Machine learning and image analysis for morphological galaxy classification

Calleja, Jorge de la; Fuentes, Olac

doi:10.1111/j.1365-2966.2004.07442.x

Cited by 93 publications

(47 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other approaches to this problem attempt to forgo any form of handcrafted feature extraction by applying principal component analysis (PCA) to preprocessed images in combination with a neural network (De La Calleja & Fuentes 2004), or by applying kernel SVMs directly to raw pixel data (Polsterer et al 2012).…”

Section: Related Workmentioning

confidence: 99%

Rotation-invariant convolutional neural networks for galaxy morphology prediction

Dieleman¹,

Willett²,

Dambre³

2015

Monthly Notices of the Royal Astronomical Society

705

604

View full text Add to dashboard Cite

Measuring the morphological parameters of galaxies is a key requirement for studying their formation and evolution. Surveys such as the Sloan Digital Sky Survey (SDSS) have resulted in the availability of very large collections of images, which have permitted population-wide analyses of galaxy morphology. Morphological analysis has traditionally been carried out mostly via visual inspection by trained experts, which is time-consuming and does not scale to large ( 10 4 ) numbers of images.Although attempts have been made to build automated classification systems, these have not been able to achieve the desired level of accuracy. The Galaxy Zoo project successfully applied a crowdsourcing strategy, inviting online users to classify images by answering a series of questions. Unfortunately, even this approach does not scale well enough to keep up with the increasing availability of galaxy images.We present a deep neural network model for galaxy morphology classification which exploits translational and rotational symmetry. It was developed in the context of the Galaxy Challenge, an international competition to build the best model for morphology classification based on annotated images from the Galaxy Zoo project.For images with high agreement among the Galaxy Zoo participants, our model is able to reproduce their consensus with near-perfect accuracy (> 99%) for most questions. Confident model predictions are highly accurate, which makes the model suitable for filtering large collections of images and forwarding challenging images to experts for manual annotation. This approach greatly reduces the experts' workload without affecting accuracy. The application of these algorithms to larger sets of training data will be critical for analysing results from future surveys such as the LSST.

show abstract

Section: Related Workmentioning

confidence: 99%

Rotation-invariant convolutional neural networks for galaxy morphology prediction

Dieleman¹,

Willett²,

Dambre³

2015

Monthly Notices of the Royal Astronomical Society

705

604

View full text Add to dashboard Cite

show abstract

“…Palomar images are available in three passbands of the Thuan-Gunn system: g, r, and i. [22] Lowell images are in two passbands (J and R) of the filter system developed by [23] (4), (5). Thus, It can be assumed that the classification accuracy can be improved when using datasets with less noise and solving the overlapping problem.…”

Section: Resultsmentioning

confidence: 99%

“…[4], used supervised ANN to classify galaxies, Difference Boosting Neural Networks was able to learn 98.3% of the galaxies correctly and identify 89.9% of galaxy images in a test set; their challenge was to develop a supervised classifier capable of sorting galaxies into subclasses using manually threshold images. [5], Presented an experimental study of machine learning and image analysis for performing automated morphological galaxy classification. They used a neural network, and a locally weighted regression method, and implemented homogeneous ensembles of classifiers.…”

Section: Introductionmentioning

confidence: 99%

Galaxy Image Classification using Non-Negative Matrix Factorization

Selim¹,

Keshk²,

Shourbugy³

2016

IJCA

View full text Add to dashboard Cite

In modern astronomy with the advent of astronomical imaging technology developments and the increased capacity of digital storage, lead to the production of photographic atlases of data which need to be processed autonomously. Galaxies morphology is an important topic to understand questions concerning the evolution and formation of galaxies and their content. In this work, morphological classification of galaxies is presented using a new method based on Non-Negative matrix factorization for images of galaxies in the Zsolt frei Catalog. The algorithm is trained using manually classified images of elliptical, spiral and lenticular galaxies. Experimental results show that galaxy images from Zsolt Frei catalog can be classified automatically with an accuracy of 93 percent compared to classifications carried out by other authors and manually classified.

show abstract

“…Galaxies can be considered as massive system of stars, gas, dust, and other forms of matter bound together gravitationally as a single physical unit [5]. The morphology of galaxies is generally an important issue in the large scale study of the Universe [1].…”

Section: Introductionmentioning

confidence: 99%

Principal Component Analysis Neural Network Hybrid Classification Approach for Galaxies Images

Elfattah

El-Bendary

Elsoud

et al. 2014

Innovations in Bio-Inspired Computing and Applications

View full text Add to dashboard Cite

Abstract. This article presents an automatic hybrid approach for galaxies images classification based on principal component analysis (PCA) neural network and moment-based features extraction algorithms. The proposed approach is consisted of four phases; namely image denoising, feature extraction, reduct generation, and classification phases. For the denoising phase, noise pixels are removed from input images, then input galaxy image is normalized to a uniform scale and Hu seven invariant moment algorithm is applied to reduce the dimensionality of the feature space during the feature extraction phase. Subsequently, for reduct generation phase, attributes in the information system table that is more important to the knowledge is generated as a subset of attributes. Rough set is used as feature reduction approach. The subset of attributed, which is called a reduct, is fully characterizing the knowledge in the database. Finally, during the classification phase, principal component analysis neural network algorithm is utilized for classifying the input galaxies

show abstract

Machine learning and image analysis for morphological galaxy classification

Cited by 93 publications

References 9 publications

Rotation-invariant convolutional neural networks for galaxy morphology prediction

Rotation-invariant convolutional neural networks for galaxy morphology prediction

Galaxy Image Classification using Non-Negative Matrix Factorization

Principal Component Analysis Neural Network Hybrid Classification Approach for Galaxies Images

Contact Info

Product

Resources

About