Abstract:Na ve simulated additive white Gaussian noise (AWGN) may not fully characterize the complexity of real world noisy images. Owing to optimal sparsity in image representation, we propose a curvelet based model for denoising real-world RGB images. Initially, the image is decomposed in three curvelet scales, namely: the approximation scale (that retains low-frequency information), the coarser scale and the finest scale (that preserves high-frequency components). Coefficients in the approximation and finest scale are estimated using NLM filter, while a scale dependent threshold is adopted for signal estimation in the coarser scale. The reconstructed image in spatial domain is further processed using Guided Image Filter (GIF) to suppress the ringing artifacts due to curvelet thresholding. The proposed approach known as CTuNLM method is extended for color image denoising using uncorrelated YUV color space. Extensive experiments on multi-channel real noisy images are conducted in comparison with eight sate-of-the-art methods. With four encouraging qualitative and quantitative measures including PSNR and SSIM, we found that CTuNLM method achieves better denoising performance in terms of noise reduction and detail preservation. We further examined the potential of proposed approach by focusing only on the Finest scale curvelet Coefficients (FC). Features like small details, edges and textures always add up to improve the overall denoising performance, while minimizing spurious details. We studied "The Curious Case of the Finest Scale" and constructed "Deep Curvelet-Net": an encoder-decoder-based CNN architecture, as a pilot work. The encoder uses multiscale spatial characteristics from noisy FC, while the decoder processes de-noised FC under the supervision of encoder's multiscale spatial attention map. The "Deep Curvelet-Net" links encoder multiscale feature modeling with decoder spatial attention supervision to learn the most essential features for denoising. The CNN-based architecture only estimates FC, while all other CTuNLM stages are left unchanged to produce the denoised output. Results presented in this article validated the design of proposed CNN architecture in curvelet domain and motivated us to search beyond classical thresholding and/or filtering approaches.

A Novel Median-Contourlet for Image Denoising Application

A Novel Improved Median Filter for Salt-and-pepper Noise from Highly Corrupted Images

Median filters combined with denoising convolutional neural network for Gaussian and impulse noises

A Fast and Reliable Switching Median Filter for Highly Corrupted Images by Impulse Noise

A median filter based on the proportion of the image variance

Modified directional weighted filter for removal of salt & pepper noise

Adaptive Window Multistage Median Filter for Image Salt-and-pepper Denoising

A New Directional Weighted Median Filter for Removal of Random-Valued Impulse Noise

Image Denoising Method Based on Nonsubsampled Contourlet Transform and Bandelet Transform

Medical Image Processing by Denoising and Contour Extraction

Image Denoising Based on Median Filter and UNet

An Efficient Switching Median Filter Based on Local Outlier Factor.

Patch-based contour prior image denoising for salt and pepper noise

Multi-Scale Based Approach for Denoising Real-World Noisy Image Using Curvelet Thresholding: Scope and Beyond

Image Denoising Based on Complex Contourlet Transform

Impulsive Noise Filter Using Median-And Partition-Based Operation

An Improved Image Denoising Method Based On Contourlet Transform And Neighshrink Algorithm

Image denoising method integrating ridgelet transform and improved wavelet threshold

An Image Denoising Algorithm Based on Clustering and Median Filtering

A Novel Image Denoising Algorithm Based On Dual-Tree Complex Wavelets And Ridgelet Transform

Image Denoising by Random Interpolation Average with Low-Rank Matrix Approximation