We present the Final Analysis of the European Large Area ISO Survey (ELAIS) 15 $\mu$m observations, carried out with the ISOCAM instrument on board the Infrared Space Observatory (ISO). The data reduction method, known as LARI method, is based on a mathematical model of the detector's behaviour and was specifically designed for the detection of faint sources in ISO-CAM/PHOT data. The method is fully interactive and leads to very reliable and complete source lists. The resulting catalogue includes 1923 sources detected with $S/N > 5$ in the {0.5 -- 100 mJy} flux range and over an area of 10.85 {deg$^2$} split into four fields, making it the largest non-serendipitous extragalactic source catalogue obtained to date from ISO data. This paper presents the concepts underlying the data reduction method together with its latest enhancements. The data reduction process, the production and basic properties of the resulting catalogue are discussed. The catalogue quality is assessed by means of detailed simulations, optical identifications and comparison with previous analyses.

We have performed 23 GHz VLA observations of 7 compact, luminous infrared galaxies, selected to have evidence of starburst activity. New and published multi-frequency data are combined to obtain the spectral energy distributions of all 7 galaxies from the near-infrared to the radio (at 1.4 GHz). These SEDs are compared with new models, for dust enshrouded galaxies, which account for both starburst and AGN components. In all 7 galaxies the starburst provides the dominant contribution to the infrared luminosity; in 4 sources no contribution from an AGN is required. Although AGN may contribute up to 50 percent of the total far--infrared emission, the starbursts always dominate in the radio. The SEDs of most of our sources are best fit with a very high optical depth of (>=50) at 1 micron. The scatter in the far-infrared/radio correlation, found among luminous IRAS sources, is due mainly to the different evolutionary status of their starburst components. The short time-scale of the star formation process amplifies the delay between the far-infrared and radio emission. This becomes more evident at low radio frequencies (below about 1 GHz) where synchrotron radiation is the dominant process. In the far-infrared (at wavelengths shorter than 100 micron) an additional source of scatter is provided by AGN, where present. AGN may be detected in the near-infrared by the absence of the knee, typical of stellar photospheres. However, near-infrared data alone cannot constrain the level at which AGN contribute because the interpretation of their observed properties, in this wave-band, depends strongly on model parameters.

We are surveying the region -87 < delta < -38.5 for multiple imaging at scales 1 < theta < 15 arsec. Candidate gravitational lenses were selected from an existing dataset of Australia Telescope Compact Array measurements of 2178 flat (4.80/8.64 GHz) spectrum PMN sources. Selection criteria were optimised using tests on simulated lens datasets.