Search

Using high-precision astrometric data from Gaia DR3 and updated systemic velocities from the literature, kinematical properties of cataclysmic variables (CVs) were investigated. By constraining the data according to the total space velocity error and Galactic population class, a reliable sample of data was obtained. Non-magnetic CVs located in the thin disk have been found to have a total space velocity dispersion of $\sigma_{\nu} = 46.33\pm4.23$ km s$^{-1}$, indicating that the thin disk CVs with a mean kinematical age of $\tau = 3.95\pm0.75$ Gyr are much younger than the local thin disk of the Galaxy with $\tau\sim$6-9 Gyr. Total space velocity dispersions of non-magnetic CVs belonging to the thin disk component of the Galaxy were found to be $\sigma_{\nu}=47.67\pm3.94$ and $\sigma_{\nu}=44.43\pm4.33$ km s$^{-1}$ for the systems below and above the orbital period gap, respectively, corresponding to kinematical ages of $\tau=4.19\pm0.71$ and $\tau=3.61\pm0.74$ Gyr. $\gamma$ velocity dispersions of the thin disk CVs below and above the gap were obtained $\sigma_{\gamma} = 27.52\pm2.28$ and $\sigma_{\gamma} = 25.65\pm2.44$ km s$^{-1}$, respectively. This study also shows that the orbital period is decreasing with increasing age, as expected from the standard theory. The age-orbital period relation for non-magnetic thin disk CVs was obtained as $dP/dt=-2.09\pm0.22\times10^{-5}$ sec yr$^{-1}$. However, a significant difference could not be found between the $\gamma$ velocity dispersions of the systems below and above the gap, which were calculated to be $\sigma_{\gamma} = 27.52\pm2.28$ and $\sigma_{\gamma} = 25.65\pm2.44$ km s$^{-1}$, respectively., Comment: 21 pages, including 15 figures and 3 tables, accepted for publication in the Astronomical Journal

Low-light environments pose significant challenges for image enhancement methods. To address these challenges, in this work, we introduce the HUE dataset, a comprehensive collection of high-resolution event and frame sequences captured in diverse and challenging low-light conditions. Our dataset includes 106 sequences, encompassing indoor, cityscape, twilight, night, driving, and controlled scenarios, each carefully recorded to address various illumination levels and dynamic ranges. Utilizing a hybrid RGB and event camera setup. we collect a dataset that combines high-resolution event data with complementary frame data. We employ both qualitative and quantitative evaluations using no-reference metrics to assess state-of-the-art low-light enhancement and event-based image reconstruction methods. Additionally, we evaluate these methods on a downstream object detection task. Our findings reveal that while event-based methods perform well in specific metrics, they may produce false positives in practical applications. This dataset and our comprehensive analysis provide valuable insights for future research in low-light vision and hybrid camera systems., Comment: 18 pages, 4 figures. Has been accepted for publication at the European Conference on Computer Vision Workshops (ECCVW), Milano, 2024. The project page can be found at https://ercanburak.github.io/HUE.html

Collecting and annotating real-world data for the development of object detection models is a time-consuming and expensive process. In the military domain in particular, data collection can also be dangerous or infeasible. Training models on synthetic data may provide a solution for cases where access to real-world training data is restricted. However, bridging the reality gap between synthetic and real data remains a challenge. Existing methods usually build on top of baseline Convolutional Neural Network (CNN) models that have been shown to perform well when trained on real data, but have limited ability to perform well when trained on synthetic data. For example, some architectures allow for fine-tuning with the expectation of large quantities of training data and are prone to overfitting on synthetic data. Related work usually ignores various best practices from object detection on real data, e.g. by training on synthetic data from a single environment with relatively little variation. In this paper we propose a methodology for improving the performance of a pre-trained object detector when training on synthetic data. Our approach focuses on extracting the salient information from synthetic data without forgetting useful features learned from pre-training on real images. Based on the state of the art, we incorporate data augmentation methods and a Transformer backbone. Besides reaching relatively strong performance without any specialized synthetic data transfer methods, we show that our methods improve the state of the art on synthetic data trained object detection for the RarePlanes and DGTA-VisDrone datasets, and reach near-perfect performance on an in-house vehicle detection dataset., Comment: Submitted to and presented at SPIE Defense + Commercial Sensing 2024, 13 pages, 4 figures, 3 tables

