Search

Although the guided image filter (GIF) is an excellent edge-preserving filter, it generally suffers from halo artifacts due to the local property and the fixed regularization parameter. To address the problem, a weighted guided image filter (WGIF) was proposed by incorporating an edge-aware weighting into the GIF. In the filtering process, WGIF employs an averaging strategy for edge-aware weighting. Although the averaging strategy is a highly efficient method, it is susceptible to extreme values and tends to obscure critical factors, so it often leads to inaccurate results. Consequently, the output results quality of the WGIF is often degraded. To remedy the deficiency, a weighted guided image filter with entropy evaluation weighting (EEW-WGIF) is proposed in this paper. EEW-WGIF employs an edge-aware weighting strategy based on entropy evaluation method to detect edges more accurately, and incorporates an explicit constraint based on the gradient variation to better preserve edges. To verify the filtering effectiveness of the EEW-WGIF, it was applied to edge-preserving smoothing filtering, exposure images fusion, single image detail enhancement, structure-transferring filtering and image denoising. Experimental results show that the proposed filter can achieve excellent performance in both visual quality and objective evaluation. [Display omitted] • An edge-aware weighting strategy based on entropy evaluation method is proposed, which is more reliable in calculating the importance of each edge-aware factor. • The proposed EEW-WGIF incorporates an explicit constraint based on gradient variation and hyperbolic function to handle the edges so that the edges can be better preserved. • The EEW-WGIF was applied to edge-preserving smoothing filtering, exposure images fusion, single image detail enhancement, structure-transferring filtering and image denoising. [ABSTRACT FROM AUTHOR]

A methodology to control the flexural wave propagation via modulating the beam profile is presented in this paper. Firstly, the power series method is utilized to theoretically solve the flexural waves in a non-uniform Timoshenko beam with arbitrarily contoured profiles. It is demonstrated that this method is effective for most of inhomogeneous beams as long as the convergence criterion about the beam thickness is satisfied in advance. After validation by comparing the theoretical results with those from the finite element analysis for different thickness variations, lenses are designed by using the quadratic thickness variation pattern with the aim of focusing the wave emitted from a point source on one or two particular positions. Additionally, according to the generalized Snell's law, the thickness-induced phase modulation of the flexural wave is proved from the views of theoretical analysis and numerical simulations, based on which lenses for focusing a plane flexural wave are also realized. It is illustrated through systematic analysis in the frequency domain that all of the lenses designed can exhibit good performances with the evidently increased energy at actual focal positions and the focusing sizes smaller than one working wavelength, which has great potentials in versatile engineering applications. • Flexural wave in a beam with arbitrary profiles is solved by the power series method. • A flexural-wave-based lens for energy focusing is designed via beam thickness modulation. • Energy focusing is realized with the aid of phase control. [ABSTRACT FROM AUTHOR]

We analyzed a three-dimensional thermal-electrical coupled model based on COMSOL to characterize the thermal damage response of a woven carbon fiber reinforced polymer (CFRP) composite with and without a lightning strike protection (LSP) system. Fuzzy fiber (FF) fabric serving as a carbon-based protection layer was attached to the outermost ply of the CFRP composite to fabricate a fuzzy fiber reinforced polymer (FFRP) composite. CFRP and FFRP composites with temperature-dependent properties were inspected to predict lightning-induced damage resulting from a 20- and 40-kA peak current for 100 μs. The predicted area of thermal damage and the appearance of the composite surface agreed fairly well with post-lightning damage observed from experiments, thus demonstrating the credibility of the numerical model. LSP characteristics were evaluated for a range of CFRP composite properties in the outermost layer by enhancing the functional conductivity of the top layer in the in-plane and out-of-plane directions. The irreversible thermal damage region in the in-plane and thickness directions was dramatically mitigated by enhanced electrical conductivity, whereas a slight matrix decomposition damage change was observed by varying the thermal conductivity. Due to the increased electrical conductivity and integration of the lightweight FF carbon-based protection layer into the uppermost composite layer, the depth and area of damage can be limited by decreasing the thermal damage penetration through the underlying composites. These results reveal that a highly conductive FF layer may serve as a lightweight and effective anti-lightning strike layer for protecting the underlying composite. [ABSTRACT FROM AUTHOR]

