Search

Recently, unpaired medical image enhancement is one of the important topics in medical research. Although deep learning-based methods have achieved remarkable success in medical image enhancement, such methods face the challenge of low-quality training sets and the lack of a large amount of data for paired training data. In this article, a dual input mechanism image enhancement method based on Siamese structure (SSP-Net) is proposed, which takes into account the structure of target highlight (texture enhancement) and background balance (consistent background contrast) from unpaired low-quality and high-quality medical images. Furthermore, the proposed method introduces the mechanism of the generative adversarial network to achieve structure-preserving enhancement by jointly iterating adversarial learning. Experiments comprehensively illustrate the performance in unpaired image enhancement of the proposed SSP-Net compared with other state-of-the-art techniques.

Heterogeneous image fusion (HIF) is an enhancement technique for highlighting the discriminative information and textural detail from heterogeneous source images. Although various deep neural network-based HIF methods have been proposed, the most widely used single data-driven manner of the convolutional neural network always fails to give a guaranteed theoretical architecture and optimal convergence for the HIF problem. In this article, a deep model-driven neural network is designed for this HIF problem, which adaptively integrates the merits of model-based techniques for interpretability and deep learning-based methods for generalizability. Unlike the general network architecture as a black box, the proposed objective function is tailored to several domain knowledge network modules to model the compact and explainable deep model-driven HIF network termed DM-fusion. The proposed deep model-driven neural network shows the feasibility and effectiveness of three parts, the specific HIF model, an iterative parameter learning scheme, and data-driven network architecture. Furthermore, the task-driven loss function strategy is proposed to achieve feature enhancement and preservation. Numerous experiments on four fusion tasks and downstream applications illustrate the advancement of DM-fusion compared with the state-of-the-art (SOTA) methods both in fusion quality and efficiency. The source code will be available soon.

In recent years, the constraint based correlation filter has shown good performance in unmanned aerial vehicle (UAV) tracking, which gains a lot popularity in many intelligence transportation applications. In this work, a distribution-based temporal knowledge driven method is proposed to leverage the temporal translation property in UAV tracking. Instead of focusing on the traditional issues in the correlation filter, we provide a new method of learning parametric distribution on temporal knowledge by Wasserstein distance which is successfully embedded to solve the problem of temporal degeneration in learning process of tracking. Furthermore, we approximate optimal response reasoning with low-rank constraint over response consistency. Furthermore, the proposed method is solved by a simple iterative scheme with alternating direction multiplication ADMM algorithm. We demonstrate the superior tracking performance in several public standard UAV tracking benchmarks compared with state-of-the-art algorithms.

Stripe noise removal of multispectral images (MSIs) is a challenging topic and has attracted substantial research attention in remote sensing areas. Existing destriping methods mainly concentrate on patch-based model representation, which ignores the heterogeneous components across the degraded MSIs and poorly utilizes the prior knowledge of the image cubes of the MSIs on high-level perception. In this article, we present a component pixel-aware MSIs destriping method. Intuitively, the image cube of the MSIs is divided into three different components (the jump, transition, and gentle components) based on the gradients of each pixel. Then, three different regularization operators are adopted adaptively for different components in a pixel-level manner. In addition, to properly depict the stripe cube of the MSIs, we exploit the nonconvex nonsmooth iteratively reweight nuclear norm. By incorporating both improved terms, our proposed method can estimate the component-variant linear representation coefficients and model the structural information of both the stripe cube and image cube. The alternating direction method of multipliers (ADMM) is adopted to solve the proposed method. Evaluating various destriping performance measurements, the algorithm proposed in this article outperforms other state-of-the-art methods under extensive experiments on synthetic and real datasets.