12 research outputs found
Feature guided training and rotational standardization for the morphological classification of radio galaxies
No full textState-of-the-art radio observatories produce large amounts of data which can be used to study the properties of radio galaxies.
However, with this rapid increase in data volume, it has become unrealistic to manually process all of the incoming data, which
in turn led to the development of automated approaches for data processing tasks, such as morphological classification. Deep
learning plays a crucial role in this automation process and it has been shown that convolutional neural networks (CNNs) can
deliver good performance in the morphological classification of radio galaxies. This paper investigates two adaptations to the
application of these CNNs for radio galaxy classification. The first adaptation consists of using principal component analysis
(PCA) during pre-processing to align the galaxies’ principal components with the axes of the coordinate system, which will
normalize the orientation of the galaxies. This adaptation led to a significant improvement in the classification accuracy of the
CNNs and decreased the average time required to train the models. The second adaptation consists of guiding the CNN to look
for specific features within the samples in an attempt to utilize domain knowledge to improve the training process. It was found
that this adaptation generally leads to a stabler training process and in certain instances reduced overfitting within the network,
as well as the number of epochs required for training
Feature Guided Training and Rotational Standardisation for the Morphological Classification of Radio Galaxies
No full textState-of-the-art radio observatories produce large amounts of data which can
be used to study the properties of radio galaxies. However, with this rapid
increase in data volume, it has become unrealistic to manually process all of
the incoming data, which in turn led to the development of automated approaches
for data processing tasks, such as morphological classification. Deep learning
plays a crucial role in this automation process and it has been shown that
convolutional neural networks (CNNs) can deliver good performance in the
morphological classification of radio galaxies. This paper investigates two
adaptations to the application of these CNNs for radio galaxy classification.
The first adaptation consists of using principal component analysis (PCA)
during preprocessing to align the galaxies' principal components with the axes
of the coordinate system, which will normalize the orientation of the galaxies.
This adaptation led to a significant improvement in the classification accuracy
of the CNNs and decreased the average time required to train the models. The
second adaptation consists of guiding the CNN to look for specific features
within the samples in an attempt to utilize domain knowledge to improve the
training process. It was found that this adaptation generally leads to a
stabler training process and in certain instances reduced overfitting within
the network, as well as the number of epochs required for training.Comment: 20 pages, 17 figures, this is a pre-copyedited, author-produced PDF
of an article accepted for publication in the Monthly Notices of the Royal
Astronomical Societ
FGT-RS-CNN: Version 1 Release
No full text<p>This is the code used for the experiments associated with the paper "Feature Guided Training and Rotational Standardisation for Morphological Classification of Radio Galaxies" by Brand et al. accepted for publication by Monthly Notices of the Royal Astronomical Society (MNRAS) on 29 March 2023.</p>
FRGMRC - FIRST Radio Galaxy Morphology Reference Catalogue
No full textFRGMRC - FIRST Radio Galaxy Morphology Reference Catalogue - Version 1.0
Brand et al. 2023 - https://ui.adsabs.harvard.edu/abs/2023MNRAS.522..292B/abstrac
Deep supervised hashing for fast retrieval of radio image cubes
No full textThe shear number of sources that will be detected by next-generation radio surveys will be astronomical, which will result in serendipitous discoveries. Data-dependent deep hashing algorithms have been shown to be efficient at image retrieval tasks in the fields of computer vision and multimedia. However, there are limited applications of these methodologies in the field of astronomy. In this work, we utilize deep hashing to rapidly search for similar images in a large database. The experiment uses a balanced dataset of 2708 samples consisting of four classes: Compact, FRI, FRII, and Bent. The performance of the method was evaluated using the mean average precision (mAP) metric where a precision of 88.5% was achieved. The experimental results demonstrate the capability to search and retrieve similar radio images efficiently and at scale. The retrieval is based on the Hamming distance between the binary hash of the query image and those of the reference images in the database.</p
Advances on the morphological classification of radio galaxies: A review
No full textModern radio telescopes will generate, on a daily basis, data sets on the scale of exabytes for systems like the Square Kilometre Array (SKA). Massive data sets are a source of unknown and rare astrophysical phenomena that lead to discoveries. Nonetheless, this is only plausible with the exploitation of machine learning to complement human-aided and traditional statistical techniques. Recently, there has been a surge in scientific publications focusing on the use of machine/deep learning in radio astronomy, addressing challenges such as source extraction, morphological classification, and anomaly detection. This study provides a comprehensive and concise overview of the use of machine learning techniques for the morphological classification of radio galaxies. It summarizes the recent literature on this topic, highlighting the main challenges, achievements, state-of-the-art methods, and the future research directions in the field. The application of machine learning in radio astronomy has led to a new paradigm shift and a revolution in the automation of complex data processes. However, the optimal exploitation of machine/deep learning in radio astronomy, calls for continued collaborative efforts in the creation of high-resolution annotated data sets. This is especially true in the case of modern telescopes like MeerKAT and the LOw-Frequency ARray (LOFAR). Additionally, it is important to consider the potential benefits of utilizing multi-channel data cubes and algorithms that can leverage massive datasets without relying solely on annotated datasets for radio galaxy classification.<br/
Classification of radio galaxies with trainable COSFIRE filters
No full textRadio galaxies exhibit a rich diversity of morphological characteristics, which make their classification into distinct types a complex challenge. To address this challenge effectively, we introduce an innovative approach for radio galaxy classification using COSFIRE filters. These filters possess the ability to adapt to both the shape and orientation of prototype patterns within images. The COSFIRE approach is explainable, learning-free, rotation-tolerant, efficient, and does not require a large training set. To assess the efficacy of our method, we conducted experiments on a benchmark radio galaxy data set comprising of 1180 training samples and 404 test samples. Notably, our approach achieved an average accuracy rate of 93.36 %. This achievement outperforms contemporary deep learning models, and it is the best result ever achieved on this data set. Additionally, COSFIRE filters offer better computational performance, ∼20 × fewer operations than the DenseNet-based competing method (when comparing at the same accuracy). Our findings underscore the effectiveness of the COSFIRE filter-based approach in addressing the complexities associated with radio galaxy classification. This research contributes to advancing the field by offering a robust solution that transcends the orientation challenges intrinsic to radio galaxy observations. Our method is versatile in that it is applicable to various image classification approaches
Calibration artefacts in radio interferometry - I. Ghost sources in Westerbork Synthesis Radio Telescope data
No full textThis work investigates a particular class of artefacts, or ghost sources, in radio in-
terferometric images. Earlier observations with (and simulations of) the Westerbork
Synthesis Radio Telescope (WSRT) suggested that these were due to calibration with
incomplete sky models. A theoretical framework is derived that validates this sug-
gestion, and provides predictions of ghost formation in a two-source scenario. The
predictions are found to accurately match the result of simulations, and qualitatively
reproduce the ghosts previously seen in observational data. The theory also provides
explanations for many previously puzzling features of these artefacts (regular geom-
etry, PSF-like sidelobes, seeming independence on model
ux), and shows that the
observed phenomenon of
ux suppression a ecting unmodelled sources is due to the
same mechanism. We demonstrate that this ghost formation mechanism is a funda-
mental feature of calibration, and exhibits a particularly strong and localized signature
due to array redundancy. To some extent this mechanism will a ect all observations
(including those with non-redundant arrays), though in most cases the ghosts remain
hidden below the noise or masked by other instrumental artefacts. The implications
of such errors on future deep observations are discussed.South African Research Chairs Initiative of the Department of Science and Technology and National Research Foundationhttp://mnras.oxfordjournals.orghb201
