Faint compact object detection in wide field interferometric radio images using a hybrid method based on local thresholding and wavelet decomposition.

J. M. Pareces investigador correspondiente del IAR.

Proceedings of the conference held 23-26 September, 2007, in Kensington Town Hall, London, United Kingdom. Edited by Robert W. Argyle, Peter S. Bunclark, and James R. Lewis.

Resumen: Automated detection of compact objects in wide field images has classically been produced using thresholding techniques based on local noise estimation. More recently Multiscale Vision Models that use wavelet decomposition have been proposed for the same aim. We propose a hybrid method where both techniques are used at different stages: In a first step bright sources are detected using classical local thresholding and a residual image that does not contain them is produced. In a second step wavelet decomposition is applied to the residual image in order to detect faint compact objects. We show the obtained results using a wide field radio map of the Cygnus OB2 region obtained at 49 cm with the GMRT interferometer and compare them with the published catalogue produced from this image.