Search

AlCrTiZrMo high-entropy alloy (HEA) films with strong amorphization were obtained by co-filter cathode vacuum arc deposition, and the effect of thermal shock on the films was investigated in order to explore the protection mechanism of HEA films against mechanical components in extreme service environments. The results show that after annealing at 800 °C for 1 h, the formation of a dense ZrTiO4 composite oxide layer on the surface actively prevents the oxidation from continuing, so that the AlCrTiZrMo HEA film exhibits excellent oxidation resistance at 800 °C in air. In the friction-corrosion coupling environment, the AlCrTiZrMo HEA film annealed at 800 °C for 1 h shows the best tribocorrosion resistance due to the stable dense microstructure and excellent mechanical properties, and its ΔOCP, COF and wear rate possess the smallest values of 0.055, 0.04 and 1.34 × 10−6 mm−3·N−1·m−1. [ABSTRACT FROM AUTHOR]

Recently, leveraging deep neural networks for automated colorectal polyp segmentation has emerged as a hot topic due to the favored advantages in evading the limitations of visual inspection, e.g., overwork and subjectivity. However, most existing methods do not pay enough attention to the uncertain areas of colonoscopy images and often provide unsatisfactory segmentation performance. In this paper, we propose a novel boundary uncertainty aware network (BUNet) for precise and robust colorectal polyp segmentation. Specifically, considering that polyps vary greatly in size and shape, we first adopt a pyramid vision transformer encoder to learn multi-scale feature representations. Then, a simple yet effective boundary exploration module (BEM) is proposed to explore boundary cues from the low-level features. To make the network focus on the ambiguous area where the prediction score is biased to neither the foreground nor the background, we further introduce a boundary uncertainty aware module (BUM) that explores error-prone regions from the high-level features with the assistance of boundary cues provided by the BEM. Through the top-down hybrid deep supervision, our BUNet implements coarse-to-fine polyp segmentation and finally localizes polyp regions precisely. Extensive experiments on five public datasets show that BUNet is superior to thirteen competing methods in terms of both effectiveness and generalization ability. • A boundary uncertainty aware network is proposed for accurate polyp segmentation. • A boundary exploration module is proposed to explore boundary cues of polyps. • A boundary uncertainty aware module is proposed to seek error-prone regions. • The proposed network achieves high performance on six public datasets. [ABSTRACT FROM AUTHOR]

Multi-view clustering divides similar objects into the same class through using the fused multiview information. Most multi-view clustering methods obtain clustering result by only analyzing structure relationship among samples, ignoring the analysis of intrinsic features of each sample, while a few methods operate the original feature on the corresponding high-dimensional kernel matrices. However, noisy and redundant features of samples are inevitably mixed in original multi-view data or high-dimensional kernel matrices. To address this problem, we propose a novel multiview clustering method, which unifies structure learning and feature learning to a framework. Specifically, we obtain a consensus structure information from multiple views via sparse subspace structure learning with weight tensor nuclear norm constraint. Then our feature learning seeks projection directions to obtain data representation by data pseudo labels, which are obtained via the fused consensus structure information. Furthermore, we use the manifold regularization term to establish the relationship between data structure information and learnt data presentation. At last, the two subtasks are alternately iterated and optimized to acquire accurate structure and discriminative data presentation. Experimental results on different datasets validate the proposed method is superior to the state-of-the-art methods. • Propose a Collaborative Structure and Feature Learning for multi-view clustering. • Propose a feature learning method with data pseudo-labels. • Propose structure learning with weighted tensor term for multiple subspace fusion. • The both learning can be iteratively boosted by using the result of the other one. [ABSTRACT FROM AUTHOR]

Traditional image editing techniques cannot be directly used to process stereoscopic media, as extra constraints are required to ensure consistent changes between the left and right images. In this paper, we propose a hybrid warping model for stereoscopic image stitching by combining projective and content-preserving warping. First, a uniform homography algorithm is proposed to prewarp the left and right images, and thus ensure consistent changes. Second, a content-preserving warping is introduced to locally refine alignment and reduce vertical disparities. Finally, a seam-cutting-based algorithm is used to find a blending seam, and the multiband blending algorithm is used to produce the final stitched image. Experimental results show that the proposed method can effectively stitch stereoscopic images, which not only avoids local distortions, but also reduces vertical disparities reasonably. [ABSTRACT FROM PUBLISHER]

