Efficient single-image super-resolution (ESISR) primarily aims to enhance super-resolution (SR) performance while keeping model complexity low, making it more suitable for deployment on edge devices. However, the limited receptive field caused by conventional convolution's locality restricts the backbone networks and attention modules from effectively capturing nonlocal features, leading to suboptimal SR performance. Additionally, insufficient interaction between high- and low-frequency feature information results in incomplete feature representation. To overcome these limitations, we propose a novel ESISR network called the lightweight dual-kernel information aggregation network (LDIAN). First, we design a dual-kernel convolution (DKC) that combines depth-wise 1-D convolution and dilated convolution to efficiently extract richer image features in an expanded receptive field while minimizing model complexity. Building upon DKC, we further develop a dual-kernel enhanced convolution (DEConv) and a dual-kernel enhanced distillation block (DEDB). Additionally, we propose a lightweight dual-kernel attention (DKA) mechanism to focus on more representative features for SR reconstruction. Second, we design an innovative feature fusion structure named the information aggregation block (IAB) to integrate spatial features and strengthen the interaction between high- and low-frequency information, thereby improving feature representation. Extensive quantitative and qualitative experiments demonstrate that the LDIAN achieves state-of-the-art performance with an optimal balance between model performance and complexity. Notably, compared to SRFormer-light, LDIAN-L delivers superior performance across five standard datasets while requiring only about 50% of the model's FLOPs.

The balanced extraction of both non-local and local features represents a critical requirement for effective image super-resolution (SR). While transformer-based self-attention (SA) mechanisms demonstrate superior non-local modeling capabilities, their substantial computational demands limit practical deployment. To address this efficiency-performance trade-off, the Spatial-Gate Self-Distillation Network (SGSDN) implements a dual-capacity architecture combining: an SA-like (SAL) module employing strategically dilated 1D depthwise convolutions in horizontal and vertical orientations for efficient non-local feature extraction, and a lightweight local spatial-gate (LKG) block optimized for local detail preservation. Moreover, the proposed spatial-gate self-distillation block (SGSDB) further enhances performance through an optimized distillation structure that simultaneously processes both feature types while minimizing memory overhead. Experimental results demonstrate SGSDN’s superior performance-complexity balance, with benchmark evaluations showing comparable accuracy to SwinIR-light while requiring only 25% of the computational resources (FLOPs) and 25% of parameters, attributable to its avoidance of computationally intensive matrix operations.

Deep learning technologies such as YoloX have achieved impressive progress in face mask detection recently. However, the neck network used in YoloX network may lead to severe confounding effect in feature mapping due to the inherent defect of channel reduction in hybrid fusion, which affects its precise localization ability of mask-wearing targets. To tackle this issue, we present a new FPN network structure (ICE-FPN) based on channel-enhanced feature pyramid network (CE-FPN) in this paper, which can mitigate the YoloX network confounding effect while reducing the number of parameters and computational effort caused by CE-FPN. Experiments conducted on the WMD dataset show that the mAP0.5 of the model improves from 99.54% to 99.62% and the mAP0.75 improves from 89.47% to 91.35%. The ablation and comparison experiments demonstrate that the proposed ICE-YoloX has achieved superior performance over existing methods.

Abstract Facial expressions are an important part of human emotional signals and their recognition has become an important topic of research in the field of pattern recognition. Deep learning based methods have achieved great success in the recognition of facial expressions. However, with the evolution of convolution neural networks and the increased network depth, these methods suffer from problems such as degraded network performance and loss of feature information. To address these problems, a novel facial expression recognition algorithm based on an improved residual neural network is proposed. First, a residual neural network is designed to extract deep features while retaining the shallow ones. This can effectively prevent the degradation of network performance. Moreover, when the gradient of the Rectified Linear Units activation function used in the residual module is 0, it will inactivate the neurons and cause a loss of feature information. To address this, the Mish activation function is used instead. The slight allowance for negative values in Mish improves the gradient flow. Next, an inception module is introduced to obtain richer feature information under the same receptive field. Finally, by conducting verification experiments on the public datasets CK+ and KDEF, the authors manage to solve the problems of degraded network performance and insufficient information from extracted features, achieving recognition accuracy rates of 96.37% and 93.38% on the two datasets, respectively.

