Underwater image super-resolution using multi-stage information distillation networks

Recently, single image super-resolution (SISR) has been widely applied in the fields of underwater robot vision and obtained remarkable performance. However, most current methods generally suffered from the problem of a heavy burden on computational resources with large model sizes, which limited their real-world underwater robotic applications. In this paper, we introduce and tackle the super resolution (SR) problem for underwater robot vision and provide an efficient solution for near real-time applications. We present a novel lightweight multi-stage information distillation network, named MSIDN, for better balancing performance against applicability, which aggregates the local distilled features from different stages for more powerful feature representation. Moreover, a novel recursive residual feature distillation (RRFD) module is constructed to progressively extract useful features with a modest number of parameters in each stage. We also propose a channel interaction & distillation (CI&D) module that employs channel split operation on the preceding features to produce two-part features and utilizes the inter channel-wise interaction information between them to generate the distilled features, which can effectively extract the useful information of current stage without extra parameters. Besides, we present USR-2K dataset, a collection of over 1.6K samples for large-scale underwater image SR training, and a testset with an additional 400 samples for benchmark evaluation. Extensive experiments on several standard benchmark datasets show that the proposed MSIDN can provide state-of-the-art or even better performance in both quantitative and qualitative measurements.     

WavNet — Visual saliency detection using Discrete Wavelet Convolutional Neural Network

In the recent advancements in image and video analysis, the detection of salient regions in the image becomes the initial step. This plays a crucial role in deciding the performance of such algorithms. In this work, a Multi-Resolution Feature Extraction (MRFE) technique that makes use of Discrete Wavelet Convolutional Neural Network (DWCNN) for generating features is employed. An Enhanced Feature Extraction (EFE) module extracts additional features from the high level features of the DWCNN, which are used to frame both channel as well as spatial attention models for yielding contextual attention maps. A new hybrid loss function is also proposed, which is a combination of Balanced Cross Entropy (BCE) loss and Edge based Structural Similarity (ESSIM) loss that effectively identifies and segments the salient regions with clear boundaries. The method is tested exhaustively with five different benchmark datasets and is proved superior to the existing state-of-the-art methods with a minimum Mean Absolute error (MAE) of 0.03 and F-measure of 0.956.     

Residual attention-based tracking-by-detection network with attention-driven data augmentation

Tracking-by-detection (TBD) is a significant framework for visual object tracking. However, current trackers are usually updated online based on random sampling with a probability distribution. The performance of the learning-based TBD trackers is limited by the lack of discriminative features, especially when the background is full of semantic distractors. We propose an attention-driven data augmentation method, in which a residual attention mechanism is integrated into the TBD tracking network as supplementary references to identify discriminative image features. A mask generating network is used to simulate changes in target appearances to obtain positive samples, where attention information and image features are combined to identify discriminative features. In addition, we propose a method for mining hard negative samples, which searches for semantic distractors with the response of the attention module. The experiments on the OTB2015, UAV123, and LaSOT benchmarks show that this method achieves competitive performance in terms of accuracy and robustness.     

Recent advances in carbon nanotubes-based microwave absorbing composites

With the blossom of information industry, electromagnetic wave technology shows increasingly potential in many fields. Nevertheless, the trouble caused by electromagnetic waves has also drawn extensive attention. For instance, electromagnetic pollution can threaten information safety in vital fields and the normal function of delicate electronic devices. Consequently, electromagnetic pollution and interference become an urgent issue that needs to be addressed. Carbon nanotubes (CNTs) have become a potential candidate to deal with these problems due to many advantages, such as high dielectric loss, remarkable thermodynamic stability, and low density. With the appearance of climbing demands, however, the carbon nanotubes combining various composites have shown greater prospects than the single CNTs in microwave absorbing materials. In this short review, recent advances in CNTs-based microwave absorbing materials were comprehensively discussed. Typically, we introduced the electromagnetic wave absorption mechanism of CNTs-based microwave absorbing materials and generalized the development of CNTs-based microwave absorbers, including CNTs-based magnetic metal composites, CNTs-based ferrite composites, and CNTs-based polymer composites. Ultimately, the growing trend and bottleneck of CNTs-based composites for microwave absorption were analyzed to provide some available ideas to more scientific workers.     