A novel four-arc co-filter cathodic vacuum arc deposition technique which was optimized was introduced to deposit (AlCrMoTiV)N x high-entropy alloy nitride films that were corrosion-resistant, high hardness and adhesion. The effect of nitrogen on the microstructure, corrosion resistance and mechanical properties of films was systematically investigated. As the N 2 gas flow rate continues to increase, the phase structure of the (AlCrMoTiV)N x films transforms from a body-centred cubic structure to an amorphous structure, and then to a nanocomposite structure with nano- crystallites embedded in the amorphous matrix. The hardness of the films is proportional to the ratio of N 2 :Ar and reaches maximum (34.08 GPa) at a ratio of 7:4. The adhesion of deposited films was higher than 20 N owing to the bombardment of the substrate by energetic ion beams during the deposition of coatings. The enhanced hardness at higher N contents is mainly gives rise to the combined effect of formation of strong metal‑nitrogen bonding, solid solution strengthening, the formation of the nanocomposite structure and ion beam assisted bombardment strengthening. Furthermore, the excellent corrosion resistance can be obtained by the (AlCrMoTiV)N x films. • AlCrMoTiV HEA nitride films were prepared by co-filter cathodic vacuum arc deposition. • Effect of gas flow ratio of N 2 :Ar on structure and properties of film was studied. • The phase structure changes from BCC to amorphous and then to nanocomposite structure. • (AlCrMoTiV)N x films prepared under N 2 :Ar = 7:4 show a high-hardness of 35.7 GPa. • AlCrMoTiV HEA nitride films show better corrosion resistance than 304 stainless steel. [ABSTRACT FROM AUTHOR]

The present study reports room temperature ferromagnetism observed in Co implanted, Er implanted and (Co, Er) co-implanted ZnO films deposited by filtered cathode vacuum arc (FCVA) on sapphire substrates. The microstructures, surface morphologies, electrical and magnetic properties of pure and implanted ZnO films were investigated. X-ray diffraction results confirm that the implantation of Co and Er ions cause the substitution of Zn ions without the formation of secondary phases. The results of X-ray photoelectron spectroscopy indicate that the Er ions and Co ions are in trivalent and divalent states, respectively. After ions implantation, the electrical properties of the films are improved and the carrier concentration of Er implanted ZnO film is higher than that of Co implanted ZnO film. Substantial intrinsic ferromagnetism has been observed for all the implanted ZnO films. The saturation magnetization for the Er implanted ZnO film is higher than that of the Co implanted ZnO film and the magnetization is further enhanced by (Co, Er) co-implanting. The carrier mediated exchange mechanism can well describe the origin of ferromagnetism in the doped ZnO films. Our results can provide a reference for preparing high-quality diluted magnetic semiconductors (DMSs) by RE elements and TM elements co-doping in ZnO films. [Display omitted] • A filtered cathode vacuum arc (FCVA) deposition system was used to grow ZnO films on sapphire substrates. • The saturation magnetization for the Er implanted ZnO film is higher than that of Co implanted ZnO. • The magnetization is enhanced by Co-Er co-doping compared with Co doped and Er doped film. • The magnetization shows carrier concentration-dependent behavior. [ABSTRACT FROM AUTHOR]

The paper proposes a content- and disparity-adaptive stereoscopic image retargeting. To simultaneously avoid the saliency content and disparity distortion, firstly, we calculate the image saliency region distortion difference, and conclude the factors causing visual distortion. Then, the proposed method via a convex quadratic programming can simultaneously avoid the distortion of the salient region and adjust disparity to a target area, by considering the relationship of the scaling factor of salient region and the disparity scaling factor. The experimental results show that the proposed method is able to successfully adapt the image disparity to the target display screen, while the salient objects remain undistorted in the retargeted stereoscopic image. [ABSTRACT FROM AUTHOR]

Metal bipolar plates (BPPs) are candidates with great promise to replace conventional carbon-based BPPs. However, the drawbacks caused by the corrosion and passivation of metal bipolar plates need to be addressed urgently. In present study, corrosion-resistant and interfacial conductive high-entropy alloy and ceramic coatings possess great potential in the application of surface modification of bipolar plates in proton exchange membrane fuel cells (PEMFCs). An original co-filtered cathodic vacuum arc deposition (C-FCVAD) is designed to prepare AlTiVCrMo high-entropy alloy (HEA) and (AlTiVCrMo)N x high-entropy ceramic (HEC) coatings with different nitrogen content, and their feasibility as surface modification methods of bipolar plates is investigated by analyzing the microstructure, corrosion properties, interfacial conductivity and hydrophobicity of the coatings. The obvious body-centered cubic (BCC) structure can be observed in HEA coating, and amorphous structure is gradually formed as the nitrogen content increases to 28.12 at. %. Finally, polycrystal structure is formed when nitrogen content reached the maximum. The electrochemical corrosion and interfacial contact resistance (ICR) results significantly demonstrate that the coatings have enhanced corrosion resistance and conductivity. The bipolar plates with AlTiVCrMo HEA and (AlTiVCrMo)N x HEC coatings as surface modification methods have good application potential in PEMFCs operating environment. • The novel idea of HEA and HEC coatings adopting for bipolar plates in PEMFCs. • Development of Co-filtered cathodic vacuum arc deposition technique for HEC coating. • The HEC coatings show excellent corrosion resistance and super-low ICRs. [ABSTRACT FROM AUTHOR]

