Research Of Curvelet Transform In Digital Image Denosing And Compression

Compared with Fourier analysis, wavelet analysis can give more "sparse" in one-dimensional piecewise smooth function or bounded variance function. In terms of one-dimensional signals, wavelet representation is optimal. While in terms of image signals, curvelet transform is more suitable to analyze the curve or straight edge features than wavelet, since curvelet transform approximates with higher accuracy, better sparse expression, can achieve more excellent effect.A new threshold method of compensating energies at different levels is proposed in this paper, on the basis of comparing different methods of choosing the thresholds of curvelet denoising. This method has been applied to image denoising, which shows better experimental results, and the peak signal to noise ratio and visual effects of denoised image has improved greatly.At the same time, in the process of studying the application of curvelet transform in image compression, by choosing appropriate classification to images and reordering the coefficients experiencing, curvelet transform improves the processing speed of curvelet transform and the performance of compressing images. Experiments show the image compression coding based on the wavelet transform has a higher peak signal to noise ratio, as well as a better visual effect after decoding.Whether the curvelet transform is used in image denoising or image compression, this paper is of great significance to practical application, also providing the basis for the application of curvelet transform in other image processing fields.