Search

Video saliency prediction (VSP) aims to imitate eye fixations of humans. However, the potential of this task has not been fully exploited since existing VSP methods only focus on modeling visual saliency of the input previous frames. In this paper, we present the first attempt to extend this task to video saliency forecasting (VSF) by forecasting attention regions of consecutive future frames. To tackle this problem, we propose a video saliency forecasting transformer (VSFT) network built on a new encoder-decoder architecture. Different from existing VSP methods, our VSFT is the first pure-transformer based architecture in the VSP field and is freed from the dependency of the pretrained S3D model. In VSFT, the attention mechanism is exploited to capture spatial-temporal dependencies between the input past frames and the target future frame. We propose cross-attention guidance blocks (CAGB) to aggregate multi-level representation features to provide sufficient guidance for forecasting. We conduct comprehensive experiments on two benchmark datasets, DHF1K and Hollywoods-2. We investigate the saliency forecasting and predicting abilities of existing VSP methods by modifying the supervision signals. Experimental results demonstrate that our method achieves superior performance on both VSF and VSP tasks. [ABSTRACT FROM AUTHOR]

In recent years, hydrogels have been widely used as drug carriers, especially in the area of protein delivery. The natural silk fibroin produced from cocoons of the Bombyx mori silkworm possesses excellent biocompatibility, significant bioactivity, and biodegradability. Therefore, silk fibroin-based hydrogels are arousing widespread interest in biomedical research. In this study, a process for extracting natural silk fibroin from raw silk textile yarns was established, and three aqueous solutions of silk fibroin with different molecular weight distributions were successfully prepared by controlling the degumming time. Silk fibroin was dispersed in the aqueous solution as "spherical" aggregate particles, and the smaller particles continuously accumulated into large particles. Finally, a silk fibroin hydrogel network was formed. A rheological analysis showed that as the concentration of the silk fibroin hydrogel increased its storage modulus increased significantly. The degradation behavior of silk fibroin hydrogel in different media verified its excellent stability, and the prepared silk fibroin hydrogel had good biocompatibility and an excellent drug-loading capacity. After the protein model drug BSA was loaded, the cumulative drug release within 12 h reached 80%. We hope that these investigations will promote the potential utilities of silk fibroin hydrogels in clinical medicine. [ABSTRACT FROM AUTHOR]

With the rapid development of communication technology and portable electronic devices, the demand for the design of electronic devices with superior electromagnetic interference shielding and fire safety performance is increasing. Herein, epoxy resins (EPs) with typical segregated networks were designed by emulsion polymerization and adhesion self-assembly strategies and novel polyphosphonitrile microspheres (PZS) synthesized by precipitation polymerization were used to improve the fire safety of EP composites. It is worth noting that the oil-water (EP to water) ratio in emulsion polymerization is closely related to the microscopic morphology of the segregated network. A smaller ratio of EP to water results in more paths for electrical charge transfer through the EP composites. As a result, EP/PT-3 with a minimum oil-to-water ratio exhibits a high EMI SE (electromagnetic interference shielding effectiveness) in the X-band ranging from 40.12 to 40.30 dB at a loading of 7.20 vol% Ti 3 C 2 Al x. Besides, the synergistic action of the adjustable Ti 3 C 2 Al x network with PZS endows EP superior flame retardant performance. Compared with EP, the PHRR and THR values of EP/PT-3 are down 55.7 % and 72.6 %, respectively. There is no doubt that EP/PT-3 has provided a new route for the preparation of Ti 3 C 2 Al x -based electromagnetic shielding and fireproof materials. [Display omitted] [ABSTRACT FROM AUTHOR]