In this work, amorphous hydrogenated carbon and titanium carbide composite (a-C: H/TiC) films are deposited on titanium bipolar plates which are used in proton exchange membrane fuel cells (PEMFCs). It was fabricated by high power pulsed magnetron sputtering (HiPIMS) with TiN and TiCN transition layers. The microstructure results indicated that the titanium element was introduced in the films with a form of titanium carbide, and the increasing ID/IG values imply the graphitization of films. Amorphous carbon with a large content of sp2 hybrid bonds leads to an excellent conductivity of 1.6 mΩ·cm2 at 1.4 MPa and improves the corrosion resistance by inhibiting the growth of columnar crystal. In addition, a number of sp2-riched clusters on the surface enhance the hydrophobicity of the film. The a-C: H/TiC films prepared in this work significantly improved the hydrophobicity, corrosion resistance and interfacial conductivity of titanium bipolar plates, showing a great potential for applications in PEMFCs. [Display omitted] • Amorphous hydrogenated carbon and TiC composite films are deposited on Ti bipolar plates in PEMFCs. • The interfacial contact resistance of the modified bipolar plate is as low as 1.6 mΩ·cm2 at 1.4 MPa. • The corrosion current density of the modified bipolar plate is less than 10−6 A/cm2 in PEMFCs. [ABSTRACT FROM AUTHOR]

To overcome the moisture sensitivity of flexible organic light-emitting diode devices (OLEDs), physical vapor deposition (PVD) technology can be used to prepare films with excellent density for thin film encapsulation (TFE). The internal stress in the film should be modulated to enhance the flexibility, which can prevent structural degradation. To address these issues, a filtered cathode vacuum arc (FCVA) equipped with a superlow duty ratio high-bias pulsed was used to prepare Al 2 O 3 films with low internal stress (−57.6 MPa), excellent visible light transmittance, high packaging density, and good bending resistance. After 1000 bending cycles, the water vapor transmission rate (WVTR) for a film is increased to 5.57 × 10−3 g/m2/day at 85 °C and 85% relative humidity (R.H.). In addition, the molecular dynamics (MD) is employed to investigate the mechanism of film internal stress release, as well as the shallow implantation of ions brought about by high-bias pulse. [ABSTRACT FROM AUTHOR]

Estimating crowd density is a demanding task that has garnered significant research attention in urban planning, intelligent transportation, and other related fields. This study utilizes RGB and thermal images to leverage multimodal information and introduces a coordinated attention guidance network (CAGNet) for RGB-thermal (RGB-T) crowd counting. The framework enhances the expressive capabilities of backbone features by incorporating overall and local relationships through the information enhancement unit module, which utilizes the context coordination perception module for horizontal information mining, interactively compensating for the continuity of spatial information. Subsequently, it utilizes diverse multi-level information features for hierarchical intersection and progression, resulting in an accurate crowd counting density map. Experimental results on the RGBT-CC benchmark dataset demonstrate the robustness and effectiveness of CAGNet for RGB-T crowd counting. Furthermore, the proposed CAGNet can be extended to crowd density estimation and has achieved high performance on the ShanghaiTechRGBD and DroneRGBT datasets. The former dataset comprises paired RGB images and depth maps. The code and model for this research are available at https://github.com/WBangG/CAGNet. [ABSTRACT FROM AUTHOR]

In this paper, N ions are implanted into ZnO and ZnAlO films with different carrier concentrations prepared by radio frequency reactive magnetron sputtering on sapphire substrates. The structural, electrical, optical and magnetic properties of N-doped and (Al, N) co-doped ZnO films is investigated, and the particular emphasis is placed on the effects of carrier concentration on the defects induced magnetism in the films. Our results show that all the doped ZnO films are ferromagnetic at room temperature. A trace amount of additional Al doping has a significant effect on the improvement of ferromagnetic properties, and a maximum saturation magnetization of 73 emu/cm3 is obtained for (Al, N) co-doped ZnO films. The optical band gaps of ZnO films increase with the increasing Al doping content, which is owing to the combined effect of high carrier concentration and the Burstein Moss effect. The average transmittance of all ZnO films exceeds 80%. Our results confirm that appropriate electron doping can effectively enhance the magnetic moment in N implanted ZnO which may also apply to other ZnO systems with d0 magnetism. • N-doped ZnO and AZO films were made by ion implantation and thus contain both N ions and defects due to implantation. • The band gap and carrier concentration of ZnO films increase after Al doping. • Electron doping can effectively enhance the d0 magnetism in the N doped films. • Magnetization was correlated with lattice defects and carrier concentration. [ABSTRACT FROM AUTHOR]

Large view stereoscopic images can provide users with immersive depth experience. Image stitching techniques aim to obtain large view stitched images, and there have been various image stitching algorithms proposed recently. However, there is still no effective objective quality assessment for stereoscopic stitched images. In this paper, we propose a new perceptual objective stereoscopic stitched image quality assessment (S-SIQA) method by considering different distortion types in the existing stitching methods, including color distortion, ghost distortion, structure distortion(shape distortion, information loss), and disparity distortion. The quality evaluation methods for these distortion types are designed by using the color difference coefficient, points distance, matched line inclination degree, information loss, and disparity difference. Then we fuse these measures in the proposed S-SIQA model by an optimally weighted linear combination. In addition, to evaluate the performance of the proposed S-SIQA, we build a subjective quality assessment database for stereoscopic stitched images. Experimental results have confirmed the proposed method can effectively measure the perceptual quality of stereoscopic stitched images. [ABSTRACT FROM AUTHOR]

