Large radio surveys, such as those proposed for the Australian SKA Pathfinder (ASKAP), require excellent source-finding algorithms. Here, we discuss advanced methods for finding sources in spectral line, radio continuum and polarisation surveys. We also look into data visualisation to explore the survey results together with their multi-wavelength counterparts.

Detection thresholds in polarized intensity and polarization bias correction are investigated for surveys where the polarization information is obtained from RM synthesis. The false detection rate of sources in polarized intensity is investigated for images with non-Gaussian noise in Stokes Q and U.

We have developed a method for simultaneously finding and fitting HI absorption lines in radio data by using multi-nested sampling, a Bayesian Monte Carlo algorithm. The method is tested on a simulated ASKAP data cube, and is shown to be produce reliable detections in low signal-to-noise data.

In this paper, we implement a 2D–1D wavelet decomposition to obtain an efficient way of denoising data cubes from spectroscopic imaging surveys. We conduct different simulations to evaluate the usefulness of the algorithm as part of a source finding pipeline.

We have compared duchamp and manual identification methods for identifying masers in the Galactic plane survey HOPS. We find noise in the data is the limiting factor determining efficiency of both methods. duchamp may be useful for very large datasets, but the manual method is quicker for smaller datasets.

We discuss optimal source detection with multiwavelength imaging data using a color-neutral approach representing a more general version of the so-called ‘chi-squared’ technique. We discuss its performance with four-band data from the WISE mission and suggest applications to the multifrequency data cubes of the ASKAP surveys.

We discuss a method to determine the reliability of source finders within the context of upcoming Hi surveys. Assuming that the noise is symmetric and that real sources have positive total flux, we use sources with negative flux to assign to each positive detection a probability of being real.

This work presents a method for determining the accuracy of a source finder algorithm for spectral line radio astronomy data and the Source Finder Accuracy Evaluator, a program that implements this method.

Existing telescopes rely on human expertise to reduce raw data and then analyse the images for non-pointlike objects; however, the next generation of radio telescopes will make this infeasible. We explore the Circle Hough Transform as a detection method on a sample of different extended circular or arc-like astronomical objects.

We present a framework to interactively volume-render 3D data cubes over a cluster of workstations powered by GPUs and a multi-core CPU. Our target is to provide 3D interactive views of terabyte-sized data cubes. The framework proved to be scalable to render a 204 GB data cube with 30 fps.

In this paper we compare four different methods of pre-processing ASKAP Hi spectral line observations to improve source finder performance. We test two conventional methods, linear smoothing and wavelet reconstruction (as implemented by Duchamp). The other two methods we test are iterative median smoothing and mathematical morphology opening subtraction. Both of these methods are not typically used.

We present algorithms used to enhance the source-finding capabilities of duchamp for use in the processing pipelines for the Australian Square Kilometre Array Pathfinder (ASKAP). These algorithms cover parallel processing, variable detection thresholds and two-dimensional fitting. We also discuss the development process and community involvement therein.