A5058690922- Radio Astronomy Observations and Technology
- Remote-Sensing Image Classification
- Remote Sensing in Agriculture
- Antenna Design and Optimization
- Soil Moisture and Remote Sensing
- Astrophysics and Cosmic Phenomena
- Land Use and Ecosystem Services
- Galaxies: Formation, Evolution, Phenomena
- Algorithms and Data Compression
- Time Series Analysis and Forecasting
- Remote Sensing and Land Use
- Astronomical Observations and Instrumentation
- Advanced Wireless Communication Techniques
- Radio Wave Propagation Studies
- Error Correcting Code Techniques
- Astronomy and Astrophysical Research
- Image Retrieval and Classification Techniques
- Mathematics and Applications
- semigroups and automata theory
- Structural Health Monitoring Techniques
- Electromagnetic Scattering and Analysis
- Automated Road and Building Extraction
- Maritime Navigation and Safety
- Pulsars and Gravitational Waves Research
- Spectroscopy and Chemometric Analyses
Stellenbosch University
2018-2025
Laboratoire d'Informatique de Paris-Nord
2025
Rhodes University
2014-2020
SKA Telescope, South Africa
2014-2016
University of Pretoria
2011-2013
Council for Scientific and Industrial Research
2011-2012
Tampere University
1995
It has recently been shown that radio interferometric gain calibration can be expressed succinctly in the language of complex optimisation. In addition to providing an elegant framework for further development, it exposes properties problem which exploited accelerate traditional non-linear least squares solvers such as Gauss-Newton and Levenberg-Marquardt. We extend existing derivations chains Jones terms: products several gains model different aberrant effects. doing so, we find useful...
New radio (MeerKAT and Parkes) X-ray (XMM-Newton, Swift, Chandra, NuSTAR) observations of PSR J1622-4950 indicate that the magnetar, in a quiescent state since at least early 2015, reactivated between 2017 March 19 April 5. The flux density, while variable, is approximately 100x larger than during its dormant state. one month after reactivation was 800x quiescence, has been decaying exponentially on 111+/-19 day timescale. This high-flux state, together with radio-derived rotational...
The morphological classification of radio sources is important to gain a full understanding galaxy evolution processes and their relation with local environmental properties. Furthermore, the complex nature problem, its appeal for citizen scientists large data rates generated by existing upcoming telescopes combine make an ideal test case application machine learning techniques. One approach that has shown great promise recently Convolutional Neural Networks (CNNs). Literature, however,...
We present the results of a four-month campaign searching for low-frequency radio transients near North Celestial Pole with Low-Frequency Array (LOFAR), as part Multifrequency Snapshot Sky Survey (MSSS). The data were recorded between 2011 December and 2012 April comprised 2149 11-minute snapshots, each covering 175 deg^2. have found one convincing candidate astrophysical transient, duration few minutes flux density at 60 MHz 15-25 Jy. transient does not repeat has no obvious optical or...
This work investigates a particular class of artefacts, or ghost sources, in radio interferometric images. Earlier observations with (and simulations ) the Westerbork Synthesis Radio Telescope suggested that these were due to calibration incomplete sky models. A theoretical framework is derived validates this suggestion, and provides predictions formation two-source scenario. The are found accurately match result simulations, qualitatively reproduce ghosts previously seen observational data....
Abstract The morphologies of astronomical sources are highly complex, making it essential not only to classify the identified into their predefined categories but also determine that most similar a given query source. Image-based retrieval is essential, as allows an astronomer with source under study ask computer sift through large archived database find ones. This particular interest if does fall “known” category (anomalous). Our work uses trainable COSFIRE (Combination Shifted Filter...
Abstract With the advent of modern radio interferometers, a significant influx in data is expected. This will render manual inspection samples infeasible and thus necessitates development automated approaches to find sources with anomalous morphologies. In this paper, we investigate use autoencoders for source detection, based on assumption that reconstruct anomalies poorly. Specifically, compare an autoencoder architecture from literature two other architectures, as well four conventional...
The key challenge in the observation of redshifted 21-cm signal from cosmic reionization is its separation much brighter foreground emission. Such relies on different spectral properties two components, although, real life, intrinsic spectrum often corrupted by instrumental response, inducing systematic effects that can further jeopardize measurement signal. In this paper, we use Gaussian Process Regression to model both emission and systematics $\sim 2$ hours data Hydrogen Epoch...
ABSTRACT State-of-the-art radio observatories produce large amounts of data which can be used to study the properties galaxies. However, with this rapid increase in volume, it has become unrealistic manually process all incoming data, turn led development automated approaches for processing tasks, such as morphological classification. Deep learning plays a crucial role automation and been shown that convolutional neural networks (CNNs) deliver good performance classification This paper...
A mathematical model is given for describing activity dynamics, learning, and associative memory in the olfactory bulb. Numerical bifurcation analysis calculation of Lyapunov-exponents suggest that chaotic behavior only occurs case strong excitatory coupling mitral layer. Hebbian-type learning rule, supplemented with a nonlinear decay term selective decreasing term, defined analyzed. Slow modifies bulbar dynamics hence it plays crucial role odor information processing. © 1995 John Wiley & Sons, Inc.