Convolutional neural-network-based autoencoders, which can integrate the spatial correlation between pixels well, have been broadly used for hyperspectral unmixing and obtained excellent performance. Nevertheless, these methods are hindered in their performance by the fact that they treat all spectral bands and spatial information equally in the unmixing procedure. In this article, we propose an adaptive spectral–spatial attention autoencoder network, called SSANet, to solve the mixing pixel problem of the hyperspectral image. First, we design an adaptive spectral–spatial attention module, which refines spectral–spatial features by sequentially superimposing the spectral attention module and spatial attention module. The spectral attention module is built to select useful spectral bands, and the spatial attention module is designed to filter spatial information. Second, SSANet exploits the geometric properties of endmembers in the hyperspectral image while considering abundance sparsity. We significantly improve the endmember and abundance results by introducing minimum volume and sparsity regularization terms into the loss function. We evaluate the proposed SSANet on one synthetic dataset and four real hyperspectral scenes, i.e., Samson, Jasper Ridge, Houston, and Urban. The results indicate that the proposed SSANet achieved competitive unmixing results compared with several conventional and advanced unmixing approaches with respect to the root mean square error and spectral angle distance. [ABSTRACT FROM AUTHOR]

Al2O3 and MgO composite (Al2O3/MgO) films were rapidly deposited at low temperatures using filtered cathode vacuum arc (FCVA) technology, aiming to achieve good barrier properties for flexible organic light emitting diodes (OLED) thin-film encapsulation (TFE). As the thickness of the MgO layer decreases, the degree of crystallinity decreases gradually. The 3:2 Al2O3:MgO layer alternation type has the best water vapor shielding performance, and the water vapor transmittance (WVTR) is 3.26 × 10−4 g·m−2·day−1 at 85 °C and 85% R.H, which is about 1/3 of that of a single layer of Al2O3 film. Under the action of ion deposition, too many layers will cause internal defects in the film, resulting in decreased shielding ability. The surface roughness of the composite film is very low, which is about 0.3–0.5 nm depending on its structure. In addition, the visible light transmittance of the composite film is lower than that of a single film and increases with the increase in the number of layers. [ABSTRACT FROM AUTHOR]

Lane line detection is a fundamental and critical task for geographic information perception of driverless and advanced assisted driving. However, the traditional lane line detection method relies on manual adjustment of parameters, and has poor universality, a heavy workload, and poor robustness. Most deep learning-based methods make it difficult to effectively balance accuracy and efficiency. To improve the comprehensive perception ability of lane line geographic information in a natural traffic environment, a lane line detection algorithm based on a mixed-attention mechanism residual network (ResNet) and row anchor classification is proposed. A mixed-attention mechanism is added after the backbone network convolution, normalization and activation layers, respectively, so that the model can focus more on important lane line features to improve the pertinence and efficiency of feature extraction. In addition, to achieve faster detection speed and solve the problem of no vision, the method of lane line location selection and classification based on the row direction is used to detect whether there are lane lines in each candidate point according to the row anchor, reducing the high computational complexity caused by segmentation on a pixel-by-pixel basis of traditional semantic segmentation. Based on TuSimple and CurveLane datasets, multi-scene, multi-environment, multi-linear road image datasets and video sequences are integrated and self-built, and several experiments are designed and tested to verify the effectiveness of the proposed method. The test accuracy of the mixed-attention mechanism network model reached 95.96%, and the average time efficiency is nearly 180 FPS, which can achieve a high level of accuracy and real-time detection process. Therefore, the proposed method can meet the safety perception effect of lane line geographic information in natural traffic environments, and achieve an effective balance between the accuracy and efficiency of actual road application scenarios. [ABSTRACT FROM AUTHOR]

• The intensive investigation and comparison of the degradation behavior and structural transformation of Mylar, Teonex, Vectra, Kapton and Teflon under AO exposure and Fe+-Ion irradiation in a space environment. • ReaxFF-MD simulations verify the essential interatomic interactions between AO/Fe+and the selected polymers and draw additional information. Polymers are ubiquitous in the aerospace industry, and their atomic oxygen (AO) exposure and ion irradiation stability are a chief concern. In this work, the AO and ion-irradiation degradation mechanisms for Mylar, Teonex, Vectra, Kapton and Teflon were investigated by adopting ground-based simulated environments (AO exposure and Fe+ irradiation tests) and reactive molecular dynamics (ReaxFF-MD) simulations. Mylar, Teonex, Vectra and Kapton show analogous AO degradation mechanisms, which are associated with the fracture and oxidation of the molecular chain skeleton and the emission of volatile small molecular species, resulting in mass loss. Teflon exhibits distinctive AO erosion kinetics, and the existence of an overall drift of the macromolecular chain with AO emission greatly compromises its stability. Moreover, Mylar, Teonex, Vectra and Kapton also exhibit comparable irradiation damage mechanisms and radical evolution. The elimination of C=O, C O-C and C N (only for Kapton) bonds and the generation of phenyl, phenoxyl and ketone radicals are the primary consequences of Fe+ irradiation. However, only subtle C-F bond breakage for Teflon can be detected during Fe+ irradiation, and the corresponding peroxide radicals are formed after contact with air atmosphere. [Display omitted] [ABSTRACT FROM AUTHOR]