In order to make the weighted guided image filter (WGIF) more fully utilize the edge direction of the image, a steering kernel weighted guided image filter (SKWGIF) is proposed by employing the steering kernel to learn the edge direction and incorporating the learning results into the WGIF. However, SKWGIF does not provide a good compromise between the two possibly contradictory objectives of edge-preserving and smoothing, where the image edges are inevitably smoothed. To overcome the drawback, a SKWGIF with gradient constraint (GC-SKWGIF) is proposed by introducing the gradient constraints into the SKWGIF. The gradient constraints allow the filter to take into account the gradient variation of the edges in the filtering process, and therefore the image edges can be better preserved. To verify the effectiveness of the proposed filter, the GC-SKWGIF is applied to edge-aware smoothing, tone mapping of high range dynamic images, image denoising and haze removal. Both theoretical analysis and experimental results show that the proposed filter can produce good resultant images. [Display omitted] • A new edge-aware weighting is defined by using local variance and local gradient magnitude of 3 × 3 windows of all pixels. • A steering kernel weighted guided image filter with gradient constraint (GC-SKWGIF) is proposed by considering the gradient variation at the edges and introducing the gradient constraint. • Experiments such as edge-aware smoothing, tone mapping of high dynamic images, image denoising and haze removal show that GC-SKWGIF has very outstanding performance. [ABSTRACT FROM AUTHOR]

The combination of deep learning and the medical field has recently achieved great success, particularly in recommending medicine for patients. However, patients' clinical records often contain repeated medical information that can significantly impact their health condition. Most existing methods for modeling longitudinal patient information overlook the impact of individual diagnoses and procedures on the patient's health, resulting in insufficient patient representation and limited accuracy of medicine recommendations. Therefore, we propose a medicine recommendation model called KEAN, which is based on an attention aggregation network and enhanced graph convolution. Specifically, KEAN can aggregate individual diagnoses and procedures in patient visits to capture significant features that affect patients' diseases. We further incorporate medicine knowledge from complex medicine combinations, reduce drug–drug interactions (DDIs), and recommend medicines that are beneficial to patients' health. The experimental results on the MIMIC-III dataset demonstrate that our model outperforms existing advanced methods, which highlights the effectiveness of the proposed method. [Display omitted] • An attention aggregation model KEAN is proposed for medicine recommendation. • The enhanced graph convolution module is employed to capture domain knowledge from medicine combinations. • The proposed KEAN achieves the state-of-the-art performance on MIMIC-III datasets. [ABSTRACT FROM AUTHOR]

Recent research in the analysis of human images, such as human parsing and pose estimation, usually requires input images to have a sufficiently high resolution. However, small images of people are commonly encountered in our daily lives, particularly in surveillance applications. This paper aims to ultra-resolve a tiny person image to its high-resolution counterpart by learning effective feature representations and exploiting useful human body prior knowledge. First, we propose the Residual Clique Block (RCB) to fully exploit compact feature representations for image Super-Resolution (SR). Second, a series of RCBs are cascaded in a coarse-to-fine manner to construct the Pyramid Residual Clique Network (PRCN), which simultaneously reconstructs multiple SR results (e.g. 2 × , 4 × , and 8 ×) in one feed-forward pass. Third, we utilize the human parsing map as the shape prior, and the high-frequency sub-bands of Uniform Discrete Curvelet Transform (UDCT) as the texture prior to enhance the details of reconstructed human body image. Experimental results demonstrate that our proposed method achieves state-of-the-art performance with superior visual quality and PSNR/SSIM scores. Moreover, we show that our results can considerably enhance the performance of human parsing and pose estimation tasks. • We propose a new feature extraction block to fully utilize the hierarchical features from LR images. • We develop a framework to progressively reconstruct HR images and estimate human body prior in a coarse-to-fine manner. • Two kinds of human body priors are estimated in our framework and then used to improve SR performance. • Our method outperforms other methods in image super-resolution, pose estimation and human parsing. [ABSTRACT FROM AUTHOR]

