Radio-astronomical image reconstruction with a conditional denoising diffusion model

Open Access

Table 3

Comparative results for the fraction of the detected sources whose estimated flux is within 50 µJy of the true flux.

Model	Normalization	Aggregation

		Mean	Median	Single run	Detect-aggregate
CLEAN+PyBDSF	–			0.57
	γ = 1	0.85	0.66	0.048	0.015
	γ = 2	0.96	0.97	0.88	0.97
	γ = 10	0.92	0.93	0.71	0.89
Diffusion	γ = 20	0.82	0.83	0.53	0.72
	γ = 30	0.74	0.75	0.47	0.66

Notes. The detect-aggregate method results are reported for a reliability threshold of 30%. We compare our results with PyBDSF. Italic font indicates the optimal metric value across all conducted experiments.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.