• AMPAP-NH 4 + is synthesized through a one-pot method, avoiding the need for organic solvents or halogens in the process. • AMPAP-NH 4 + could engage in chemical reactions with cellulose, resulting in the formation of stable P O C bonds. • According to the TL848-2017 testing standard, the finished wood pulp paper demonstrates excellent durable flame retardancy. Wood pulp paper (WPP) is an exemplary substrate material, yet its flammability impedes the widespread application in diverse functional fields. Herein, a highly effective, colorless and transparent phosphorus–nitrogen flame retardant, namely ammonium salt of tris(hydroxymethyl)aminomethane diphosphonic acid (AMPAP-NH 4 +) without halogen and formaldehyde, has been synthesized for WPP. Chemical grafting was employed to confer highly durable flame retardancy to the WPP, enabling it to maintain excellent fire resistance after functionalization. X-ray photoelectron spectroscopy (XPS), Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM) and Energy Dispersive Spectroscopy (EDS) analysis demonstrated that AMPAP-NH 4 + was directly linked to cellulose fibers via P O C bonds, effectively introducing P and N elements into the WPP without altering the surface morphology of the fibers. Results from thermogravimetric analysis, vertical burning test, limiting oxygen index test, cone calorimeter test and thermogravimetric-Fourier infrared spectrometer test demonstrated that the treated WPP exhibited outstanding flame retardancy. The peak heat release rate (pHRR) and total heat release (THR) values decreased by 80.8% and 70.9%, respectively, and the limiting oxygen index (LOI) increased to 41.4%. Furthermore, due to the stability of the P O C bond, the treated samples remained unignited after 168 h, with a high LOI value of 35.3%. In summary, AMPAP-NH 4 + serves as a phosphorus–nitrogen synergistic transparent flame retardant that can effectively and durably impart paper with flame retardancy without altering its surface morphology. [Display omitted] [ABSTRACT FROM AUTHOR]

With the development of microelectronics and flexible wearable electronic devices, poly(vinyl alcohol) (PVA) based thermal interface materials (TIMs) have received much attention. However, poor heat dissipation and high flammability have emerged as the key issues that restrict their application. In this work, the directional alignment of phytic acid (PA) modified boron nitride nanosheets (BNNS) and Ti 3 C 2 T x MXene in PVA matrix was achieved by ice-templated assembly, which simultaneously enhanced the flame retardancy and thermal conductivity of PVA composite films (PVA/BT x). Owing to the synergistic effect between oriented BNNS and Ti 3 C 2 T x MXene on forming the thermally conductive network within PVA matrix, the in-plane thermal conductivity of PVA/BT x -30 composite film was increased to 8.54 W m−1 K−1, much higher than that of PVA film (0.30 W m−1 K−1). Meanwhile, the heat dissipation process of PVA films applied as TIMs was simulated, which showed that PVA/BT x -30 films exhibited excellent thermal conduction and significantly decreased the operation temperature of electronic devices. Additionally, the peak heat release rate (PHRR) and total heat release (THR) of PVA/BT x -30 films were reduced by 60.3% and 62.0%, respectively, showing good flame retardancy. Importantly, the flexible Ti 3 C 2 T x nanosheets further enhanced the mechanical properties of PVA/BT x composite films. Therefore, this work provides an effective method for the preparation of high thermal conductivity and flame retardant composite materials, which has important potential applications in the microelectronics industry. [Display omitted] • The directional arrangement of BNNS and Ti 3 C 2 T x MXene was achieved by ice-templated assembly. • The flame retardancy of PVA/BT x -30 composite film was greatly improved. • The PVA/BT x -30 composite film exhibited excellent in-plane thermal conductivity. [ABSTRACT FROM AUTHOR]

[Display omitted] • The synergistic effect in the flame retardant system improves the fire safety of materials. • The release of toxic fumes limits the application of EVA materials. • The melt produced by piperazine pyrophosphate can effectively inhibit the "popcorn effect" of expandable graphite. Ethylene vinyl acetate copolymer combined with the expandable graphite (EG), pyrophosphate piperazine (PAPP) and melamine polyphosphate (MPP) was adopted to prepare the flame retardant EVA (FREVA) composites by melt blending. The research results showed that the PAPP/EG/MPP mixture with a mass ratio of 2:2:1 presented the optimal flame retardant effect. With the 25 wt% additive amount, EVA composites passed UL-94 V-0 level (1.6 mm), and the limiting oxygen index (LOI) reached 35.2 %. The peak of heat release rate (PHRR), total heat release (THR) and total smoke release decreased by 67.6 %, 60.9 % and 67.4 %, respectively. The Fourier Transform Infrared Spectroscopy (FTIR), X-ray Photoelectron Spectra (XPS), scanning electron microscope (SEM), and thermal gravimetric analyzer (TGA) were used to research the synergistic effect flame retardant mechanism of PAPP, EG and MPP in combination with the FTIR spectrometer (TG-FTIR) and the cone calorimeter test (CCT). The polyphosphoric acid non-volatile substances generated by PAPP, attached to the surface of the EG carbon layer, effectively restrained the "popcorn effect", and the char layer presented a compact and stable structure. [ABSTRACT FROM AUTHOR]

