Search

Image inpainting is a classical yet challenging inverse ill-posed problem. In this paper, we introduce the multi-filters guided low-rank tensor coding as a prior information to tackle it. The key innovation is to formulate multiple feature-domain tensors by convoluting the target image with multi-filters. Furthermore, by exploring a low-rank tensor coding, it can reduce the redundancy between sparse feature vectors at neighboring locations and improve the efficiency of the overall representation. The resulting non-convex model is iteratively tackled by gradient descent procedure for updating the image and by low-rank pursuit procedure for updating the multi-view features. Besides, we explore an aggregation version of proposed method for further improving the inpainting performance. Experimental results demonstrate that the proposed algorithms can faithfully recover image and outperform the current state-of-the-art approaches in terms of visual inspection, the quantitative peak signal-to-noise ratio (PSNR), and structural similarity (SSIM).

Nowadays, with the rapid spread of Corona Virus Disease 2019 (COVID-19), this epidemic has become a threatening risk for global public health. Medical workers and researchers all over the world are struggling against the novel coronavirus in the front line. Because the computed tomography (CT) images from infected patients exposure characteristic abnormalities, automatic CT analyzers based on AI-based algorithms are extensively employed as effective weapons to aid clinicians. However, unbalanced data and lack of annotations obstruct AI-based algorithms applying in aided diagnosis because of their low performance. Therefore, in order to solve the above problems, a general-purpose solution is proposed to synthesize COVID-19 CT scans from non-COVID-19 data for providing high-quality negative-positive paired CT scans. Particularly, we introduce an elastic registration algorithm of CT images to manufacture paired training data. Then, a conditional Generative Adversarial Networks (GANs) based image-to-image translation model is implemented to synthesize COVID-19 CT scans from non-COVID-19 data. The effectiveness of our proposed algorithm used in COVID-19 aided diagnosis is verified in the experiments, and the identification and detection capacities of the classification models have been enhanced with the generated CT scans. Specifically, the precise lesion location is achieved by the generated data with a weakly supervised algorithm of class activation mapping (CAM). The model and code of this paper are publicly available at https://github.com/lihengbit/Synthesis-of-COVID-19-CT-Scan.

Most low-rank tensor approximations are NP-hard problems. In this paper, we introduce a novel concept: field-of-experts (FoE) filters guided tensor completion, which aims to integrate the strengths of the emerging tensor completion method and the conventional FoE filters. Specifically, the target image is convolved by FoE filters to produce multiview features as a high-order tensor, which captures complementary information from multiple views. In order to impose the concept, we employ two strategies to model the new tensor, one is called FoE filters guided low-rank tensor completion, and another is called FoE filters guided simultaneous tensor decomposition and completion (FoE-STDC). The resulting objectives are solved efficiently by alternating minimization. Extensive experimental results validate the superior performance and robustness of the proposed methods over their corresponding counterparts in all cases. Particularly, the proposed FoE-STDC is superior to the state-of-the-art tensor completion methods.

Image prior information is a determinative factor to tackling with the ill-posed problem. In this paper, we present multi-filters guided low-rank tensor coding (MF-LRTC) model for image restoration. The appeal of constructing a low-rank tensor is obvious in many cases for data that naturally comes from different scales and directions. The MF-LRTC takes advantages of the low-rank tensor coding to capture the sparse convolutional features generated by multi-filters representation. Using such a low-rank tensor coding would reduce the redundancy between feature vectors at neighboring locations and improve the efficiency of the overall sparse representation. In this work, we are committed to achieving this goal by convoluting the target image with Filed-of-Experts (FoE) filters to formulate multi-feature images. Then similarity-grouped cube set extracted from the multi-features images is regarded as a low-rank tensor. The resulting non-convex model is addressed by an efficient ADMM technique. The potential effectiveness of this tensor construction strategy is demonstrated in image restoration including image deblurring and compressed sensing (CS) applications.

This paper presents a novel semi-reference inspired color-to-gray conversion model for faithfully preserving the contrast details of the color image, essentially differs from most of the no-reference and reference approaches. In the proposed model, on the basic assumption that a good gray conversion should make the conveyed gradient values (i.e., contrast) to be maximal, we present a projection maximum function to model the decolorization procedure. We further incorporate weights of the original gradients into the maximum function. The Gaussian weighted factor consisting of the gradients of each channel of the input color image is employed to better reflect the degree of preserving feature discriminability and color ordering in color-to-gray conversion. The projected gradient descent and discrete searching techniques are developed to solve the proposed model with and without nonnegative constraint, respectively. Extensive experimental evaluations on two existing datasets, containing abundant colors and patterns, show that the proposed method outperforms the state-of-the-art methods quantitatively and qualitatively.

Image restoration (IR) is a long-standing challenging problem in low-level image processing. It is of utmost importance to learn good image priors for pursuing visually pleasing results. In this paper, we develop a multi-channel and multi-model-based denoising autoencoder network as image prior for solving IR problem. Specifically, the network that trained on RGB-channel images is used to construct a prior at first, and then the learned prior is incorporated into single-channel grayscale IR tasks. To achieve the goal, we employ the auxiliary variable technique to integrate the higher-dimensional network-driven prior information into the iterative restoration procedure. In addition, according to the weighted aggregation idea, a multi-model strategy is put forward to enhance the network stability that favors to avoid getting trapped in local optima. Extensive experiments on image deblurring and deblocking tasks show that the proposed algorithm is efficient, robust, and yields state-of-the-art restoration quality on grayscale images.

Magnetic Resonance (MR) images often suffer from noise pollution during image acquisition and transmission, which limits the accuracy of quantitative measurements from the data. Noise in magnitude MR images is usually governed by Rician distribution, due to the existence of uncorrelated Gaussian noise with zero-mean and equal variance in both the real and imaginary parts of the complex K-space data. Different from the existing MRI denoising methods that utilizing the spatial neighbor information around the pixels or patches, this work turns to capture the pixel-level distribution information by means of supervised network learning. A progressive network learning strategy is proposed via fitting the distribution of pixel-level and feature-level intensities. The proposed network consists of two residual blocks, one is used for fitting pixel domain without batch normalization layer and another one is applied for matching feature domain with batch normalization layer. Experimental results under synthetic, complex-valued and clinical MR brain images demonstrate great potential of the proposed network with substantially improved quantitative measures and visual inspections.

Image inpainting is a classical inverse ill-posed problem. In this paper, we introduce a multi-filters guided low-rank tensor coding as a priori information to tackle it. The key innovation is to formulate multiple feature-domain tensors by convoluting the target image with multi-filters. Furthermore, by exploring a low-rank tensor coding, it can reduce the redundancy between sparse feature vectors at neighboring locations and improve the efficiency of the overall representation. The resulting non-convex model is iteratively tackled by gradient descent and low-rank pursuit procedure. The experimental results demonstrate that the proposed algorithm can faithfully recover image and outperform the current state-of-the-art approach.