The Research On Image Compression Based On Curvelet Transform

The algorithm of image compression has been a very popular field of study. After years of research, there has been a lot of mature technology of image compression. Wavelet transform has achieved great development in signal processing, which profits from its time-frequency analysis and near-ideal ability to represent one dimension function of bounded variation. It's a pity that wavelet analysis can't take full advantage of geometrical characteristic of the data to get the sparse representation of the function.Curvelet analysis is a new methods of oriented multiscale method,which was develop from wavelet and ridgelet analysis. Rideglet transform,which has one more parameter-orientation,besides the scale and location,is especially suitable for representing two dimensions signals with linear or hyperplane singularities,and has high approach precision. Curvelet transform is firstly proposed by using the partitioning method and the ridgelet. It is good for describing the curve singularities singals.Nowadays,Curvelet Transform has been widely used in image processing.However,there is very little algorithm of Curvelet transform used in image compression .In this paper,we firstly introduces the basic knowledge of Curvelet theory;then,we put forword a image compression algorithm based on Curvelet transform and do some researches on the algorithm. The algorithm is based on the second generation of the Curvelet transform which named Curvelet implementation based on frequency domain.By the experiment,we can see that the result of image compression based on Curvelet transform is realy good. Beacuse the process of the transform product a lot of parameters ,it takes much time to do compute. In this paper ,we improve the algorithm to solve this problem. We change the Fourier transform into Fast Fourier transform. By the experiment ,we know that ,this change significantly improve the efficiency of the implementation of the procedures. In addition,this algorithm reduce computation,improve the decoding speed and get better edge parts representation.