[Display omitted] • A phytic acid-chitosan-polyacrylamide (PA-CS-PAM) coating is prepared in one step; • PA-CS-PAM coating endows with an flame retardant performance and conductivity; • The resulted humidity sensors exhibit repeatability and reliability for hydrophilia; • The signal output is stable in the strain interval due to collapsible paper; Paper fabric is a good alternative for flexible sensors, but it remains challenging to engineer a versatile paper-based sensor for monitoring the body's condition. Even worse, the practical applications are strongly restricted due to their intrinsically high flammability, especially the paper-based construction in the short-circuit environment will cause fire, threatening life and health. Here, we report a facile strategy for constructing a flame retardant, healthy and eco-friendly paper-based sensor by assembly of natural phytic acid, chitosan and acrylamide. Successfully engineering paper-based sensors with great humidity and pressure sensing could lead to breakthroughs in life monitoring. The resulting paper-based sensors exhibit excellent repeatability and reliability for respiratory monitoring. Specifically, the signal output is stable in the maximum and minimum strain interval, this makes the resulting paper highly durable and suitable as a pressure sensor. Notably, the modified paper exhibits excellent flame retardant properties due to the synergistic effect between the gas phase and the condensation phase, thus, they exhibit a limited oxygen index value of 27.8 %, the paper self-extinguishes immediately after the ignition of 15 s. And even, the carbon length of the modified paper is less than 120 mm in the vertical combustion monitoring. This work offers an environmentally friendly, safe, and sustainable method for the fabrication of paper-based flexible sensors derived from renewable materials, and hold great potential for the development of environmentally friendly electronic devices. [ABSTRACT FROM AUTHOR]

In this paper, a novel distributed adaptive neurodynamic approach (DANA) based on proportional integral technique is proposed to solve distributed optimization problem on multi-agent systems. The goal is that all agents reach consensus in finite time and converge to the optimal solution of the global objective function in fixed time. In the proposed approach, the proportional technique drives all agents to reach consensus, and the integral technique is used to offset the influence of the gradient term of the objective function. On the other hand, in order to avoid the prior estimation of gain parameter and the global gradient information, as the main contribution of this paper, the adaptive idea is considered into proportional integral technique. The results show that the adaptive integral technique can automatically adjust the gain according to the maximum consensus error between agents, so as to ensure that agents can achieve consensus in finite time. Then the theoretical results are applied to voltage distribution and logistic regression. Numerical simulation verifies the effectiveness of DANA. [ABSTRACT FROM AUTHOR]

6 DoF pose estimation problem aims to estimate the rotation and translation parameters between two coordinates, such as object world coordinate and camera world coordinate. Although some advances are made with the help of deep learning, how to full use scene information is still a problem. Prior works tackle the problem by pixel-wise feature fusion but need to randomly select numerous points from images, which can not satisfy the demands of fast inference simultaneously and accurate pose estimation. In this work, we present a novel deep neural network based on dynamic keypoint selection designed for 6DoF pose estimation from a single RGBD image. Our network includes three parts, instance semantic segmentation, edge points detection and 6DoF pose estimation. Given an RGBD image, our network is trained to predict pixel category and the translation to edge points and center points. Then, a least-square fitting manner is applied to estimate the 6DoF pose parameters. Specifically, we propose a dynamic keypoint selection algorithm to choose keypoints from the foreground feature map. It allows us to leverage geometric and appearance information. During 6DoF pose estimation, we utilize the instance semantic segmentation result to filter out background points and only use foreground points to finish edge points detection and 6DoF pose estimation. Experiments on two commonly used 6DoF estimation benchmark datasets, YCB-Video and LineMoD, demonstrate that our method outperforms the state-of-the-art methods and achieves significant improvements over other same category methods time efficiency. • A simple but effective dynamic keypoint selection algorithm that leverages texture and geometry information of object. • A filtering background points algorithm significantly improves pose estimation time efficiency. • State-of-the-art 6DoF pose estimation performance on the YCB-Video and LineMOD datasets. [ABSTRACT FROM AUTHOR]