Microstructural and magnetic properties of ion implantation method prepared Co doped Zn polar and O polar ZnO wafers are investigated by a combined experimental and theoretical approach. Substantial intrinsic ferromagnetism is observed in all the doped ZnO wafers. The ferromagnetism of Co implanted oxygen polar ZnO wafer is found to be larger than that of zinc polar ZnO. First principle calculations reveal that the magnetic moments mainly come from the Co-3d and O-2p states. The O-2p state has a stronger hybridization with the Co-3d state in oxygen polar ZnO, resulting in higher magnetic moments of Co implanted oxygen polar ZnO. These results not only benefit to a better understanding of the origin of surface-polarity dependent ferromagnetism in ZnO diluted magnetic semiconductors but also may open up the possibility to employ these polar ZnO in spintronic devices. abstract [Display omitted] [ABSTRACT FROM AUTHOR]

Surveillance video has been widely used in business, security, search, and other fields. Identifying and locating specific pedestrians in surveillance video has an important application value in criminal investigation, search and rescue, etc. However, the requirements for real-time capturing and accuracy are high for these applications. It is essential to build a complete and smooth system to combine pedestrian detection, tracking and re-identification to achieve the goal of maximizing efficiency by balancing real-time capture and accuracy. This paper combined the detector and Re-ID models into a single end-to-end network by introducing a new track branch to YOLOv5 architecture for tracking. For pedestrian detection, we employed the weighted bi-directional feature pyramid network (BiFPN) to enhance the network structure based on the YOLOv5-Lite, which is able to further improve the ability of feature extraction. For tracking, based on Deepsort, this paper enhanced the tracker, which uses the Noise Scale Adaptive (NSA) Kalman filter to track, and adds adaptive noise to strengthen the anti-interference of the tracking model. In addition, the matching strategy is further updated. For pedestrian re-identification, the network structure of Fastreid was modified, which can increase the feature extraction speed of the improved algorithm by leaps and bounds. Using the proposed unified network, the parameters of the entire model can be trained in an end-to-end method with the multi-loss function, which has been demonstrated to be quite valuable in some other recent works. Experimental results demonstrate that pedestrians detection can obtain a 97% mean Average Precision (mAP) and that it can track the pedestrians well with a 98.3% MOTA and a 99.8% MOTP on the MOT16 dataset; furthermore, high pedestrian re-identification performance can be achieved on the VERI-Wild dataset with a 77.3% mAP. The overall framework proposed in this paper has remarkable performance in terms of the precise localization and real-time detection of specific pedestrians across time, regions, and cameras. [ABSTRACT FROM AUTHOR]

At present, with the advance of satellite image processing technology, remote sensing images are becoming more widely used in real scenes. However, due to the limitations of current remote sensing imaging technology and the influence of the external environment, the resolution of remote sensing images often struggles to meet application requirements. In order to obtain high-resolution remote sensing images, image super-resolution methods are gradually being applied to the recovery and reconstruction of remote sensing images. The use of image super-resolution methods can overcome the current limitations of remote sensing image acquisition systems and acquisition environments, solving the problems of poor-quality remote sensing images, blurred regions of interest, and the requirement for high-efficiency image reconstruction, a research topic that is of significant relevance to image processing. In recent years, there has been tremendous progress made in image super-resolution methods, driven by the continuous development of deep learning algorithms. In this paper, we provide a comprehensive overview and analysis of deep-learning-based image super-resolution methods. Specifically, we first introduce the research background and details of image super-resolution techniques. Second, we present some important works on remote sensing image super-resolution, such as training and testing datasets, image quality and model performance evaluation methods, model design principles, related applications, etc. Finally, we point out some existing problems and future directions in the field of remote sensing image super-resolution. [ABSTRACT FROM AUTHOR]

Traffic signs detection and recognition is an essential and challenging task for driverless cars. However, the detection of traffic signs in most scenarios belongs to small target detection, and most existing object detection methods show poor performance in these cases, which increases the difficulty of detection. To further improve the accuracy of small object detection for traffic signs, this paper proposed an optimization strategy based on the YOLOv4 network. Firstly, an improved triplet attention mechanism was added to the backbone network. It was combined with optimized weights to make the network focus more on the acquisition of channel and spatial features. Secondly, a bidirectional feature pyramid network (BiFPN) was used in the neck network to enhance feature fusion, which can effectively improve the feature perception field of small objects. The improved model and some state-of-the-art (SOTA) methods were compared on the joint dataset TT100K-COCO. Experimental results show that the enhanced network can achieve 60.4% mAP(Mean Average Precision), surpassing the YOLOv4 by 8% with the same input size. With a larger input size, it can achieve a best performance capability of 66.4% mAP. This work provides a reference for research on obtaining higher accuracy for traffic sign detection in autonomous driving. [ABSTRACT FROM AUTHOR]