Despite significant progress made in image semantic segmentation, most research has primarily focused on daytime scenes. Semantic segmentation of nighttime images is equally critical for autonomous driving; however, this task poses more challenges due to inadequate lighting and difficulties associated with obtaining accurate manual annotations. In this paper, we propose a novel method called Latent Domain Knowledge Distillation (LDKD) for accurate nighttime semantic segmentation. Within a teacher-student framework, our LDKD combines domain alignment and knowledge distillation in a mutually reinforcing manner. First, we introduce a bidirectional photometric alignment module capable of generating daytime-like and nighttime-like latent images in real-time. This module effectively bridges the appearance gap between the source domain (daytime) and the target domain (nighttime). Second, we enhance collaboration between the teacher and student networks. The student network unifies image-level and feature-level domain alignment, while the teacher network generates reliable pseudo-labels by distilling knowledge from the latent domain. Furthermore, to account for potential noise in pseudo-labels, we propose a noise-tolerant learning method aimed at mitigating the risks associated with overreliance on pseudo-labels during model training. Extensive experiments on Dark Zurich and ACDC datasets show that our LDKD achieves state-of-the-art performance, demonstrating the effectiveness of our method for nighttime semantic segmentation. [ABSTRACT FROM AUTHOR]

[Display omitted] • The addition of CaO and SO 3 can increase the strength of geopolymer cured at ordinary temperature; • The CaO and SO 3 increase the water immersion expansion rate of geopolymer; • Ettringite and calcium silicate hydrate gel are formed in fly ash based geopolymer; • The slight expansion of ettringite effectively reduces the drying shrinkage of geopolymer. Fly ash based geopolymer is a green cementitious material with the potential to replace portland cement. Due to the changes of raw coal and coal combustion technology in recent years, the fly ash often has high CaO and SO 3 contents. The action mechanism of CaO and SO 3 on the mechanical properties and dimensional stability of fly ash-based geopolymers is not clear, which restricts its application. The influence of CaO and SO 3 content on the flexural strength, compressive strength, water immersion expansion rate and drying shrinkage rate of fly ash based geopolymer were studied, and X-ray diffractography (XRD), scanning electron microscope (SEM), energy dispersive spectrometer (EDS) and 29Si nuclear magnetic resonance (NMR) were used to study the influence mechanism of calcium and sulfur on the geopolymer. The results show that the mechanical properties of fly ash based geopolymers increase initially and then decrease with the increase of CaO and SO 3 content. When the contents of CaO and SO 3 are 11% and 4% respectively, the properties of geopolymer prepared at ordinary temperature (20 ℃) are the optimum. The 28d flexural strength and compressive strength can be increased by 17.91% and 15.56% respectively, and compared with the unmodified fly ash based geopolymer, the drying shrinkage is reduced by 60.65%. Microscopic test results show that the addition of CaO and SO 3 promoted the geopolymer reaction, increased the gel products and generated ettringite, which enhanced the strength of the matrix and compensated the shrinkage of geopolymer. [ABSTRACT FROM AUTHOR]

During the multi-attribute group decision-making (MAGDM) processing, the individuals often hold different opinions about the alternatives. It is necessary to aggregate the different individual opinions into a unified group opinion. In the real world, experts sometimes use linguistic expressions to evaluate attributes in uncertain environments. To address the problem of reducing the information loss of expert opinion aggregation in MAGDM, this paper proposes a MAGDM approach based on linguistic concept lattices in the context of uncertain linguistic expression. A linguistic concept lattice for multi-expert linguistic formal context is first constructed based on linguistic truth-valued lattice implication algebra, which can express both comparable and incomparable linguistic information in the decision-making process. Different expert opinions are aggregated via the extent of fuzzy linguistic concepts, which can reduce information loss in the aggregation process. Second, meet-irreducible elements in the linguistic concept lattice are introduced to reduce the computational complexity of obtaining all fuzzy linguistic concepts in the decision-making process. the distance between the intents of different fuzzy linguistic concepts is considered to enhance the rationality of linguistic decision results. In addition, the expert's decision-making process for each alternative is visualized via linguistic concept lattices. Finally, the case study and comparative analysis illustrate the validity and rationality of the proposed approach in MAGDM with linguistic information. • The fuzzy linguistic concepts are used to aggregate expert opinions. • The comparable and incomparable linguistic information is handled in MAGDM. • Visualize the decision-making process of experts by constructing concept lattices. • A novel linguistic concept lattice-based MAGDM approach is proposed. [ABSTRACT FROM AUTHOR]