Accurate detection of crystal defects from microstructure micrographs is essential for evaluating the mechanical and functional properties of metallic materials. Emerging deep learning techniques have great potential for application in crystal defects detection, but the excellent detection accuracy requires domain knowledge and micrograph datasets. Here we target to detect and distinguish nanoscale dislocation loops (DLs) and stacking fault tetrahedrons (SFTs) in transmission electronic microscope (TEM) micrographs. The micrograph datasets of crystal defects derive from copper prepared by rapid quenching and post-annealing. Applying standard U-Net architecture, the maximum Mean Intersection over Union (mIOU max) of 3-class model is 54.29%, while it is 73.81% and 74.67% of DL 2-class and SFT 2-class model. Results suggest that excellent detection accuracy of deep learning model could be achieved by learning features from specific defect datasets. These findings provide a new strategy for accurate detection of crystal defects in the field of material engineering. Accurate detection of crystal defects from microstructure micrographs is essential for evaluating the mechanical and functional properties of metallic materials. Emerging deep learning techniques have great potential for application in crystal defects detection, but the excellent detection accuracy requires domain knowledge and micrograph datasets. Here we target to detect and distinguish nanoscale dislocation loops (DLs) and stacking fault tetrahedrons (SFTs) in transmission electronic microscope (TEM) micrographs. The micrograph datasets of crystal defects derive from copper prepared by rapid quenching and post-annealing. Applying standard U-Net architecture, the maximum Mean Intersection over Union (mIOU max) of 3-class model is 54.29%, while it is 73.81% and 74.67% of DL 2-class and SFT 2-class model. Results suggest that excellent detection accuracy of deep learning model could be achieved by learning features from specific defect datasets. These findings provide a new strategy for accurate detection of crystal defects in the field of material engineering. [Display omitted] • Quenching coupled with post-annealing techniques is an efficient strategy for introducing vacancy defects in FCC metals. The subsequent quantitative TEM characterization supports the construction of vacancy defect micrograph dataset, which is suitable for training deep learning semantic segmentation models. • Deep learning semantic segmentation technology is successfully applied on the analysis of nanoscale defects in TEM micrographs. Results suggest that constructing a defect micrograph dataset containing clear microstructure features can further improve the comprehensive performance of deep learning models. [ABSTRACT FROM AUTHOR]

Saikosaponin a (SSa), a triterpene saponin extracted from Bupleurumfalcatum L. , has been reported to have anti-inflammatory activity. In the present study, we investigated the effects of SSa on lead-induced kidney injury in common carp. Kidney histological changes were detected by H&E staining. The levels of TNF-α, IL-1β, MPO, MDA, GSH, and SOD activity were also measured. Furthermore, the NF-κB and Nrf2 signaling pathways were tested by western blot analysis. The results showed that lead-induced kidney histological change was attenuated by SSa. Lead-induced TNF-α, IL-1β, MPO, and MDA production were also suppressed by SSa. Meanwhile, lead could decrease GSH level and SOD activity and the decreases were inhibited by SSa. Furthermore, we found SSa significantly inhibited lead-induced NF-κB translocation. In addition, the expression of Nrf2 and HO-1 were increased by the treatment of SSa and Keap1 expression was decreased by SSa. In conclusion, this study indicated that SSa inhibited lead-induced kidney injury in carp through suppressing inflammatory and oxidative responses, and the mechanism may be involved in the inhibition of NF-κB and activation of Nrf2 signaling pathway. Unlabelled Image • Saikosaponin a (SSa) significantly inhibited lead-induced TNF-α, IL-1β, MPO, and MDA production. • The activity of SOD and the content of GSH were increased by SSa. • SSa dose-dependently suppressed lead-induced NF-κB activation. • SSa increased Nrf2 and HO-1 expression in a dose-dependent manner. [ABSTRACT FROM AUTHOR]