Developments on various relations among stellar variables such as the main sequence empirical mass-luminosity (MLR), mass-radius (MRR) and mass-effective temperature (MTR) relations were reviewed. Conceptual changes in their understanding and usages were discussed. After its discovery, MLR was treated as one of the fundamental secrets of the cosmos. Differences between fundamental laws and statistical relations were used to understand long-term developments of MLR, MRR and MTR. Developments show a break point, initiated by Andersen et al. (1991), in the line of progress. Before the break when reliable data were limited, MLR and MRR were calibrated using M, L, and R of binary components of all kinds visual, spectroscopic, and eclipsing for two purposes: i) obtaining mean mass, mean luminosity, and mean radius, ii) to estimate M and R of single stars. By the time of the break, the number of solutions from detached double-lined eclipsing binaries (DDEB) giving accurate M and R within a few percent levels are increased. Parameters from very close, semi-detached, and contact binaries were excluded for refinement, however, MLR and MRR diagrams were found insufficient to derive MLR and MRR functions because the dispersions are not only due to random observational errors but also due to chemical composition and age differences. Then, a new trend was adopted by replacing classical MLR and MRR with empirical M and R predicting relations. Thus, the purpose one was suppressed also because the new trend found a fruitful application in determining M and R of exoplanet hosting single stars. Corrections on misnames and devising new classical MLR, MRR, and MTR, giving mean values are encouraged since they are still useful and needed by astrophysical models requiring such mean values, not only beneficial to astrophysics, but also beneficial to Galactic, extragalactic search, even cosmological models., Comment: 26 pages, 7 figures and 2 tables, accepted for publication in Physics and Astronomy Reports

Rewriting logic is a natural and expressive framework for the specification of concurrent systems and logics. The Maude specification language provides an implementation of this formalism that allows executing, verifying, and analyzing the represented systems. These specifications declare their objects by means of terms and equations, and provide rewriting rules to represent potentially non-deterministic local transformations on the state. Sometimes a controlled application of these rules is required to reduce non-determinism, to capture global, goal-oriented or efficiency concerns, or to select specific executions for their analysis. That is what we call a strategy. In order to express them, respecting the separation of concerns principle, a Maude strategy language was proposed and developed. The first implementation of the strategy language was done in Maude itself using its reflective features. After ample experimentation, some more features have been added and, for greater efficiency, the strategy language has been implemented in C++ as an integral part of the Maude system. This paper describes the Maude strategy language along with its semantics, its implementation decisions, and several application examples from various fields.

A detailed analysis of ground-based CCD UBV photometry and space-based Gaia Data Release 3 (DR3) data for the open clusters King 6 and NGC 1605 was performed. Using the pyUPMASK algorithm on Gaia astrometric data to estimate cluster membership probabilities, we have identified 112 stars in King 6 and 160 stars in NGC 1605 as the statistically most likely members of each cluster. We calculated reddening and metallicity separately using UBV two-color diagrams to estimate parameter values via independent methods. The color excess $E(B-V)$ and photometric metallicity [Fe/H] for King 6 are $0.515 \pm 0.030$ mag and $0.02 \pm 0.20$ dex, respectively. For NGC 1605, they are $0.840 \pm 0.054$ mag and $0.01 \pm 0.20$ dex. With reddening and metallicity kept constant, we have estimated the distances and cluster ages by fitting PARSEC isochrones to color-magnitude diagrams based on the Gaia and UBV data. Photometric distances are 723 $\pm$ 34 pc for King 6 and 3054 $\pm$ 243 pc for NGC 1605. The cluster ages are $200 \pm 20$ Myr and $400 \pm 50$ Myr for King 6 and NGC 1605, respectively. Mass function slopes were found to be 1.29 $\pm$ 0.18 and 1.63 $\pm$ 0.36 for King 6 and NGC 1605, respectively. These values are in good agreement with the value of Salpeter (1955). The relaxation times were estimated as 5.8 Myr for King 6 and 60 Myr for NGC 1605. This indicates that both clusters are dynamically relaxed since these times are less than the estimated cluster ages. Galactic orbit analysis shows that both clusters formed outside the solar circle and are members of the young thin-disc population., Comment: 25 pages, including 14 figures and 7 tables, accepted for publication in the Astronomical Journal

Social tipping points are promising levers to achieve net-zero greenhouse gas emission targets. They describe how social, political, economic or technological systems can move rapidly into a new state if cascading positive feedback mechanisms are triggered. Analysing the potential of social tipping for rapid decarbonization requires considering the inherent complexity of social systems. Here, we identify that existing scientific literature is inclined to a narrative-based account of social tipping, lacks a broad empirical framework and a multi-systems view. We subsequently outline a dynamic systems approach that entails (i) a systems outlook involving interconnected feedback mechanisms alongside cross-system and cross-scale interactions, and including a socioeconomic and environmental injustice perspective (ii) directed data collection efforts to provide empirical evidence for and monitor social tipping dynamics, (iii) global, integrated, descriptive modelling to project future dynamics and provide ex-ante evidence for interventions. Research on social tipping must be accordingly solidified for climate policy relevance.

