Hao Fang

Taylor expansion is a polynomial for approximating a function constructed by the coefficients of its derivatives at a certain point, where it is a challenging research to utilize the powerful learning ability of deep learning to characterize the polynomial parts for pursuing its approximate solution. In this paper, we develop a cascading residual context sharing Transformer (RCST) to approximate Taylor expansion for remote sensing (RS) image denoising. Our RCST method includes the following key procedures: Firstly, a mapping function about a latent clean RS image patch is built by employing the low-rank characteristic of its neighborhood RS image blocks, and is expanded into a polynomial with Taylor expansion for its approximate solution. Secondly, the intrinsic recursive relationship of the neighborhood derivatives is analyzed and is mathematically formulated, which provides a theoretical interpretability for the construction of our RCST model. Thirdly, a lightweight residual network is developed to estimate the base layer while the residual context sharing Transformer (RCST) is shared to calculate the derivative parts. Finally, to transfer as many rich multi-scale details from noisy RS images to estimated results as possible, we adopt a down-/up-sampling architecture. Specifically, a spatial-Fourier up-sampling (SFUS) operator is reported to preserve both local and global information. Quantitatively and qualitatively experimental results validate that our RCST denoising method can achieve competitive performance, and is even superior to other SOTA denoising approaches.

Tactical Mobile Ad hoc Networks (MANETs) often rely on multicast to carry their traffic, e.g., time-sensitive and critical mission data and situational awareness information. Implementing efficient and scalable multicast forwarding in MANETs can be challenging. Traditional protocols require per-flow state at intermediate nodes, leading to high overhead and vulnerability to nodes’ mobility. BIER is a recent multicast proposal (for static networks) that eliminates the need for explicit distribution trees or group membership information by addressing multiple destinations via a bit-string in each packet’s header. This paper identifies and addresses several challenges in designing a BIER-like multicast scheme for MANETs, including using hashing to map bit indices (from the bit-string) to IP addresses (slots in a routing table implemented as a hash-table), proposing header constructions to preserve bandwidth, and supporting forwarding using longest prefix matching.

Optical flow represents vector motion of targets, and its estimation is a hot research in computer vision. The traditional prior-based methods build mathematical models using different regularizers on object characteristics. These methods are straightforward and easily understood, but they suffer from a high computational burden. Deep learning based approaches greatly improve their efficiency, but their filter sizes are fixed and receptive fields are large. To solve these problems, this paper proposes a novel recurrent feature refinement network based on short-term dense connection modules for optical flow estimation (RFRFlow), including the following key parts: firstly, we utilize short-term dense concatenate modules (STDC) as a backbone part for constructing a feature extracting network which can preserve rich details and use global averaging pooling to extract global context information with a large receptive field. Secondly, a 4D correlation volume is constructed using context features of image pairs to extract multi-scale 4D correlation information. Thirdly, a motion feature encoder is reported to transfer the multi-scale 4D correlation information into 2D motion features. Finally, 2D context features and 2D motion features are input into a global motion aggregation (GMA) decoder for pursuing the final optical flow. In the experiments, we firstly present and analyze the results of ablation studies on the selection of batch size and image size. Then, both quantitative and qualitative results on Sintel and KITTI datasets are compared, showing that the proposed RFRFlow is effective and even performs better than the state-of-the-art optical flow estimation methods on performance improvement and objects’ shape preservation.

The additive Gaussian white noise (AGWN) level in real-life images is usually unknown, for which the empirical setting will make the denoising methods over-smooth fine structures or remove noise incompletely. The previous noise level estimation methods are easily lost in accurately estimating them from images with complicated structures. To cope with this issue, we propose a novel noise level estimation scheme based on Chi-square distribution, including the following key points: First, a degraded image is divided into many image patches through a sliding window. Then, flat patches are selected by using a patch selection strategy on the gradient maps of those image patches. Next, the initial noise level is calculated by employing Chi-square distribution on the selected flat patches. Finally, the stable noise level is optimized by an iterative strategy. Quantitative, with association, to qualitative results of experiments on synthetic real-life images validate that the proposed noise level estimation method is effective and even superior to the state-of-the-art methods. Extensive experiments on noise removal using BM3D further illustrate that the proposed noise level estimation method is more beneficial for achieving favorable denoising performance with detail preservation.

Remotely sensed images degraded by additive white Gaussian noise (AWGN) are not beneficial for the analysis of their contents. Such a phenomenon is usually modeled as an inverse problem which can be solved by model-based optimization methods or discriminative learning approaches. The former pursue their pleasing performance at the cost of a highly computational burden while the latter are impressive for their fast testing speed but are limited by their application range. To join their merits, this letter proposes a nonlocal self-similar (NSS) block-based deep image denoising scheme, namely deep low-rank prior (DLRP), which includes the following key points: First, the low-rank property of the neighboring NSS patches ordered lexicographically is utilized to model a global objective function (GOF). Second, with the aid of an alternative iteration strategy, the GOF can be easily decomposed into two independent subproblems. One is a quadratic optimization problem, and has a closed-form solution. While the other is a low-rank minimization denoising problem and is learned by deep convolutional neural network (DCNN). Then, the deep denoiser, acted as a modular part, is plugged into the model-based optimization method with adaptive noise level estimation to solve the inverse problem. In the experiments, we first discuss parameter setting and the convergence. Then, quantitative/qualitative comparisons of experimental results validate that the DLRP is a flexible and powerful denoising method to achieve competitive performance which even outperforms those produced by state-of-the-arts.

Patch-based modelling methods (i.e. sparse coding) had proven their great ability in solving the degradation problem caused by additive white Gaussian noise (AWGN) in remotely sensing images. However, these methods usually pursue surprising performance with sacrificing computational efficiency, and the learned dictionary is universal, resulting in a weak model representation flexibility. To address these issues, this paper proposes a deep sparse representation with learnable dictionary (DSRD) scheme, where the major difference from the previous sparse coding methods is that sparse representation coefficients and dictionaries are both deeply learned, acted as two modular parts to be plugged into the unfolded sparse coding model, to speed up for a stable solution as well as yielding competitive performance. The ablation studies illustrate that our DSRD strategy is efficient for a fixed solution while a flexible and powerful capacity in producing pleasing denoising performance. Meanwhile, the comparisons of quantitatively and qualitatively experimental results further demonstrate that it is effective and can generate enjoyable denoising performance that even outperforms that produced by other state-of-the-arts.