Heat-stable antifungal factor (HSAF) isolated from Lysobacter enzymogenes has shown a broad-spectrum of antifungal activities. However, little is known about its mode of action. In this study, we used the model filamentous fungus Neurospora crassa to investigate the antifungal mechanism of HSAF. We first used HSAF to treat the N. crassa strain at different time points. Spore germination, growth phenotype and differential gene expression analysis were conducted by utilizing global transcriptional profiling combined with genetic and physiological analyses. Our data showed that HSAF could significantly inhibit the germination and aerial hyphae growth of N. crassa. RNA-seq analysis showed that a group of genes, associated with cell wall formation and remodeling, were highly activated. Screening of N. crassa gene deletion mutants combined with scanning electron microscopic observation revealed that three fungal cell wall integrity-related genes played an important role in the interaction between N. crassa and L. enzymogens. In addition, Weighted Gene Co-Expression Network Analysis (WGCNA), accompanied by confocal microscopy observation revealed that HSAF could trigger autophagy-mediated degradation and eventually result in cell death in N. crassa. The findings of this work provided new insights into the interactions between the predatory Lysobacter and its fungal prey. [ABSTRACT FROM AUTHOR]

The application of wireless electronic devices inevitably leads to electromagnetic pollution. Three-dimensional porous foam structures have been shown to be an effective strategy to construct high-performance microwave absorbers. Herein, the honeycomb-like hierarchical foams were fabricated via melamine-formaldehyde (MF) foam as substrate coupled with flaky carbonyl iron (CI)-MXene lamellae. Inside the synthesized composite foam, flaky CI is interlaced into MXene nanosheets to form hierarchical structures, which effectively alleviates the originally self-stacking phenomenon and provides numerous heterogeneous interfaces, resulting in the optimization of impedance matching. In addition, owing to the construction of consecutive electromagnetic networks, the CI-MXene/MF foams exhibit significant dielectric loss and magnetic loss capability. Moreover, the honeycomb-like structure offers multiple reflections/scattering for electromagnetic waves, which effectively prolongs the transmission paths, further improving the microwave attenuation ability. Hence, the synthesized composite foam exhibits superior microwave absorption performance with a minimum refection loss of −61.4 dB and a maximum effective absorption band of 7.18 GHz. The construction of unique microstructure and regulation of electromagnetic components is demonstrated to be feasible strategies for the design of efficient microwave absorbers in future. [Display omitted] • The hierarchical structure formed by flaky carbonyl iron (CI)-MXene lamellae effectively alleviates the self-stacking phenomenon. • The honeycomb-like coherent electromagnetic network is conducive to optimizing impedance matching and enhancing microwave dissipation capability. • The ultralight composite foams achieve superior microwave absorption performance with broad effective absorption bands. [ABSTRACT FROM AUTHOR]

One-dimensional rod-shaped Ag 2 Mo 2 O 7 /BiOI composite photocatalysts were prepared via a facile hydrothermal-calcining method. The structure, morphologies, and elemental nature of the photocatalysts were characterized. Ag 2 Mo 2 O 7 /BiOI was formed through the attachment of BiOI nanosheets on the surface of Ag 2 Mo 2 O 7 rods, resulting in a 1D/2D heterojunction with larger interface area, more surface-active sites, and faster charge transfer channel. The Ag 2 Mo 2 O 7 /BiOI heterogeneous exhibited excellent and stable photodegradation performances for Rhodamine B (RhB) and Tetracycline (TC) under visible light irradiation. At the optimum composition of 15 wt% BiOI, the Ag 2 Mo 2 O 7 /BiOI heterojunction exhibited a highest degradation rate for RhB and TC, which were 70 times and 16 times higher than those of Ag 2 Mo 2 O 7 , respectively. The improved photocatalytic performance can be attribute to the efficient separation of photoinduced charges caused by the formation of the n-n heterojunction between Ag 2 Mo 2 O 7 and BiOI. Additionally, the dominant active species of photocatalytic reaction were determined to be •O 2 − and h+ by free radical capture experiments and ESR test. In addition, a toxicity test with E. coli was carried out, which shows that the photodegraded Tetracycline solution is basically harmless to E. coli. Based on the experimental results and hybrid density functional theory calculation, a plausible photocatalytic mechanism of n-n heterojunction was proposed. [Display omitted] • One-dimensional rod-shaped Ag 2 Mo 2 O 7 /BiOI were synthesized by a solvothermal method. • Ag 2 Mo 2 O 7 /BiOI can efficiently degrade Tetracycline. • The toxicity of the degraded tetracycline solution was low by E. coli activity detection. • Charge carriers were separated due to the n-n heterojunction and large interface area. [ABSTRACT FROM AUTHOR]

