Search

Human skeleton contains intuitive information of actions and has high robustness in dynamic environment. Therefore, it has been widely studied in action recognition tasks. Most existing methods of skeleton recognition are based on graph convolutional networks (GCNs), which extract the topological structure of graphs to describe the dependencies between joints. However, the GCNs pay excessive attention to skeleton structure and neglect the feature information of skeleton joints. Accordingly, how to fuse the feature of both skeleton structure and joints is a problem to be solved. In addition, non-linear temporal convolutional network (TCN), which has more robustness and learning capability, is rarely investigated in existing methods. With the comprehensive consideration of the dependence between structure and feature on graphs, we propose a novel structure-feature fusion adaptive GCN (SFAGCN) for skeleton based action recognition. The topological structure of the skeleton graph and the feature of the joints can be fused by the decoupled spatiotemporal correlation in our model effectively. The relevance of spatiotemporal data can be preserved well by the fusion strategy, with data integrity ensured. Moreover, Gated TCN is used to extract temporal feature, improving the network performance further. We choose two-stream adaptive GCNs and shift-GCN as the baseline. To demonstrate the effectiveness of our methods, extensive experiments are implemented on the three large-scale datasets, namely, NTU-RGBD 60, NTU-RGBD 120 and Kinetics-Skeleton 400. The accuracy of top-1 on above datasets are improved by more than 0.6% on average, where the performance of SFAGCN exceeds the state-of-the-art methods.

In this study, a Superhydrophobic (SH) surface with ultrahigh adhesive force and good anti-scratching property was created on elastomeric substrate by thiol-ene click chemistry. The elastomer substrate was prepared by solution mixing of styrene butadiene rubber (SBR), polar fluorinated monomer, multi-functional thiol, and photoinitiator. Then the silanized nano-silica was simply sprayed on the elastomer substrate to prepare the SH coating. Through UV irradiation, thiol-ene click reaction between multi-functional thiol and double bonds of SBR, fluorinated monomer and silanized nano-silica simultaneously occurs, leading to the crosslinking of SBR, the polymerization of fluorinated monomer, the graft of fluorinated monomer on silanized nano-silica, and the covalently bonding of fluorinated monomer with SBR substrate. As a result, the polymerized fluorinated monomer bonded silanized nano-silica is covalently bonded with the SBR substrate. Meanwhile, the silanized nano-silica forms the micro-nano structure on the substrate and the polymerized fluorinated monomer decreases the surface energy of the substrate. A stabilized SH coating with ultrahigh adhesive force (123 μN) to water was achieved. In addition, the SH surface shows strong interfacial bonding and a good anti-scratching property because of the covalent bonding of SH coating with the SBR substrate. This is a facile, simple and efficient method to fabricate SH coating on the elastomeric substrate, which may open a door to fabricate SH coating on other surfaces and provide more potential applications.

A glass-based IPD band-pass chip filter for 5G N77 application is presented in this paper. This filter is composed of five shunt LC resonators. The low loss passband performance is realized by two LC resonators connected in shunt near the input and output terminals. The high rejection property for the stopband is realized by three LC resonators connected in series on the main branch of this filter. This filter is successfully manufactured with a commercial glass IPD process. Its chip size is $\mathbf{1.6\times 0.8{mm}^{2}}$ . The measurement results show that, the working frequency band of this filter is 3.2-4.0GHz, the maximum passband loss is only 2.0dB, and the stopband rejection is higher than 30dB at the designed zero-point frequencies.

We study a Schrödinger system of four equations with linear coupling functions and nonlinear couplings, including the case that the corresponding elliptic operators are indefinite. For any given nonlinear coupling β > 0 {\beta>0} , we first use minimizing sequences on a normalized set to obtain a minimizer, which implies the existence of positive solutions for some linear coupling constants μ β , ν β {\mu_{\beta},\nu_{\beta}} by Lagrange multiplier rules. Then, as β → ∞ {\beta\to\infty} , we prove that the limit configurations to the competing system are segregated in two groups, develop a variant of Almgren’s monotonicity formula to reveal the Lipschitz continuity of the limit profiles and establish a kind of local Pohozaev identity to obtain the extremality conditions. Finally, we study the relation between the limit profiles and the optimal partition for principal eigenvalue of the elliptic system and obtain an optimal partition for principal eigenvalues of elliptic systems.

