scholarly journals The FIRST Classifier: compact and extended radio galaxy classification using deep Convolutional Neural Networks

2018 ◽  
Author(s):  
Wathela Alhassan

Upcoming surveys with new radio observatories such as the Square Kilometer Array will generate a wealth of imaging data containing large numbers of radio galaxies. Different classes of radio galaxies can be used as tracers of the cosmic environment, including the dark matter density field, to address key cosmological questions. Classifying these galaxies based on morphology is thus an important step toward achieving the science goals of next generation radio surveys. Radio galaxies have been traditionally classified as Fanaroff-Riley (FR) I and II, although some exhibit more complex 'bent' morphologies arising from environmental factors or intrinsic properties. In this work we present the FIRST Classifier, an on-line system for automated classification of Compact and Extended radio sources. We developed the FIRST Classifier based on a trained Deep Convolutional Neural Network Model to automate the morphological classification of com- pact and extended radio sources observed in the FIRST radio survey. Our model achieved an overall accuracy of 97% and a recall of 98%, 100%, 98% and 93% for Compact, BENT, FRI and FRII galaxies respectively. The current version of the FIRST classifier is able to identify the morphological class for a single source or for a list of sources as Compact or Extended (FRI, FRII and BENT).

Author(s):  
Micah Bowles ◽  
Anna M M Scaife ◽  
Fiona Porter ◽  
Hongming Tang ◽  
David J Bastien

Abstract In this work we introduce attention as a state of the art mechanism for classification of radio galaxies using convolutional neural networks. We present an attention-based model that performs on par with previous classifiers while using more than 50% fewer parameters than the next smallest classic CNN application in this field. We demonstrate quantitatively how the selection of normalisation and aggregation methods used in attention-gating can affect the output of individual models, and show that the resulting attention maps can be used to interpret the classification choices made by the model. We observe that the salient regions identified by the our model align well with the regions an expert human classifier would attend to make equivalent classifications. We show that while the selection of normalisation and aggregation may only minimally affect the performance of individual models, it can significantly affect the interpretability of the respective attention maps and by selecting a model which aligns well with how astronomers classify radio sources by eye, a user can employ the model in a more effective manner.


2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Song-Quan Ong ◽  
Hamdan Ahmad ◽  
Gomesh Nair ◽  
Pradeep Isawasan ◽  
Abdul Hafiz Ab Majid

AbstractClassification of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse) by humans remains challenging. We proposed a highly accessible method to develop a deep learning (DL) model and implement the model for mosquito image classification by using hardware that could regulate the development process. In particular, we constructed a dataset with 4120 images of Aedes mosquitoes that were older than 12 days old and had common morphological features that disappeared, and we illustrated how to set up supervised deep convolutional neural networks (DCNNs) with hyperparameter adjustment. The model application was first conducted by deploying the model externally in real time on three different generations of mosquitoes, and the accuracy was compared with human expert performance. Our results showed that both the learning rate and epochs significantly affected the accuracy, and the best-performing hyperparameters achieved an accuracy of more than 98% at classifying mosquitoes, which showed no significant difference from human-level performance. We demonstrated the feasibility of the method to construct a model with the DCNN when deployed externally on mosquitoes in real time.


2017 ◽  
Vol 2017 ◽  
pp. 1-6 ◽  
Author(s):  
Atsushi Teramoto ◽  
Tetsuya Tsukamoto ◽  
Yuka Kiriyama ◽  
Hiroshi Fujita

Lung cancer is a leading cause of death worldwide. Currently, in differential diagnosis of lung cancer, accurate classification of cancer types (adenocarcinoma, squamous cell carcinoma, and small cell carcinoma) is required. However, improving the accuracy and stability of diagnosis is challenging. In this study, we developed an automated classification scheme for lung cancers presented in microscopic images using a deep convolutional neural network (DCNN), which is a major deep learning technique. The DCNN used for classification consists of three convolutional layers, three pooling layers, and two fully connected layers. In evaluation experiments conducted, the DCNN was trained using our original database with a graphics processing unit. Microscopic images were first cropped and resampled to obtain images with resolution of 256 × 256 pixels and, to prevent overfitting, collected images were augmented via rotation, flipping, and filtering. The probabilities of three types of cancers were estimated using the developed scheme and its classification accuracy was evaluated using threefold cross validation. In the results obtained, approximately 71% of the images were classified correctly, which is on par with the accuracy of cytotechnologists and pathologists. Thus, the developed scheme is useful for classification of lung cancers from microscopic images.


Author(s):  
R. R. Andreasyan ◽  
H. V. Abrahamyan

It is brought the physical and morphological data of 267 nearby radio galaxies identified with elliptical galaxies brighter than 18th magnitude (sample 1) and for 280 extragalactic radio sources with known position angles between the integrated intrinsic radio polarization and radio axes (sample 2).