Semantic segmentation of high-resolution remote sensing images plays an important role in the remote sensing community. However, many indistinguishable objects are prevalent within urban remote sensing images, and some objects belonging to the same class are different and many objects that do not belong to the same class are similar. These tricky objects make the images exhibit low-interclass variance and high-intraclass variance, which significantly limits segmentation performance. Therefore, a fresh insight was presented to alleviate this issue by incorporating the fuzzy pattern recognition method and deep-learning method. Specifically, we proposed a multiscale fuzzy dual-domain attention network (MFDAN). In MFDAN, a two-dimensional Gaussian fuzzy learning module is proposed to eliminate those factors that influence the intraclass and interclass variance. In addition, a dual-domain attention module is proposed to derive more informative semantic representations in the channel and spatial domains, respectively. These two modules will be integrated in a multiscale perspective. Extensive experiments on the benchmark datasets illustrate qualitatively and quantitatively that the proposed MFDAN is competitive. [ABSTRACT FROM AUTHOR]

Most existing image dehazing methods rely on the solution of the atmospheric scattering model or supervised learning based on paired images. However, owing to incomplete prior knowledge and the lack of paired hazy and haze‐free images of the same scenes as training samples, their performances for single image dehazing are unsatisfactory. Here, the authors present an unpaired image learning method based on the attention mechanism for single image dehazing problems. The method uses the constraint transfer learning ability and circulatory structure of CycleGAN to carry out an unsupervised image dehazing task for unpaired data. Considering the complexity of the haze distribution in actual imaging and human visual characteristics, the improved channel attention and domain attention mechanisms are integrated into the network to process different features and different regions non‐uniformly. The experimental results show that the proposed method achieves good results on both synthetic datasets and real hazy images. [ABSTRACT FROM AUTHOR]

Organic light-emitting diode devices (OLEDs) industry is developing rapidly, however, OLEDs are very sensitive to oxygen and water, determining how to achieve better barrier effect is one of the main challenges of flexible OLEDs technology. Based on the filtered cathode vacuum arc (FCVA) technology, a dense aluminum oxide (Al 2 O 3) film with a thickness of about 100 nm was grown at room temperature for OLEDs encapsulation. The composition, structure, surface roughness, refractive index, light transmittance and other physical properties of the film were analyzed. The water vapor transmission rates (WVTRs) of the films were tested at 85 °C and 85% relative humidity (R.H.). Photoluminescence spectrum (PL) test was carried out to study the luminescence ability of tris(8-hydroxyquinoline) aluminum (Alq 3) before and after the deposition of Al 2 O 3 film. The results show that the amorphous Al 2 O 3 film with dense structure prepared by FCVA has high visible light transmittance. The Al 2 O 3 film reduces the WVTRs of the polyethylene naphthalate (PEN) film by four orders of magnitude and maintains the light-emitting performance of the organic light-emitting layer. These results indicate that the FCVA technology is a feasible method for preparing barrier films of flexible OLEDs. • The FCVA is applied to the field of OLEDs encapsulation. • Rapidly deposited Al 2 O 3 (15 nm/min) by FCVA do not reduce PL intensity of Alq 3. • Al 2 O 3 film with 100 nm thickness reaches low WVTRs value of 9.897 × 10−4 g/m2/day. [ABSTRACT FROM AUTHOR]

Recent years have seen a surge in interest in graph neural networks (GNNs) due to their superior performance in a range of graph and network mining applications. Graph embedding attempts to convert nodes in graph data to a low-dimensional vector representation by capturing the edges between them. However, because the bulk of GNNs currently use unstable graph structures, they perform well on graphs with a high degree of homogeneity and badly on those with a low degree of homogeneity. As a result, GNNs that rely solely on the original graph structure may give unsatisfactory results. In this research, we introduce an adaptive multi-channel Bayesian graph neural network (AMBGN 1 1 Code available at https://github.com/DongYang26/AMBGN.) for estimating new graph structures and adaptively fusing some depth-related information between the original topological structures. The key idea is to follow the GNN mechanism by estimating the ideal graph structure using Bayesian inference and extracting specific and common embeddings from estimated graph structures, topological structures, and their combinations. We are able to maximize both estimated graph structure learning and node embedding in an iterative framework by using the attention approach to learn the significant weights of these three node embeddings. Our extensive research on a number of benchmark datasets with varied degrees of homogeneity demonstrated that AMBGN reliably estimates the graph structure and effectively learns the most relevant node information in both graph representations, confirming the usefulness of AMBGN. • Explore graph structure learning, advocate GNN's concurrent learning from original and estimated graphs. • Adaptive integration enhances GNN's classification using useful information from both graphs. • Introduce Bayesian inference for learning posterior graph structures. • Propose attention mechanisms to quantify the importance of graphs in classification tasks. [ABSTRACT FROM AUTHOR]