Cyanobacterial blooms continue to serve as one of the most serious global issues threatening water supply and human health. During cyanobacterial bloom season, a large •OH-yield equipment was developed and installed after coagulation settling in a 12000 ton/day drinking water treatment system in Xiamen, China. An •OH concentration of 7.76∼57.8 μmol/L was formed by using the oxygen activated species generated by strong ionisation discharge combining with the effect of water jet cavitation. •OH pre-treatment at a dose of 1.0 mg/L inactivated cyanobacterial blooms in the process of conveying bloom water within only 20s, which were then removed by sand filtration. Under SEM observation, dominant Microcystis sp. colonies connected by mucilage were dispersed into individuals that still retained the cell integrity, indicating no release of intracellular organic matter (IOM). According to a flow cytometry analysis, the main cause of •OH inactivation was the breakage of DNA strands. Meanwhile, the •OH-mineralized microcystin-LR was by breaking the C=C conjugated diene bond and crucial opening the persistent benzene ring to carboxylic acid m/z 158.0. During •OH pre-treatment of 1.0 mg/L and NaClO disinfection of 0.5 mg/L, all water quality indexes and disinfection by-product (DBP) contents complied with the Chinese Sanitary Standards for Drinking Water. Therefore, the •OH based on the strong ionisation discharge showed great prospect for large-scale drinking water treatment in the removal of cyanobacterial blooms while retaining cell integrity as well as the degradation of toxins.

A hydroxyl radical ( OH) ballast water treatment system (BWTS) was developed and applied to inactivate entrained organisms in a 10,000-ton oceanic ship, where OH was produced by a strong ionization discharge combined with a water jet cavitation effect. The calculated OH generation rate was 1373.4 μM min−1 in ballast water, which is much higher than that in other advanced oxidative processes such as photocatalysis. As a result, non-indigenous red tide algae were inactivated to meet the ballast water discharge standards ( OH inactivation. A risk assessment of the OH treatment system was conducted, and the ratios of predicted environmental concentrations to predicted no effect concentrations of all detected disinfection byproducts were less than 1, even at a worst-case oxidant concentration of 2.41 mg L−1. Ship ballast water treated using OH inactivation is safe for marine environments. Finally, the energy consumption and operational costs of the OH BWTS were found to be 0.033 kWh m−3 and CNY 0.03 m-3, respectively.

In this paper, we study the nonlinear Klein-Gordon systems arising from relativistic physics and quantum field theories $$\left\{\begin{array}{lll} u_{tt}- u_{xx} +bu + \varepsilon v + f(t,x,u) =0,\; v_{tt}- v_{xx} +bv + \varepsilon u + g(t,x,v) =0 \end{array}\right. $$ where $u,v$ satisfy the Dirichlet boundary conditions on spatial interval $[0, ��]$, $b>0$ and $f$, $g$ are $2��$-periodic in $t$. We are concerned with the existence, regularity and asymptotic behavior of time-periodic solutions to the linearly coupled problem as $\varepsilon$ goes to 0. Firstly, under some superlinear growth and monotonicity assumptions on $f$ and $g$, we obtain the solutions $(u_\varepsilon, v_\varepsilon)$ with time-period $2��$ for the problem as the linear coupling constant $\varepsilon$ is sufficiently small, by constructing critical points of an indefinite functional via variational methods. Secondly, we give precise characterization for the asymptotic behavior of these solutions, and show that as $\varepsilon\rightarrow 0$, $(u_\varepsilon, v_\varepsilon)$ converge to the solutions of the wave equations without the coupling terms. Finally, by careful analysis which are quite different from the elliptic regularity theory, we obtain some interesting results concerning the higher regularity of the periodic solutions.

This paper considers the spreading speed of cooperative nonlocal dispersal system with irreducible reaction functions and non-uniform initial data. Here the non-uniformity means that all components of initial data decay exponentially but their decay rates are different. It is well-known that in a monostable reaction-diffusion or nonlocal dispersal equation, different decay rates of initial data yield different spreading speeds. In this paper, we show that due to the cooperation and irreducibility of reaction functions, all components of the solution with non-uniform initial data will possess a uniform spreading speed which non-increasingly depends only on the smallest decay rate of initial data. The nonincreasing property of the uniform spreading speed further implies that the component with the smallest decay rate can accelerate the spatial propagation of other components. In addition, all the methods in this paper can be carried over to the cooperative system with classical diffusion (i.e. random diffusion).

In this paper, we study important Schr��dinger systems with linear and nonlinear couplings \begin{equation}\label{eq:diricichlet} \begin{cases} -��u_1-��_1 u_1=��_1 |u_1|^{p_1-2}u_1+r_1��|u_1|^{r_1-2}u_1|u_2|^{r_2}+��(x)u_2~\hbox{in}~\mathbb{R}^N,\\ -��u_2-��_2 u_2=��_2 |u_2|^{p_2-2}u_2+r_2��|u_1|^{r_1}|u_2|^{r_2-2}u_2+��(x)u_1~ \hbox{in}~\mathbb{R}^N,\\ u_1\in H^1(\mathbb{R}^N), u_2\in H^1(\mathbb{R}^N),\nonumber \end{cases} \end{equation} with the condition $$\int_{\mathbb{R}^N} u_1^2=a_1^2, \int_{\mathbb{R}^N} u_2^2=a_2^2,$$ where $N\geq 2$, $��_1,��_2,a_1,a_2>0$, $��\in\mathbb{R}$, $2