Texture entropy algorithm for automatic detection of oil spill from RADARSAT-1 SAR data

This work presents a method based on the utilisation of texture algorithms for the discrimination of oil spill areas from the surrounding features, e. g. sea surface and look-alikes, using RADARSAT-1 SAR Wide beam mode (W1), Standard beam mode (S2) and Standard beam mode (S1) data acquisition under different wind speeds. The results show that entropy texture algorithm is able to discriminate between oil spills and look-alike areas. The results also illustrate that the entropy texture algorithm identifies well the deficiency of oil spills in pairs of S2 data. Further, the W1 mode data, however, show an error standard deviation of 0.002, thus performing a better discrimination of oil spills than the S1 and S2 mode data.