Semantic segmentation of remote sensing images is a fundamental task in computer vision, with significant applications in forest and farmland cover surveys, geological disaster monitoring, and other related fields. The inclusion of digital surface models can enhance the segmentation performance compared to using unimodal imaging alone. However, most existing methods simply combine the features from both modalities without considering their differences and complementarity, leading to a loss of spatial details. In order to address this issue and improve segmentation accuracy, we propose a novel network called Progressive Adjacent-Layer Coordination Symmetric Cascade Network (PACSCNet). This network employs a two-stage fusion symmetric cascade encoder to leverage the similarities and differences between adjacent features for cross-layer fusion, thereby preserving spatial details. Additionally, our network includes a new dual-pyramid symmetric cascade decoder that extracts similarities in multimodal and cross-layer fusion features for merging. Furthermore, a pyramid residual integration module progressively integrates features at four scales to mitigate noise interference during large-scale fusion. Extensive experimental evaluations on the Vaihingen and Potsdam datasets demonstrate that PACSCNet achieves strong semantic segmentation performance, comparable to state-of-the-art approaches, in terms of accuracy and intersection-over-union. The source code and results of our proposed PACSCNet are publicly available at https://github.com/F8AoMn/PACSCNet. [ABSTRACT FROM AUTHOR]

Landcover classifications have large uncertainty related to the heterogeneity of similar objects and complex spatial correlations in satellite images, making it difficult to obtain ideal classification results using traditional classification methods. Therefore, to address the uncertainty in landcover classifications based on remotely sensed information, we propose a novel fuzzy c-means algorithm, which integrates adaptive interval-valued modelling and spatial information. It dynamically adjusts the interval width according to the fuzzy degree of the target membership without pre-setting any parameters, controls the fuzziness of the target, and mines the inherent distribution of the data. Furthermore, reliability-based spatial correlation modelling is used to describe the spatial relationship of the target and to improve both robustness and accuracy of the algorithm. Experimental data consisting of SPOT5 (10-m spatial resolution) or Thematic Mapper (30-m spatial resolution) satellite data for three case study areas in China are used to test this algorithm. Compared with other state-of-the-art fuzzy classification methods, our algorithm markedly improved the ground-object separability. Moreover, it balanced improvement of pixel separability and suppression of heterogeneity of intra-class objects, producing more compact landcover areas and clearer boundaries between classes. [ABSTRACT FROM AUTHOR]

Multi-view clustering aims to group data points into their classes. Exploiting the complementary information underlying multiple views to benefit the clustering performance is one of the topics of multi-view clustering. Most of existing multi-view clustering methods only constrain diversity and consistency in the data space, but not consider the diversity and consistency in the learned label space. However, It is natural to take the impacts of diversity in the learned label matrix into consideration, because different view would generate different clustering label matrix, in which some labels are consistent and some are diverse. To overcome this issue, we propose a novel multi-view clustering method (DCMSC) by constraining diversity and consistency in both the learned clustering label matrix and data space. Specifically, in the learned label space, we relax the learned common label matrix into consistent part and diverse part. Meanwhile, by applying an introduced row-aware diversity representation and l 2,1 -norm to constrain diverse part, wrong-labels and the influences of noises on the consistent part are reduced. In the data space, we weight each view by using a self-weight strategy. Furthermore, we conduct clustering in spectral embedded spaces instead of original data spaces, which suppresses the effect of noises and decreases redundant information. An augmented Lagrangian multiplier with alternating direction minimization (ALM-ADM) based optimization solution can guarantee the convergence of our method. Extensive experimental results on both synthetic datasets and real-world datasets demonstrate the effectiveness of our method. [ABSTRACT FROM AUTHOR]

To alleviate the atomic oxygen (AO) degradation of the polyimide deployed in the low Earth orbit (LEO) environment, we fabricate a set of amorphous TiAlCrZrSi/polyimide films under various negative bias values via the 3rd generation co-filter cathode vacuum arc (C-FCVA). Our rational technical design realizes the noninterference regulation and filtration of mixed multielement plasma compositions. In addition, the atomic size difference (δ , 0.075–0.081), mixing enthalpy (ΔH mix , −48.98 to −46.84 kJ mol−1), and mixing entropy (ΔS mix , 12.83–12.91 J K−1 mol−1) are all within the range of parameters required for the formation of amorphous TiAlCrZrSi high-entropy alloys (HEAs). Moreover, the TiAlCrZrSi/polyimide (no bias) film can realize an equilibrium of AO resistance and durability (E y = 2.02 × 10−26 cm3 atom−1), which is only 0.67 % of that calculated for pristine polyimide. This can be attributed to the excellent AO penetration barrier ability of the generated HEA oxide scales. Furthermore, the thickening of HEA films can reduce the sensitivity of residual defects to AO undercutting erosion, as discussed by the existing empirical theoretical models. [Display omitted] • Fabricated amorphous TiAlCrZrSi/polyimide films by noninterference regulation and filtration of mixed multielement plasmas. • Atomic oxygen erosion yield of the TiAlCrZrSi/polyimide film is 2.02 × 10-26 cm3 atom-1. • The generation of high-entropy alloy oxide scales results in enhanced AO penetration barrier ability. • The thickening of high-entropy alloy films can reduce the sensitivity of residual defects to AO undercutting erosion. [ABSTRACT FROM AUTHOR]

