With the aid of satellite remote sensing imagery, it is now possible to monitor the seawater with adequate temporal and spatial coverage for marine pollution and biological productivity. In this paper, we present case studies on two major coastal oil spills, which have been captured by the remote sensing imagery. The detection and tracking of the oil slicks are achieved by means of twodimensional Gaussian-based wavelet analysis. This work demonstrates that this technique is a IIIseful and promising tool for monitoring of coastal waters.
Satellite remote sensing technology can provide sea surface data with far better spatial and temporal coverage than that of the conventional in situ measurements. With the aid of the satellite remote sensing imagery, it is now possible to monitor the seawater, particularly coastal water, with adequate temporal and spatial coverage for marine pollution and biological productivity. In this paper, we present case studies on two major coastal oil spills, which have been captured by the SAR (synthetic aperture radar) and the AVHRR (advanced very-high resolution radiometer) sensors on board several satellites. In both cases, the dispersion route and pattern have been identified by means of two-dimensional Gaussian-based wavelet analysis.
In a nutshell, the two-dimensional wavelet transform is a form of spatial differentiation. The resulting image is differential image with accentuated amplitude changes. Furthermore, the Laplacian of Gaussian wavelet transform, one of the two employed in this analysis, works as a band-passed filter. Ideally only features with length scale within a chosen band are retained upon transformation. Two types of wavelets are used in this analysis, the first order derivative and the second order derivative of a two·dimensional Gaussian function. The wavelet·transformed images, therefore, reflect the properties of the spatial first order derivative and second order derivative of the original image, respectively.