Janus membranes are ideal devices for sustainable fog collection to help alleviate the global water shortage. However, it remains challenging to obtain Janus membranes with sufficient flexibility, low cost, chemical stability and anti-pollution properties. In this work, flexible Janus membranes based on cellulose substrate were fabricated for effective fog collection. We undertook a detailed study of the formation mechanisms of the Janus membranes, which contain an asymmetric growth of titania and a single-side hydrophobic modification. The proper thickness and discontinuity of the hydrophobic layer allow unidirectional water penetration, and the punched holes provide rapid water conveyance channels. The perforated Janus membranes showed the highest fog collection efficiency (∼3 times as much as that of a totally hydrophilic membrane) of the various types that were prepared. Furthermore, the Janus membranes exhibit desirable anti-acid, anti-pollution and reusability properties, which demonstrate application potential in complicated practical settings. • Asymmetric in-situ TiO 2 growth on flexible cellulose material. • Novel fabrication methods for cellulose Janus membranes. • Rapid unidirectional water penetration based on perforated structure. • Enhanced fog collection efficiency compared with totally hydrophilic membranes. [ABSTRACT FROM AUTHOR]

Endometritis is an inflammatory of the inner lining of the uterus caused by bacterial infections that affect female reproductive health in humans and animals. Neutrophil extracellular traps (NETs) have the ability to resist infections that caused by pathogenic invasions. It has been proved that the formation of NETs is related to certain inflammatory diseases, such as mastitis and chronic obstructive pulmonary disease (COPD). However, there are sparse studies related to NETs and endometritis. In this study, we investigated the role of NETs in lipopolysaccharide (LPS)-induced acute endometritis in mice and evaluated the therapeutic efficiency of DNaseI. We established LPS-induced endometritis model in mice and found that the formation of NETs can be detected in the mice uterine tissues in vivo. In addition, DNaseI treatment can inhibit NETs construction in LPS-induced endometritis in mice. Moreover, myeloperoxidase (MPO) activity assay indicated that DNaseI treatment remarkably alleviated the inflammatory cell infiltrations. ELISA test indicated that the treatment of DNaseI significantly inhibited the expression of the proinflammatory cytokines TNF-α, and IL-1β. Also, DNaseI was found to increase proteins expression of the uterine tissue tight junctions and suppress LPS-induced NF-κB activation. All the results indicated that DNaseI effectively inhibits the formation of NETs by blocking the NF-κB signaling pathway and enhances the expression of tight junction proteins, consequently, alleviates inflammatory reactions in LPS-induced endometritis in mice. • DNaseI inhibited the formation of NETs in the uterine tissues. • DNaseI inhibited the inflammatory response in uterus induced by LPS. • DNaseI inhibited the permeability of uterus barrier induced by LPS. • DNaseI inhibited the production of pro-inflammatory cytokines via inhibiting the activation of NF-κB signal pathway. [ABSTRACT FROM AUTHOR]

Abstract: CuO nanorods have been prepared through a facile microwave-hydrothermal method with the aid of polyethylene glycol 400, ethylene glycol and urea. The X-ray diffraction and transmission electron microscopy results showed that the prepared CuO nanocrystalline had a rod-like shape with diameters of 15–20nm and the average length of 80nm. Compared with commercial CuO powder, the CuO nanorods showed an enhanced catalytic activity for the synthesis of sodium humate and the degradation of Rhodamine B. The yield of sodium humate catalyzed by CuO nanorods was 75.78%, which was higher than that of the commercial CuO powder (52.8%). In addition, the CuO nanorods showed a relatively faster degradation rate of RhB. The CuO nanorods prepared by this facile method demonstrated their potential applications in the field of catalysis. [Copyright &y& Elsevier]