Image Fusion for Underwater Images using Curvelet Transform

In the real world noise is often presented in the signal. Noise in the image is inserted during the process of image acquisition or the transmission process of image. Many parameters or factors like light levels and sensor temperature are there which are responsible the for the amount of noise present in the image. Acquiring clear images in underwater environments is an important issue in ocean engineering. Generally underwater images are affected by severe blur and haze caused by light that is reflected from surface and scattered by water particles and color change due to varying degrees of light attenuation for different wavelengths. So there is contrast losses and color deviation in images. In general underwater images are polluted by noises like Gaussian, and Speckle noise. So it is necessary that to denoise the image before further process. There are different filters to denoise the image like Gaussian Filter,Mean Filter, Weiner Filter and Curvelet Transform can be used. After that Fusion is performed of best filtering method and curvelet transform using Simple Average and Average using CVT. Performance can be calculated on the basis of parameter like PSNR (peak signal-to-noise ratio)& MSE (mean square error),MAE (Mean Absolute Error), Correlation. MATLAB is used for implementation of coding and to calculate the output parameter.