Modeling Sialidosis with Neural Precursor Cells Derived from Patient-Derived Induced Pluripotent Stem Cells

Sialidosis, caused by a genetic deficiency of the lysosomal sialidase gene (NEU1), is a systemic disease involving various tissues and organs, including the nervous system. Understanding the neurological dysfunction and pathology associated with sialidosis remains a challenge, partially due to the lack of a human model system. In this study, we have generated two types of induced pluripotent stem cells (iPSCs) with sialidosis-specific NEU1^G227R and NEU1^V275A/R347Q mutations (sialidosis-iPSCs), and further differentiated them into neural precursor cells (iNPCs). Characterization of NEU1^G227R- and NEU1^V275A/R347Q- mutated iNPCs derived from sialidosis-iPSCs (sialidosis-iNPCs) validated that sialidosis-iNPCs faithfully recapitulate key disease-specific phenotypes, including reduced NEU1 activity and impaired lysosomal and autophagic function. In particular, these cells showed defective differentiation into oligodendrocytes and astrocytes, while their neuronal differentiation was not notably affected. Importantly, we found that the phenotypic defects of sialidosis-iNPCs, such as impaired differentiation capacity, could be effectively rescued by the induction of autophagy with rapamycin. Our results demonstrate the first use of a sialidosis-iNPC model with NEU1^G227R- and NEU1^V275A/R347Q- mutation(s) to study the neurological defects of sialidosis, particularly those related to a defective autophagy–lysosome pathway, and may help accelerate the development of new drugs and therapeutics to combat sialidosis and other LSDs.     

The Chemistry of DPPH· Free Radical and Congeners

Since the discovery in 1922 of 2,2-diphenyl-1-(2,4,6-trinitrophenyl) hydrazyl stable free radical (DPPH^·), the chemistry of such open-shell compounds has developed continuously, allowing for both theoretical and practical advances in the free radical chemistry area. This review presents the important, general and modern aspects of the chemistry of hydrazyl free radicals and the science behind it.     

Optimization of room-temperature TCR of polycrystalline La0.9-xSrxK0.1MnO3 ceramics by Sr adjustment

High-density La_0.9-xSr_xK_0.1MnO₃ ceramics (LSKMO, A-site = La, Sr and K, 0 ≤ x ≤ 0.25) are successfully fabricated by using facile sol-gel method. Electrical properties are performed by using combination of phenomenological percolation (PP) model, double exchange (DE) mechanism, and Jahn-Teller (JT) effect. Moreover, X-ray diffraction and scanning electron microscopy are employed to analyze the structure and morphology of LSKMO ceramics. Valence states and ionic stoichiometry are assessed by using X-ray photoemission spectrometry. Results reveal that Sr²⁺ ions, substituting La³⁺ ions, significantly influenced DE mechanism and JT effect. In addition, Sr-doping plays essential role in improving electrical properties of LSKMO ceramics. At optimal doping content of x = 0.09, peak temperature coefficient of resistance (TCR) of the resistivity is found to be 11.56% K^?1 at 297.15 K, which is optimal TCR for A-site K-occupied perovskite manganese oxides. These results confirm that polycrystalline LSKMO ceramics render high room-temperature TCR values due to Sr-doping.     

A novel approach for bulk micromachining of 4H-SiC by tool-based electrolytic plasma etching in HF-free aqueous solution

Bulk micromachining of single-crystal SiC has been challenging due to its extreme stability both mechanically and chemically. To address this issue, a novel tool-based electrolytic plasma etching method is proposed, with which micropatterns and micro-holes are fabricated in SiC in a hydrofluoric acid-free aqueous solution with no need for masks. The material removal is the result of the combined effects of electrolytic plasma chemistry and physics. The chemistry refers to the reaction of Si with hydroxyl radical to form various SiO_x and with H to form silanes, and the reactions of C to form volatile carbon oxides or hydrocarbons, all of which are accomplished and enhanced under the electrolytic plasma atmosphere. Besides, the local high temperature of plasma thermally promotes the evaporation or dissolution of SiO₂ in NaOH solution. The tool-based electrolytic plasma etching method provides alternative approaches for the fabrication of SiC-based MEMS and devices.     

结合句法依存信息的方面级情感分类

引入句法依存信息到原方面术语,提出一种新的方面术语表示方法,利用Glove词向量表示单词以及单词与单词之间的依存关系,构造出包含句法依存信息的依存关系邻接矩阵和依存关系表示矩阵,利用图卷积神经网络和多头注意力机制将句法依存信息融入到方面术语中,使得方面术语表达与上下文结构高度相关。将改进后的方面词术语表示替换到现有模型后,模型泛化能力得到有效提升。对比试验和分析结果表明:该方法具有有效性和泛化性。