This article investigates the polynomial fuzzy impulsive control problem for positive nonlinear systems subject to parameter uncertainties. The positive nonlinear systems are represented as positive interval type-2 (IT2) polynomial fuzzy-model-based systems, while the parameter uncertainties are captured by the IT2 membership functions (MFs). Considering that the controller being designed needs to be implementable and practical, in addition to ensuring the positivity and stability of the system, there are still some potential requirements for the controller, such as lower implementation cost and control cost. To reduce the implementation cost, the IT2 polynomial fuzzy impulsive controller is designed under the imperfect premise matching concept that the premise MFs and the number of rules of fuzzy controller are different from those of fuzzy model. To reduce the control cost, the impulsive controller that can tolerate larger impulse control interval (ICI) is designed by the following two novel methods. The first one is the proposed impulse-time-dependent discretized polynomial copositive Lyapunov function whose Lyapunov variable is designed as the interpolation polynomial function of time. The second one is the advanced IT2 membership-function-dependent (IT2MFD) analysis method which is improved by a novel footprint of uncertainty partitioning method proposed by this paper, so that this advanced IT2MFD method can alleviate the negative impact of parameter uncertainties on the ICI. Finally, a simulation example using lipoprotein metabolism and potassium ion transfer model as the nonlinear system to be controlled is used to verify the effectiveness of the proposed impulsive control methods.

This article focuses on expanding the estimation of the domain of attraction (DOA) for discrete-time positive nonlinear systems subject to input saturation and parameter uncertainties. To facilitate analysis and design, the interval type-2 (IT2) polynomial fuzzy model is used to represent the nonlinear plant and capture uncertainties. Combining with the IT2 polynomial fuzzy controller, the discrete-time positive IT2 polynomial fuzzy-model-based (PIT2PFMB) control system is formed to facilitate analysis. To enlarge the estimation of DOA of the discrete-time PIT2PFMB system, polyhedron is used to characterize the DOA with the help of linear copositive Lyapunov function (LCLF). Referring to the nonconvex conditions derived by LCLF, an effective convexification method is proposed in this article. For comparison purposes, the saturation-dependent-Lyapunov-function-based method is extended to the PIT2PFMB control system by adding the corresponding positivity conditions. In addition, this article attempts to enlarge the estimation of the DOA by improving the IT2 membership-function-dependent (IT2MFD) method and extending it to all conditions, including the stability conditions and the DOA estimation conditions. Finally, an example with simulation results is given to verify the effectiveness of all the methods proposed in this article for expanding the estimation of the DOA.

In this paper, the stability and positivity of positive polynomial fuzzy model based (PPFMB) control system are investigated, in which the positive polynomial fuzzy model and positive polynomial fuzzy controller are allowed to have different premise membership functions from each other. These mismatched premise membership functions can increase the flexibility of controller design; however, it will lead to the conservative results when the stability is analyzed based on the Lyapunov stability theory. To relax the positivity/stability conditions, the improved Taylor-series-membership-functions-dependent (ITSMFD) method is introduced by introducing the sample points information of Taylor-series approximate membership functions, local error information and boundary information of substate space of premise variables into the stability/positivity conditions. Meanwhile, the ITSMFD method is extended to the PPFMB control system with input saturation to relax the estimation of domain of attraction. Finally, simulation examples are presented to verify the feasibility of this method.

As a highly complex multi-input and multi-output system, blast furnace plays an important role in industrial development. Although much research has been done in the past few decades, there still exist many problems, such as the modeling and control problems. In view of these reasons, this paper is concerned with developing a Wiener model to predict the silicon content of blast furnace. Unlike traditional Wiener model, this paper avoids the optimization of high number of model parameters. The Wiener model here is composed of a basis filter filter expansion named Laguerre filter and a linear programming support vector regression (LP-SVR). They are used to represent the linear dynamic component and the nonlinear static element. Take the advantages that Laguerre filter can approximate linear systems with a lower model and order and LP-SVR can achieve a sparse solution, the proposed Wiener model not only improves the prediction accuracy but also reduces the computation complexity. Simulation results show that this Wiener model is suitable for the prediction of blast furnace silicon content.