2020 ◽  
Vol 493 (3) ◽  
pp. 4209-4228 ◽  
Author(s):  
Ting-Yun Cheng ◽  
Christopher J Conselice ◽  
Alfonso Aragón-Salamanca ◽  
Nan Li ◽  
Asa F L Bluck ◽  
...  

ABSTRACT There are several supervised machine learning methods used for the application of automated morphological classification of galaxies; however, there has not yet been a clear comparison of these different methods using imaging data, or an investigation for maximizing their effectiveness. We carry out a comparison between several common machine learning methods for galaxy classification [Convolutional Neural Network (CNN), K-nearest neighbour, logistic regression, Support Vector Machine, Random Forest, and Neural Networks] by using Dark Energy Survey (DES) data combined with visual classifications from the Galaxy Zoo 1 project (GZ1). Our goal is to determine the optimal machine learning methods when using imaging data for galaxy classification. We show that CNN is the most successful method of these ten methods in our study. Using a sample of ∼2800 galaxies with visual classification from GZ1, we reach an accuracy of ∼0.99 for the morphological classification of ellipticals and spirals. The further investigation of the galaxies that have a different ML and visual classification but with high predicted probabilities in our CNN usually reveals the incorrect classification provided by GZ1. We further find the galaxies having a low probability of being either spirals or ellipticals are visually lenticulars (S0), demonstrating that supervised learning is able to rediscover that this class of galaxy is distinct from both ellipticals and spirals. We confirm that ∼2.5 per cent galaxies are misclassified by GZ1 in our study. After correcting these galaxies’ labels, we improve our CNN performance to an average accuracy of over 0.99 (accuracy of 0.994 is our best result).


1984 ◽  
Vol 110 ◽  
pp. 39-40
Author(s):  
P. D. Barthel ◽  
G. K. Miley ◽  
R. T. Schilizzi ◽  
E. Preuss ◽  
T. J. Cornwell

The Nuclear Radio cores of several nearby extended radio galaxies (e.g. M87, 3C236) consist not only of optically thick (< 1 pc) components, but also of emission on somewhat larger scale. As extended radio sources associated with quasars have on average stronger and more luminous radio cores (see e.g. Miley, 1980, Ann. Rev. Astron. Astrophys. 18, 165), we have started a project to study the properties of these quasar cores.


2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Tuan D. Pham

Abstract The use of imaging data has been reported to be useful for rapid diagnosis of COVID-19. Although computed tomography (CT) scans show a variety of signs caused by the viral infection, given a large amount of images, these visual features are difficult and can take a long time to be recognized by radiologists. Artificial intelligence methods for automated classification of COVID-19 on CT scans have been found to be very promising. However, current investigation of pretrained convolutional neural networks (CNNs) for COVID-19 diagnosis using CT data is limited. This study presents an investigation on 16 pretrained CNNs for classification of COVID-19 using a large public database of CT scans collected from COVID-19 patients and non-COVID-19 subjects. The results show that, using only 6 epochs for training, the CNNs achieved very high performance on the classification task. Among the 16 CNNs, DenseNet-201, which is the deepest net, is the best in terms of accuracy, balance between sensitivity and specificity, $$F_1$$ F 1 score, and area under curve. Furthermore, the implementation of transfer learning with the direct input of whole image slices and without the use of data augmentation provided better classification rates than the use of data augmentation. Such a finding alleviates the task of data augmentation and manual extraction of regions of interest on CT images, which are adopted by current implementation of deep-learning models for COVID-19 classification.


