Abstract:
Based on the advantages of wavelet transform compression sensing, i.e, the sparsity of wavelet coefficients and the unconstraint of Nyquist sampling theorem, Here, the wavelet transform compression sensing is applied to the fractal image processing, and a reconstruction method of metal fracture image based on wavelet compression sensing is proposed. In the proposed method, the fractal image is sparsely sampled by wavelet transform, and then the random measurement matrix is designed to compress the fractal image. Finally, the fractal image is reconstructed by OMP or ROMP algorithm. At the same time, the two reconstruction algorithms are compared. The results show that the two reconstruction algorithms can obtain better reconstructed images when the compression ratio reaches a certain level. Comparing with the OMP algorithm, the ROMP reconstruction algorithm has higher peak signal-to-noise ratio (PSNR), and the reconstruction time of ROMP algorithm is also greatly shortened and the performance is stable.