Modified Curvelet Thresholding Algorithm for Image Denoising

Problem statement: This study introduced an adaptive thresholding method for removing additive white Gaussian noise from digital images. Approach: Curvelet transform employed in the proposed scheme provides sparse decomposition as compared to the wavelet transform methods which being nongeometrical lack sparsity and fail to show optimal rate of convergence. Results: Different behaviors of curvelet transform maxima of image and noise across different scales allow us to design the threshold operator adaptively. Multiple thresholds depending on the scale and noise variance are calculated to locally suppress the curvelet transform coefficients so that the level of threshold is different at every scale. Conclusion/Recommendations: The proposed algorithm succeeded in providing improved denoising performance to recover the shape of edges and important detailed components. Simulation results proved that the proposed method can obtain a better image estimate than the wavelet based restoration methods.