Main-sequence bolometric corrections (BC) and a standard BC-Teff relation are produced for TESS wavelengths using published physical parameters and light ratios from SED models of 209 detached double-lined eclipsing binaries. This and previous five-band (Johnson B, V, Gaia G, GBP, GRP) standard BC-Teff relations are tested by recovering luminosity (L) of the most accurate 341 single host stars (281 MS, 40 subgiants, 19 giants and one PMS). The recovered L of photometry is compared to L from published R and Teff. A very high correlation ($R^2$ = 0.9983) is achieved for this mixed sample. Error histograms of recovered and calculated L show peaks at 2 and 4 per cent respectively. The recovered L and the published Teff} were then used in $L = 4 \pi R^2 \sigma Teff^4$ to predict the standard R of the host stars. Comparison between the predicted and published R of all luminosity classes are found successful with a negligible offset associated with the giants and subgiants. The peak of the predicted R errors is found at 2 per cent, which is equivalent to the peak of the published R errors. Thus, a main-sequence BC-Teff relation could be used in predicting both L and R of a single star at any luminosity class, but this does not mean BC-Teff relations of all luminosity classes are the same because luminosity information could be more constrained by star's apparent magnitude $\xi$ than its BC since $m_{Bol} = \xi + BC_\xi$., Comment: 11 pages, 9 figures, a total of 6 tables, 5 online only tables

Event-based cameras are becoming increasingly popular for their ability to capture high-speed motion with low latency and high dynamic range. However, generating videos from events remains challenging due to the highly sparse and varying nature of event data. To address this, in this study, we propose HyperE2VID, a dynamic neural network architecture for event-based video reconstruction. Our approach uses hypernetworks to generate per-pixel adaptive filters guided by a context fusion module that combines information from event voxel grids and previously reconstructed intensity images. We also employ a curriculum learning strategy to train the network more robustly. Our comprehensive experimental evaluations across various benchmark datasets reveal that HyperE2VID not only surpasses current state-of-the-art methods in terms of reconstruction quality but also achieves this with fewer parameters, reduced computational requirements, and accelerated inference times., Comment: 20 pages, 11 figures. Accepted by IEEE Transactions on Image Processing. The project page can be found at https://ercanburak.github.io/HyperE2VID.html

Event cameras are a new type of vision sensor that incorporates asynchronous and independent pixels, offering advantages over traditional frame-based cameras such as high dynamic range and minimal motion blur. However, their output is not easily understandable by humans, making the reconstruction of intensity images from event streams a fundamental task in event-based vision. While recent deep learning-based methods have shown promise in video reconstruction from events, this problem is not completely solved yet. To facilitate comparison between different approaches, standardized evaluation protocols and diverse test datasets are essential. This paper proposes a unified evaluation methodology and introduces an open-source framework called EVREAL to comprehensively benchmark and analyze various event-based video reconstruction methods from the literature. Using EVREAL, we give a detailed analysis of the state-of-the-art methods for event-based video reconstruction, and provide valuable insights into the performance of these methods under varying settings, challenging scenarios, and downstream tasks., Comment: 19 pages, 9 figures. Has been accepted for publication at the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Vancouver, 2023. The project page can be found at https://ercanburak.github.io/evreal.html

The Ruscheweyh derivative operator is used in this paper to introduce and investigate interesting general subclasses of the function class $\Sigma_{\mathrm{m}}$ of $m$-fold symmetric bi-univalent analytic functions. Estimates of the initial Taylor-Maclaurin coefficients $\left|a_{m+1}\right|$ and $\left|a_{2 m+1}\right|$ are obtained for functions of the subclasses introduced in this study, and the consequences of the results are discussed. The results presented would generalize and improve on some recent works by many earlier authors. In some cases, our estimates are better than the existing coefficient bounds. Furthermore, within the engineering domain, this paper delves into a series of complex issues related to analytic functions, $m$-fold symmetric univalent functions, and the utilization of the Ruscheweyh derivative operator. These problems encompass a broad spectrum of engineering applications, including the optimization of optical system designs, signal processing for antenna arrays, image compression techniques, and filter design for control systems. The paper underscores the crucial role of these mathematical concepts in addressing practical engineering dilemmas and fine-tuning the performance of various engineering systems. It emphasizes the potential for innovative solutions that can significantly enhance the reliability and effectiveness of engineering applications., Comment: 15 pages