ABSTRACT Radio galaxies exhibit a rich diversity of morphological characteristics, which make their classification into distinct types complex challenge. To address this challenge effectively, we introduce an innovative approach for radio galaxy using COSFIRE filters. These filters possess the ability to adapt both shape and orientation prototype patterns within images. The is explainable, learning-free, rotation-tolerant, efficient, does not require large training set. assess efficacy our...
ABSTRACT In the realm of radio astronomy, detection frequency interference (RFI) stands as a pivotal pursuit. This study performs novel comparative analysis deep learning methodologies and introduces transfer method, called fine-tuning. We compare various aspects factors relating to this problem, including supervised fully convolutional network (FCN) architectures used within literature, loss functions, regularization techniques, training methodologies, establish most effective strategies...
In this paper, the internal operations of an Extended Kalman Filter is investigated to observe if information can be derived detect land cover change in a MODerate-resolution Imaging Spectroradiometer (MODIS) time series. The concept based on covariance matrix used by Filter, which adjusts state filter for any changes occurring expands significant reflectance value observed, followed adapting parameters compensate change. analysis shows that detection accuracy above 90% attained when...
Human settlement expansion is one of the most pervasive forms land cover change in Gauteng province South Africa. A method for detecting new developments areas that are typically covered by natural vegetation using 500 m MODIS time-series satellite data proposed. The a per pixel alarm uses temporal autocorrelation to infer index which yields or no-change decision after thresholding. Simulated was generated and used determine threshold during an off-line optimization phase. After evaluated on...
Human settlement expansion is one of the most pervasive forms land-cover change in South Africa. The use Page's cumulative sum (CUSUM) test proposed as a method to detect new developments areas that were previously covered by natural vegetation using 500-m Moderate Resolution Imaging Spectroradiometer time-series satellite data. sequential per-pixel alarm algorithm can take into account positive detection delay, probability detection, and false-alarm construct threshold. Simulated data...
Calibration artefacts, like the self-calibration bias, usually emerge when data are calibrated using an incomplete sky model. In first paper of this series, in which we analysed calibration artefacts from Westerbork Synthesis Radio Telescope, showed that these take form spurious positive and negative sources, refer to as ghosts or ghost sources. We also developed a mathematical framework with could predict pattern east–west interferometer for simple two-source test case. paper, extend our...
Abstract A critical challenge in the observation of redshifted 21 cm line is its separation from bright Galactic and extragalactic foregrounds. In particular, instrumental leakage polarized foregrounds, which undergo significant Faraday rotation as they propagate through interstellar medium, may harmfully contaminate power spectrum. We develop a formalism to describe due widefield effects visibility-based spectra measured with redundant arrays, extending delay-spectrum approach presented...
Abstract Radio interferometric gain calibration can be biased by incomplete sky models and radio frequency interference, resulting in artefacts that restrict the dynamic range of images. It has been suggested algorithms employing heavy-tailed likelihood functions are less susceptible to this due their robustness against outliers data. We present an algorithm based on a Student’s t-distribution which leverages framework complex optimisation Wirtinger calculus for efficient robust calibration....
It is proposed that the time series extracted from moderate resolution imaging spectroradiometer satellite data be modeled as a simple harmonic oscillator with additive colored noise. The noise an Ornstein-Uhlenbeck process. Fourier transform and maximum-likelihood parameter estimation are used to estimate parameters of oscillator. Two case studies in South Africa show reliable class differentiation can obtained between natural vegetation settlement land cover types, when using input...
Observations of the redshifted 21 cm line from epoch reionization have recently motivated construction low-frequency radio arrays with highly redundant configurations. These configurations provide an alternative calibration strategy – 'redundant calibration' and boost sensitivity on specific spatial scales. In this paper, we formulate interferometric as a complex optimization problem. We solve problem via Levenberg–Marquardt algorithm. This approach is more robust to initial conditions than...
Automatic Identification System (AIS) is one of the most prominent systems for monitoring vessel activity. Although significant advances in AIS coverage have been achieved recent times, a will typically still experience gaps reception within its voyage. These can vary from few seconds to multiple hours depending on route. A typical approach when dealing with these vessel's voyage move pointal domain trajectory using prediction algorithm. In this paper, we compare performance two algorithms...
The extraction of information on land cover classes using unsupervised methods has always been relevance to the remote sensing community. In this paper, a novel criterion is proposed, which extracts inherent in an fashion from time series. used fit parametric model series, derive corresponding covariance matrices parameters for model, and estimate additive noise proposed uses both spatial temporal when estimating can be extended incorporate spectral information. algorithm Kalman filter...
There has been recent developments in the use of hyper-temporal satellite time series data for land cover change detection and classification. Recently, an Autocorrelation function (ACF) method was proposed to detect development new human settlements South Africa. In this paper, extension is that produces estimate date addition metric. Preliminary results indicate comparable accuracy achievable relative original formulation, with added advantage providing date.
This is the third installment in a series of papers which we investigate calibration artefacts. Calibration artefacts (also known as ghosts or spurious sources) are created when calibrate with an incomplete model. In first two this series, developed mathematical framework enabled us to study ghosting mechanism itself. An interesting concomitant second paper was that appear symmetrical pairs. could possibly account for symmetrization. Spurious symmetrization refers appearance source (the...