• A filtered cathode vacuum arc (FCVA) system was used to grow MgO films on sapphire substrates. • The magnetic moment for the Er implanted MgO film is higher than that of Co implanted MgO. • The magnetization is enhanced by Co-Er co-doping compared with Co doped and Er doped MgO. • The origin of the magnetism can be described by the bound magnetic polaron (BMP) model. The present study reports a high ferromagnetism acquired by co-implanting Co and Er ions in MgO films prepared by filtered cathode vacuum arc on Al 2 O 3 substrates. The resulting MgO films were analyzed with respect to their structures, surface morphologies, and magnetic properties. Magnetic measurements show that substantial intrinsic ferromagnetism has been found for all the doped MgO. The ferromagnetism for the Er-doped MgO is larger than that of the Co-doped MgO and it is further strengthened by Er-Co co-doping. The X-ray photoelectron spectroscopy analysis substantiates that a large number of oxygen vacancies (V o) are generated after the doping of Er and Co ions. The origin of the magnetism can be described by the V o mediated bound magnetic polaron (BMP) model. The increased number of BMP with higher moments and the coupling interaction of Er and Co accounts for the strong ferromagnetism in the co-doped films. Our research provides a new idea for manufacturing high-quality diluted magnetic oxides by co-doping rare earth and transition metal ions in MgO samples. [ABSTRACT FROM AUTHOR]

Super-hard and corrosion-resistant (AlCrMoSiTi)N x high-entropy nitride coatings were prepared using a novel multi-arc cathodic vacuum magnetic filtration deposition at different N 2 gas flow rates (R N) of 0, 20, 50 and 70 sccm. The influence of R N on the element composition, phase structure, mechanical, tribological and corrosion properties of the coatings were systematically studied. Higher R N promoted higher N contents in the coatings that resulted in a transformation from an amorphous to face-centered-cubic (FCC) phase. At a high N content of 49 at. % (R N = 70%), the coating exhibits a simple FCC structure, and offers a superior combination of super-hardness, excellent wear and corrosion resistance. The exceptional hardness of the (AlCrMoSiTi)N coating (41.6 GPa) that surpasses most high-entropy ceramics is underlain by the formation of strong metal-nitride phase, the solid-solution strengthening and the fine grain strengthening. Therefore, (AlCrMoSiTi)N x nitride coating presents a potential application in protective coating owing to its extraordinary mechanical properties and corrosion resistance. • AlCrMoSiTi nitride coating prepared by multi-arc cathodic vacuum magnetic filtration. • Effect of N 2 flow rate on structure and properties of coating was studied. • AlCrMoSiTi nitride coatings show a phase structure change from amorphous to FCC. • The AlCrMoSiTi nitride coatings show a super-hard of 41.6 GPa and low wear rate. • Super-hard nitride coating has a low corrosion current density of 2.2 × 10−8 A cm−2. [ABSTRACT FROM AUTHOR]

• Er-doped ZnO films were made by ion implantation, thus contain both Er ions and defects. • The magnetization of ZnO can be enhanced significantly by annealing in Ar and vacuum. • The enhancement of ferromagnetism is strongly correlated with the increase of oxygen vacancies in Er-doped ZnO. • Proper annealing is a good way to enhance the magnetization for magnetic ions doped ZnO. The Er ions have been introduced into ZnO films prepared by radio-frequency magnetic sputtering (RFMS) on sapphire substrates by ion implantation. The annealing effect on structural and magnetic properties for Er-implanted ZnO films under air, Ar, and vacuum atmospheres at 650 °C have been investigated systematically by a combined characterization of X-ray diffractions, Raman spectroscopy, Photoluminescence, X-ray photoelectron spectroscopy and Superconducting Quantum Interference Device. Strong room-temperature ferromagnetism has been observed for the as-implanted and annealed films. We found that the saturation magnetization varied greatly for different annealing conditions. A considerable enhancement in magnetization is observed for Er-doped ZnO annealed in vacuum, which is accompanied by an enhancement in the concentration of oxygen vacancy. Our results give clear evidence that the enhancement of ferromagnetism is strongly correlated with the increase of oxygen vacancies in Er-doped ZnO. [ABSTRACT FROM AUTHOR]

Road traffic safety is a very important issue in the field of intelligent transportation system (ITS). Vehicle segmentation and behavior analysis are an significant part for solving this problem. However, 2D image detection technology is difficult to reconstruct missing information of damaged image. In this paper, a bottom-up analysis method is employed to study the related technical problems, and it will provide a strong data foundation for road traffic safety. Firstly, the M-BRISK descriptor algorithm is proposed to describe the local feature points. Secondly, we propose a 3D feature analysis method based on rigid motion constraints for vehicle trajectory. Thirdly, a similarity measure method is proposed for trajectory clustering. Finally, we used the obtained 3D information of vehicles to analyze the vehicle behavior to find the abnormal vehicles for road traffic safety. The experimental results confirm that the M-BRISK descriptor performs well comparing with the state-of-the-art feature descriptors, and the proposed clustering method improves the accuracy of the trajectory clustering. Moreover, the vehicle motion information contained in the trajectory data can be analyzed to recognize vehicle behavior. The presented work in this paper provides an important foundation for vehicle abnormal behavior detection for road traffic safety. • Construct a hybrid binary feature descriptor based on BRISK. • Propose a 3D feature analysis method of vehicle trajectory based on rigid motion constraints. • Construct a new similarity measure between trajectories. • Behavior model and semantic analysis of vehicles in traffic scene are built. [ABSTRACT FROM AUTHOR]