample of hot Jupiters to be used for future demographic and detailed characterization work.

We present analysis of the RR Lyrae star, LS Her, and confirm the previously reported modulation to its Blazhko cycles. We performed Fourier analysis on two sectors (Sector 24 and 25) of data from the Transiting Exoplanet Survey Satellite (TESS) spanning 53 days. We find LS Her to have a primary pulsation period of 0.2308 day and a Blazhko period of 12.7 days in keeping with previously reported results. We also identified sideband frequencies around the Blazhko multiplets suggesting the Blazhko cycle is modulated on a timescale of 112 days. Analysis of the Blazhko effect using the TESS data clearly shows a changing amplitude and phase throughout the four Blazhko cycles. We compared our modeled results, which were based on our TESS frequency analysis, to TESS data (Sector 51) taken ∼700 days later and found our modulation model was not a good representation of the data. We then coupled our TESS analysis with the modulation frequency results from Wils et al. and found excellent agreement with the Sector 51 data. To further test this result we obtained ground-based, V -magnitude observations of LS Her in the summer of 2022. This data also showed excellent agreement with our coupled modulation model. We have verified that LS Her is a Blazhko star with a modulated Blazhko period of 109 days, stability over the 862 days of observations, and possible stability lasting over 15 yr. We discuss the ramifications of the modulation for other Blazhko stars that show Blazhko effect changes over time.

Imaging through turbulence has been the subject of many research papers in a variety of fields, including defence, astronomy, earth observations, and medicine. The main goal of such research is usually to recover the original, undisturbed image, in which the impact of spatially dependent blurring induced by the phase modulation of the light wavefront is removed. The number of turbulence-disturbed image databases available online is small, and the datasets usually contain repeating types of ground objects (cars, buildings, ships, chessboard patterns). In this article, we present a database of solar images in widely varying turbulence conditions obtained from the SUTO-Solar patrol station recorded over a period of more than a year. The dataset contains image sequences of distinctive yet randomly selected fragments of the solar chromosphere and photosphere. Reference images have been provided with the data using computationally intensive image recovery with the latest multiframe blind deconvolution technique, which is widely accepted in solar imaging. The presented dataset will be extended in the next few years as new image sequences are routinely acquired each sunny day at the SUTO-Solar station.

Deep neural networks (DNNs) have been used successfully for many image classification problems. One of the most important factors that determines the final efficiency of a DNN is the correct construction of the training set. Erroneously labeled training images can degrade the final accuracy and additionally lead to unpredictable model behavior, reducing reliability. In this paper, we propose MultiNET, a novel method for the automatic detection of noisy labels within image datasets. MultiNET is an adaptation of the current state-of-the-art confident learning method. In contrast to the original, our method aggregates the outputs of multiple DNNs and allows for the adjustment of detection sensitivity. We conduct an exhaustive evaluation, incorporating four widely used datasets (CIFAR10, CIFAR100, MNIST, and GTSRB), eight state-of-the-art DNN architectures, and a variety of noise scenarios. Our results demonstrate that MultiNET significantly outperforms the confident learning method.

BRITE-Constellation is devoted to high-precision optical photometric monitoring of bright stars, distributed all over the Milky Way, in red and/or blue passbands. Photometry from space avoids the turbulent and absorbing terrestrial atmosphere and allows for very long and continuous observing runs with high time resolution and thus provides the data necessary for understanding various processes inside stars (e.g., asteroseismology) and in their immediate environment. While the first astronomical observations from space focused on the spectral regions not accessible from the ground it soon became obvious around 1970 that avoiding the turbulent terrestrial atmosphere significantly improved the accuracy of photometry and satellites explicitly dedicated to high-quality photometry were launched. A perfect example is BRITE-Constellation, which is the result of a very successful cooperation between Austria, Canada and Poland. Research highlights for targets distributed nearly over the entire HRD are presented, but focus primarily on massive and hot stars.

Dark current in charge-coupled devices (CCDs) is one of the most important sources of impulsive noise present in scientific images. While the dark current originating in the fabrication defects (mainly impurities) is stable and dependent only on temperature, the one present in the proton-irradiated sensors shows a range of metastable states which makes calibration of images almost impossible. In this paper, we show an extended analysis of such metastabilities present in Kodak KAI 11000M CCD sensors employed in the BRITE (BRIghtest Target Explorer) astrophysical mission over 7 years of in-orbit work. Our collection of dark current characteristics has an unprecedented time span, large temperature range and high number of investigated pixels. A special methodology based on the Gaussian mixture model was proposed for identification and characterization of the metastable states in the dark current. We identified several interesting properties of the metastability and found an experimental rule for the dark current in tristable defects. The results shed a new light on the dark current problems, its modeling and the mitigation in an image sensor working in space.

Most of the current image processing methods used in the near-infrared imaging of finger vascular system concentrate on the extraction of internal structures (veins). In this paper, we propose a novel approach which allows to enhance both internal and external features of a finger. The method is based on the Distance Transformation and allows for selective extraction of physiological structures from an observed finger. We evaluate the impact of its parameters on the effectiveness of the already established processing pipeline used for biometric identification. The new method was compared with five state-of-the-art approaches to features extraction (position-gray-profile-curve—PGPGC, maximum curvature points in image profiles—MC, Niblack image adaptive thresholding—NAT, repeated dark line tracking—RDLT, and wide line detector—WD) on the GustoDB database of images obtained in a wide range of NIR wavelengths (730−950 nm). The results indicate a clear superiority of the proposed approach over the remaining alternatives. The method managed to reach over 90 % identification accuracy for all analyzed datasets.

The BRightest Target Explorer (BRITE) is the pioneering nanosatellite mission dedicated for photometric observations of the brightest stars in the sky. The BRITE charge coupled device (CCD) sensors are poorly shielded against extensive flux of energetic particles which constantly induce defects in the silicon lattice. In this paper we investigate the temporal evolution of the generation of the dark current in the BRITE CCDs over almost four years after launch. Utilizing several steps of image processing and employing normalization of the results, it was possible to obtain useful information about the progress of thermal activity in the sensors. The outcomes show a clear and consistent linear increase of induced damage despite the fact that only about 0.14% of CCD pixels were probed. By performing the analysis of temperature dependencies of the dark current, we identified the observed defects as phosphorus-vacancy (PV) pairs, which are common in proton irradiated CCD matrices. Moreover, the Meyer-Neldel empirical rule was confirmed in our dark current data, yielding E M N = 24.8 meV for proton-induced PV defects.

Optical observations of wide fields of view entail the problem of selecting the best exposure time. As many objects are usually observed simultaneously, the quality of photometry of the brightest ones is always better than that of the dimmer ones, even though all of them are frequently equally interesting for astronomers. Thus, measuring all objects with the highest possible precision is desirable. In this paper, we present a new optimization algorithm, dedicated for the division of exposure time into sub-exposures, which enables photometry with a more balanced noise budget. The proposed technique increases the photometric precision of dimmer objects at the expense of the measurement fidelity of the brightest ones. We have tested the method on real observations using two telescope setups, demonstrating its usefulness and good consistency with theoretical expectations. The main application of our approach is a wide range of sky surveys, including ones performed by space telescopes. The method can be used to plan virtually any photometric observation of objects that show a wide range of magnitudes. [ABSTRACT FROM AUTHOR]