Patient outcome prediction is critical in management of ischemic stroke. In this paper, a novel machine learning model is proposed for stroke outcome prediction using multimodal Magnetic Resonance Imaging (MRI). The proposed model consists of two serial levels of Autoencoders (AEs), where different AEs at level 1 are used for learning unimodal features from different MRI modalities and a AE at level 2 is used to combine the unimodal features into compressed multimodal features. The sequences of multimodal features of a given patient are then used by an LSTM network for predicting outcome score. The proposed AE2-LSTM model is proved to be an effective approach for better addressing the multimodality and volumetric nature of MRI data. Experimental results show that the proposed AE2-LSTM outperforms the existing state-of-the art models by achieving highest AUC=0.71 and lowest MAE=0.34., Comment: The IEEE International Symposium on Biomedical Imaging (ISBI). arXiv admin note: text overlap with arXiv:2205.05545

Facial recognition is fundamental for a wide variety of security systems operating in real-time applications. In video surveillance based face recognition, face images are typically captured over multiple frames in uncontrolled conditions; where head pose, illumination, shadowing, motion blur and focus change over the sequence. We can generalize that the three fundamental operations involved in the facial recognition tasks: face detection, face alignment and face recognition. This study presents comparative benchmark tables for the state-of-art face recognition methods by testing them with same backbone architecture in order to focus only on the face recognition solution instead of network architecture. For this purpose, we constructed a video surveillance dataset of face IDs that has high age variance, intra-class variance (face make-up, beard, etc.) with native surveillance facial imagery data for evaluation. On the other hand, this work discovers the best recognition methods for different conditions like non-masked faces, masked faces, and faces with glasses.

Arbitrariness in the zero point of bolometric corrections is a nearly century-old paradigm leading to two more paradigms. "Bolometric corrections must always be negative," and "bolometric magnitude of a star ought to be brighter than its $V$ magnitude". Both were considered valid before IAU 2015 General Assembly Resolution B2, a revolutionary document that supersedes all three aforementioned paradigms. The purpose of this article is to initiate a new insight and a new understanding of the fundamental astrophysics and present new capabilities to obtain standard and more accurate stellar luminosities and gain more from accurate observations in the era after Gaia. The accuracy gained will aid in advancing stellar structure and evolution theories, and also Galactic and extragalactic research, observational cosmology and dark matter and dark energy searches., Comment: 8 pages, including 1 figure, accepted for publication in the Astronomical Journal

Standard luminosity ($L$) of 406 main-sequence stars with the most accurate astrophysical parameters are predicted from their absolute magnitudes and bolometric corrections at Johnson $B,V$, and Gaia EDR3 $G$, $G_{BP}$, $G_{RP}$ filters. Required multiband $BC$ and $BC-T_{eff}$ relations are obtained first from the parameters of 209 DDEB (Double-lined Detached Eclipsing Binaries) with main-sequence components and Gaia EDR3 parallaxes. A simplified SED is formulated to give filter dependent component light contributions and interstellar dimming, which are essential in computing $BC$ of a component virtually at any filter. The mean standard $L$ of a star is calculated from the mean $M_{Bol}$ which is a mathematical average of independent $M_{Bol}$ values predicted at different filters, while the uncertainty of $L$ is the uncertainty propagated from the uncertainty of the mean $M_{Bol}$. The mean standard $L$ of the sample stars are compared to the corresponding $L$ values according to the Stefan-Boltzmann law. A very high correlation ($R^2>0.999$) is found. Comparing histogram distributions of errors shows that uncertainties associated with the mean standard $L$ (peak at $\sim2.5$ per cent) are much smaller than the uncertainties of $L$ (peak at $\sim8$ per cent) by the Stefan-Boltzmann law. Increasing the number of filters used in predicting the mean $M_{Bol}$ increases the accuracy of the standard stellar luminosity. Extinction law, color-color relations and color excess - color excess relations for Gaia passbands are demonstrated for main-sequence stars for the first time., Comment: After submitting minor revision in Acta Astronomica on August 8th, 2022. Tables will be available as online supplementary material