HIDEEP was a deep neutral hydrggen (HI) survey carried out with the multibeam receiver on the 64-m Parkes telescope in Australia. It covered a total area of 60 sq. deg. (6° × 10°), reaching full sensitivity over the central 32 sq. deg. (4° × 8°. In this central region, the survey had an integration time of 9000 s beam-1, giving a column-density sensitivity (for a velocity width of 20 km s-1) of 7.4 × 1016 cm-2. This made it one of the most sensitive surveys ever carried out for low column-density gas. (The only more sensitive survey was the WSRT wide-field HI survey of Braun et al. (2003), which used the 14 25-m antennae of the Westerbork telescope as single-dish telescopes.)

HIDEEP detected 173 HI sources, all of which had optical counterparts. Based on the sizes of the optical counterparts, the column-densities of the sources followed the relationship NHI = 1020.65 ±0.38 cm-2. The survey did not discover any free-floating low-column-density HI clouds without optical counterparts, nor did it discover any Malin 1-like giant low-surface-brightness (LSB) galaxies that were hidden from shallower HI surveys (such as HIPASS) due to their low column-density.

The central 36 sq. deg. (6° × 6°) was observed in the optical R-band using a stack of eight Tech Pan films, each exposed for 1 hour. This region covered 129 of the 173 sources. From this region, it was determined that while LSB galaxies provided 30 ± 10 per cent of the contribution of gas-rich galaxies to the neutral hydrogen density of the Universe, they only 7 ± 3 per cent of the gas-rich galaxies' contribution to luminosity density (and thus an even smaller proportion of the contribution of all galaxies). The survey also found that low surface-brightness galaxies contributed 40 ± 20 per cent of the cross-sectional area to quasar absorption lines, a finding confirmed (and improved on) with WSRT measurements by Zwaan et al. (2005), who estimated that 44% of the cross-section would come from LSB galaxies.