The detection of small surface targets at sea is an increasing requirement for warships. The present sensors on board do not provide the required detection probabilities for these low observable targets like small rubber boats, floating mines, periscopes, people etc. The reason for the low probability of detection is not only the low contrast but also the severe clutter problem and occlusion behind waves in higher sea states and light level problems at night. Analysis of the features of the various sources of false alarms like birds, whitecaps and sunglints in the spectral bands from the UV to the thermal IR has learned about possibilities for algorithms for discrimination of the red targets. These algorithms are based on the ratios of spectral reflectance/emission in different bands and spatial and temporal effects. These algorithms are weather and direction dependent (wind, sun and observation direction). The knowledge of sea background characteristics for target detection (surface targets as well as low altitude airborne targets) has been increased in recent years thanks to measurement campaigns with advanced sensors such as LAPTEX (Greece) and thanks to the model improvements. Examples of imagery will be shown as well as results of various processing algorithms.