Latent Dirichlet allocation for image segmentation and source finding in radio astronomy images

We present exploratory work into the application of the topic modelling algorithm latent Dirichlet allocation (LDA) to image segmentation in greyscale images, and in particular, source detection in radio astronomy images. LDA performed similarly to the standard source-detection software on a representative sample of radio astronomy images. Our use of LDA underperforms on fainter and diffuse sources, but yields superior results on a representative image polluted with artefacts --- the type of image in which the standard source-detection software requires manual intervention by an astronomer for adequate results.

[1]  A. Koekemoer,et al.  Deep ATLAS Radio Observations of the Chandra Deep Field-South/Spitzer Wide-Area Infrared Extragalactic Field , 2006 .

[2]  W. Eric L. Grimson,et al.  Spatial Latent Dirichlet Allocation , 2007, NIPS.

[3]  Marc Masias Moyset,et al.  A review of source detection approaches in astronomical images , 2012 .

[4]  Fei-Fei Li,et al.  Spatially Coherent Latent Topic Model for Concurrent Segmentation and Classification of Objects and Scenes , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[5]  Jitendra Malik,et al.  A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[6]  Alexei A. Efros,et al.  Discovering object categories in image collections , 2005 .

[7]  Thomas L. Griffiths,et al.  Probabilistic Topic Models , 2007 .

[8]  D. Wells,et al.  Fits: a flexible image transport system , 1981 .

[9]  David L. Olson,et al.  Advanced Data Mining Techniques , 2008 .

[10]  R. D. Ekers,et al.  ATLBS: the Australia Telescope Low-brightness Survey , 2009, 0911.4571.

[11]  Pietro Perona,et al.  Learning object categories from Google's image search , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[12]  Matthew T. Whiting,et al.  duchamp: a 3D source finder for spectral‐line data , 2012, 1201.2710.

[13]  Alexei A. Efros,et al.  Discovering objects and their location in images , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[14]  C. Elkan,et al.  Topic Models , 2008 .

[15]  Mehran Sahami,et al.  Text Mining: Classification, Clustering, and Applications , 2009 .

[16]  R. D. Ekers,et al.  ATLBS EXTENDED SOURCE SAMPLE: THE EVOLUTION IN RADIO SOURCE MORPHOLOGY WITH FLUX DENSITY , 2012, 1202.4516.

[17]  Alexei A. Efros,et al.  Using Multiple Segmentations to Discover Objects and their Extent in Image Collections , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[18]  Antonio Torralba,et al.  Learning hierarchical models of scenes, objects, and parts , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[19]  Pietro Perona,et al.  A Bayesian hierarchical model for learning natural scene categories , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[20]  T. Budavari,et al.  Radio Continuum Surveys with Square Kilometre Array Pathfinders , 2012, Publications of the Astronomical Society of Australia.

[21]  D. J. Saikia,et al.  EMU: Evolutionary Map of the Universe , 2011, Publications of the Astronomical Society of Australia.

[22]  David M. Blei,et al.  Probabilistic topic models , 2012, Commun. ACM.

[23]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[24]  Luc Van Gool,et al.  Modeling scenes with local descriptors and latent aspects , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[25]  Christopher Hollitt,et al.  Feature Detection in Radio Astronomy using the Circle Hough Transform , 2012, Publications of the Astronomical Society of Australia.