2021 ◽  
Author(s):  
◽  
Siamak Dehghan

<p>This thesis presents an investigation of the habitat of extended radio sources, and the way in which the generation and properties of these radio sources are affected by environmental factors. We begin with a detailed structure analysis of the 0.3 deg² area of the MUSYC-ACES field, generated by applying a density-based clustering method, known as DBSCAN, to our spectroscopic and photometric samples of the field. As a result, we identify 62 over-dense regions across the field. Based on the properties of the detected structures, we classify 13 as clusters, of which 90% are associated with diffuse soft-band X-ray emission. This provides a strong and independent confirmation that both the clustering and classification methodologies are reliable for use in investigation of the environment of the radio sources in the Chandra Deep Field South (CDFS).  Using an interpolation-based method followed by a new calibration technique of using clusters of similar mass as standard candles, we are able to estimate the local environmental richness for a desired region. This methodology is applied to a sample of AGNs and star forming galaxies in the CDFS to probe whether or not the radio luminosity of the different radio sources is correlated to their environments. As a result, we do not find a significant correlation between the radio luminosity and the environment of star-forming galaxies and radio-quiet AGNs, however, a weak positive dependency is spotted for radio-loud AGNs. This may indicate that over-populated environments trigger or enhance the radio activity processes in the AGNs. We find that star-forming galaxies, unlike radio-loud AGNs, tend to avoid overpopulated environments especially at low redshifts. However, radio-loud AGN are found in both poor and rich environments. As a result, we find neither of these radio sources suitable for tracing the over-dense regions of the Universe, unlike tailed radio galaxies.  It is believed that tailed radio galaxies reside in the dense environments of clusters and groups, and therefore, may be the signatures of overdensities in large-scale structure. To evaluate the idea of using tailed radio galaxies as tracers of dense environments, a systematic study of these sources as a function of density is required. For this reason and by using the 1.4 GHz Australia Telescope Large Area Survey (ATLAS) data, we examined over four deg² area of the ATLAS-CDFS field, which includes the entire CDFS. We present a catalogue of 56 non-linear, extended, and low surface brightness sources including 45 tailed radio galaxies, two relic candidates, and a possible radio halo. We report the detection of the most distant tailed radio galaxy to date, at a redshift of 2.1688. In addition, despite the lack of deep spectroscopic data in the ATLAS field, we find two of the detected tailed radio galaxies are associated with clusters. We find three Head-Tail galaxy candidates in the CDFS field, all of which are located at high redshifts, where the magnitude constraint of our redshift sample prevents any structure detection.  One of the primary objectives of this research is to investigate the association between the morphology of tailed radio galaxies and the physical characteristics of the surrounding environment. In order to understand the role of the variety of factors that influence the radio morphology, we constructed a simple model that generates the overall radio structure of the sources in different habitats. We report the results of the simulation of the wide-angle tail radio galaxy PKS J0334-3900, which shows that both the gravitation interactions and a cluster wind are required to generate the observed radio tails. As a result, we find the morphology of the tailed radio galaxies as an invaluable tool to probe environmental characteristics.  In a supplementary study, we investigate the role of cluster dynamics on generation and alternation of extended radio sources. We present a comprehensive structure and sub-structure analysis of the Abell 3266 galaxy cluster. Based on the results of the sub-structure test, position and orientation of a radio relic candidate, and morphology of a prominent tailed radio galaxy in the cluster, we propose an ongoing merger scenario for this chaotic cluster environment. Furthermore, we verify our theory by an N-body simulation of a pre-merger cluster and an in-falling group. The results of the simulation supports our merger scenario by explaining both the orientation of the radio relic and the observed morphology of the tailed radio galaxy.  While there is a weak correlation between the luminosity of radio-loud AGNs and environmental density, tailed radio galaxies make superior probes of over-dense regions. Thus, overall we find tailed radio galaxies can be used to trace overdensities out to z ~ 2 and probe the details of the environments in which they are found.</p>


2021 ◽  
Vol 11 (22) ◽  
pp. 10528
Author(s):  
Khin Yadanar Win ◽  
Noppadol Maneerat ◽  
Syna Sreng ◽  
Kazuhiko Hamamoto

The ongoing COVID-19 pandemic has caused devastating effects on humanity worldwide. With practical advantages and wide accessibility, chest X-rays (CXRs) play vital roles in the diagnosis of COVID-19 and the evaluation of the extent of lung damages incurred by the virus. This study aimed to leverage deep-learning-based methods toward the automated classification of COVID-19 from normal and viral pneumonia on CXRs, and the identification of indicative regions of COVID-19 biomarkers. Initially, we preprocessed and segmented the lung regions usingDeepLabV3+ method, and subsequently cropped the lung regions. The cropped lung regions were used as inputs to several deep convolutional neural networks (CNNs) for the prediction of COVID-19. The dataset was highly unbalanced; the vast majority were normal images, with a small number of COVID-19 and pneumonia images. To remedy the unbalanced distribution and to avoid biased classification results, we applied five different approaches: (i) balancing the class using weighted loss; (ii) image augmentation to add more images to minority cases; (iii) the undersampling of majority classes; (iv) the oversampling of minority classes; and (v) a hybrid resampling approach of oversampling and undersampling. The best-performing methods from each approach were combined as the ensemble classifier using two voting strategies. Finally, we used the saliency map of CNNs to identify the indicative regions of COVID-19 biomarkers which are deemed useful for interpretability. The algorithms were evaluated using the largest publicly available COVID-19 dataset. An ensemble of the top five CNNs with image augmentation achieved the highest accuracy of 99.23% and area under curve (AUC) of 99.97%, surpassing the results of previous studies.


Sign in / Sign up

Export Citation Format

Share Document