ecs.vuw.ac.nz

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 im-ages. Our use of LDA underperforms on fainter and diffuse sources, but yields superior results on a representative im-age polluted with artefacts — the type of image in which the standard source-detection software requires manual in-tervention by an astronomer for adequate results