Large-scale model sizes and expensive computing costs cause the challenge of deploying and applying large pre-trained models. Hence, this paper presents a novel Triple Concepts Attention Mechanism and a lightweight TCAMixer model for edge devices to classify texts. Furthermore, the TCAMixer abstracts textual concepts in a human way, which is unmatched by other counterparts such as pNLP-Mixer (a projection-based MLP-Mixer model for Nature Language Processing) and HyperMixer (a hyper network using dynamic token-mixing layers). Experimental results on several public datasets demonstrate that the TCAMixer outperforms the counterparts by a significant margin, for example, achieving 3% higher accuracy with a smaller model size of 0.177M. Additionally, the TCAMixer achieves a performance of 85% to 98.7% compared to that of large pre-trained models but only occupies 1/3000 to 1/2000 of their size on most test datasets. [ABSTRACT FROM AUTHOR]

Visual tracking is a fundamental research topic in computer vision community, which is of great importance in many application areas including augmented reality, traffic control, medical imaging and video editing. This paper presents an ordered over-relaxation Langevin Monte Carlo sampling (ORLMC) based tracking method within the Bayesian filtering framework, in which the traditional object state variable is augmented with an auxiliary momentum variable. At the proposal step, the proposal distribution is designed by simulation of the Hamiltonian dynamics. We first use the ordered over-relaxation method to draw the momentum variable which could suppress the random walk behavior in Gibbs sampling stage. Then, we leverage the gradient of the energy function of the posterior distribution to draw new samples with high acceptance ratio. The proposed tracking method could ensure that the tracker will not be trapped in local optimum of the state space. Experimental results show that the proposed tracking method successfully tracks the objects in different video sequences and outperforms several conventional methods. [ABSTRACT FROM AUTHOR]

Predicting crowd flow is of great importance to public safety and traffic management. The crowd flow is difficult to predict accurately and timely due to the uncertainty of the future positions. In this paper, we propose a novel Active Offset Network (AONet), in which ActiveGRU (Active Gate Recurrent Unit) is designed to predict the variation of pedestrians' positions in the crowd flow. Its inner location-variant recurrent structure is implemented by utilizing convolution operation on low dimensional spatio-temporal sequences to obtain fractional offset locations. Afterwards, the sampling locations are determined by bilinear interpolation on fractional offset locations. Moreover, a probabilistic sparse strategy is introduced to reduce the links between sampling locations during supervised training. Finally, the experiments over popular benchmarks demonstrate that our method can actively characterize the future positions of pedestrians. Meanwhile, the performance of the proposed AONet is superior over state-of-art baselines with regard to both accuracy and computational savings. [ABSTRACT FROM AUTHOR]

• Alkali-activated slag modified by chitosan was used to prepare pervious concrete. • The volumetric structural parameters of pervious concrete were optimized. • The pervious concrete with water purification was designed. A kind of alkali-activated slag pervious concrete (AAS-PC) was designed to purify rainwater to prevent it from polluting groundwater resources. The pervious concrete was designed from two aspects: the cementitious material and the volumetric structure parameters of pervious concrete. Chitosan was used in the cementitious material paste to enhance the adsorption capacity for heavy metal ions and other pollutants. In consideration of adsorption, composition parameters of the cementitious material paste such as the chitosan content, the mole ratio of SiO 2 /Na 2 O, and the concentration of alkali activator wereoptimized by orthogonal experiment. The influences of the bulk porosity of aggregate (BPA) and the ratio of paste to aggregate (P/A, mass ratio) on water purification capability, water permeability, and mechanical properties of the AAS-PC were investigated. It is found that, with the BPA increasing, the adsorption to Pb2+ and the compressive strength of pervious concrete decreased gradually but its water permeability improved; meanwhile, with the P/A increasing, its adsorption to Pb2+ and the compressive strength increased, but its water permeability decreased. There is a negative correlation between the water purification and the water permeability of the pervious concrete. Considering the water purification, water permeability and strength properties of the pervious concrete, the optimum BPA was 30% and the optimum P/A was 0.25. [ABSTRACT FROM AUTHOR]