Thuong Nguyen Canh

Multi-scale/multi-resolution Kronecker compressive imaging.

As a universal sampling procedure, compressive sensing (CS) considers that all samples of compressible signal are equally important. However, it is not true in in image/video signal since human visual system is more sensitive to low frequency components. Therefore, CS theory has been extended to hybrid and multi-scale CS to better capture the low-frequency samples. The computational complexity is another challenge in CS which can be solved by multi-resolution sensing matrix. In this paper, we propose a multi-scale/multi-resolution sensing matrix for Kronecker CS (KCS) based on separable wavelet transform and address measurement allocation problem with and without information of to-be-sensed image. The proposed methods not only perform better (3.72dB gain) but also low complexity and compatible with conventional reconstruction methods.

Total variation reconstruction for Kronecker compressive sensing with a new regularization

Recovery algorithm based on total variation (TV) has shown its capability to recover high quality image in compressive sensing by preserving edges well but not fine details and textures. Recently, to improve this deficiency, characteristics of natural images are further utilized by adding some regularization terms into its recovery problem. In these efforts, this paper proposes a new regularization exploiting nonlocal properties of image using the nonlocal means filter in the gradient domain instead of the spatial domain. The Split Bregman method is applied to solve a combination of total variation and a new regularization term under the framework of Kronecker compressive sensing. Numerical experiments with the proposed and related regularizations verify significant improvement of the proposed method in term of both objective and subjective qualities.

Detail-preserving compressive sensing recovery based on cartoon texture image decomposition

In this paper, we propose a detail-preserving reconstruction method for total variation-based recovery in low subrate compressive sensing using cartoon texture image decomposition and residual reconstruction. It iteratively decomposes and reconstructs cartoon and texture image components separately. A nonlocal structure-preserving filter is utilized to reduce staircase artifacts while preserving nonlocal structures of image in the spatial domain. Experimental results show that the proposed method outperforms the conventional ones in terms of preserving small scale details of image.

Multi-Resolution Kronecker Compressive Sensing

From the perspective of reducing the sampling cost of color images at high resolution, block-based compressive sensing (CS) has attracted considerable attention as a promising alternative to conventional Nyquist/Shannon sampling. On the other hand, for storing/transmitting applications, CS requires a very efficient way of representing the measurement data in terms of data volume. This paper addresses this problem by developing a measurement-coding method with the proposed customized Huffman coding. In addition, by noting the difference in visual importance between the luma and chroma channels, this paper proposes measurement coding in YCbCr space rather than in conventional RGB color space for better rate allocation. Furthermore, as the proper use of the image property in pursuing smoothness improves the CS recovery, this paper proposes the integration of a low pass filter to the CS recovery of color images, which is the block-based -norm minimization. The proposed coding scheme l 20 shows considerable gain compared to conventional measurement coding.

Compressive sensing reconstruction via decomposition

When recovering images from a small number of Compressive Sensing (CS) measurements, a problem arises whereby image features (e.g., smoothness, edges, textures) cannot be preserved well in reconstruction, especially textures at small-scale. Since the missing information still remains in the residual measurement, we propose a novel Decomposition-based CS-recovery framework (DCR) which utilizes residual reconstruction and state-of-the-art filters. The proposed method iteratively refines residual measurement which is closely related to the denoise-boosting techniques. DCR is further incorporated with a weighted total variation and nonlocal structures in the gradient domain as priors to form the proposed Decomposition based Texture preserving Reconstruction (DETER). We subsequently demonstrate robustness of the proposed framework to noise and its superiority over the other state-of-the-art methods, especially at low subrates. Its fast implementation based on the split Bregman technique is also presented. HighlightsImage features (edge, smooth, texture regions) have different characteristics.A decomposition based CS reconstruction is proposed via denoising filters.An instance of image denoise boosting techniques in CS reconstruction.The better recovery and filter methods, the higher performance is archived.Multiple image priors improve the final CS reconstruction performance.

Edge-preserving nonlocal weighting scheme for total variation based compressive sensing recovery

Although total variation minimization technique is being widely used in compressive sensing recovery, it still suffers from the so called staircase artifact which is caused by losing fine details of image. As a solution for the problem, in this paper, we propose an edge-preserving weighting scheme utilizing nonlocal structure and histogram of natural image in the gradient domain. Experimental results show that the proposed scheme surpasses the traditional total variation and the edge-guided CS in both objective and subjective qualities.

Color Image Denoising via Cross-Channel Texture Transferring

Image denoising can reduce the perturbation inevitably generated during image signal acquisition and its subsequent processing. While the utilization of nonlocal properties can enhance the performance of the state-of-the-art denoising methods, a heavy computational burden is incurred especially for color images. Inspired by the high correlation in the texture information over color channels, for a reduction of the computational burden, this letter proposes denoising the luma channel first, and then, performing a patch-wise linear prediction to transfer the texture information of the denoised luma channel to the other two channels. The texture transferring is adapted to local characteristic (i.e., variance of the local patches) for a reduction of color smearing caused by large prediction error especially along edges. Experimental results confirm that the proposed method achieves performance improvement over the state-of-the-art color image denoising methods only at a slightly increased complexity of single-channel denoising.

Compressive sensing of video with weighted sensing and measurement allocation

This paper proposes a compressive sensing of video (CSV) framework that utilizes statistical properties of video signal. The proposed scheme periodically acquires deterministic measurements (i.e., low frequency DCT coefficients) for key frames to improve recovering both key and nonkey frames. In addition, based on temporal correlation among frames, side information is generated for nonkey frames to model their important coefficients and sparsity in transform domain. This information helps better sensing by weighted sensing and measurement allocation. Experimental results show effectiveness of the proposed techniques and their significant improvement compared to prior work.

Filter-aided recovery for block-based compressive sensing of images

This paper proposes a recovery algorithm for block-based compressive sensing of images, which iteratively performs a compressive sensing recovery and a denoising processes. The compressive sensing recovery is performed by MH-BCS-SPL, whereas the denoising is by BM3D. Experimental results show the superiority of the proposed recovery algorithm.

Hybrid Kronecker compressive sensing for images

Natural images has certain level of both similarity and difference which can be efficiently represented by deterministic and random sensing matrices in compressive sensing. In this context, a hybrid sensing matrix which combines a deterministic DCT and a random matrix, is recently investigated. In this paper, we bring the concept of hybrid sensing matrix into Kronecker compressive sensing (KCS) of images. Extensive experiment has shown that the proposed hybrid KCS method performs better than either fully random or deterministic DCT matrix, and comparatively with other state-of the-art sensing schemes in terms of reconstruction quality.