Ocean pollution by oil slicks is one of the major environmental hazards. Oil tanker accidents (such as Exxon Valdez, Erika, and Prestige), although spectacular and highlighted, are responsible for only 5% of the total oil pollution worldwide, 95% coming from illegal discharges [4]. In the Mediterranean Sea, oil pollution monitoring is normally carried out by aircrafts or ships. This is expensive and is constrained by the limited availability of these resources. In

order to provide all-weather and global monitoring of such events, spaceborne synthetic aperture radar (SAR) has been recognized as a cornerstone. Radar imagery can provide a significant contribution for this field, identifying probable spills and tracking ships’ routes over very large areas, then guiding aerial surveys for precise observations in specific locations. Automatic detection of oil spills in SAR images has been widely researched, with specific focus being given to the detection and classification of oil spills. In this chapter an algorithm for oil-spill detection and classification is proposed whose block scheme is shown in Figure 8.1 and whose main steps are

1. Despeckling, which aims at reducing the speckle noise on the SAR image for its better interpretation

2. Image enhancement 3. Image segmentation 4. Oil-spill classification 5. Oil-spill shape features extraction

A more detailed description of these steps is provided in the following sections. Results of the application of such algorithm to COSMO-SkyMed SAR images are shown in Section 8.6.