In smoke interference scenarios, visible image degradation significantly limits the availability of detailed information for fusion. Moreover, a single fusion network often struggles to effectively extract complementary and key information. To overcome these challenges, we propose a infrared and visible image fusion method tailored for smoke interference, termed the Repeated Key Feature Embedding Fusion Network (REFusion). To fully leverage low-frequency and high-frequency information from source images, we design the network with a pre-fusion stage and a fine-fusion stage. To address the heterogeneity between infrared and visible images, we introduce a Cross-Source Context Association Network (CSCAN). By integrating the self-attention mechanism and a cascading pyramid pooling module, we compute attention interaction weights across modes to capture the most relevant context for fusion. To prevent the fusion from favoring one source disproportionately, we enhance the fine-fusion stage with a Differential Information Reinforcement Module (DIRM) and a three-branch network. This structure establishes long-term dependencies between complementary information and excels in capturing high-frequency details. We further design a tailored loss function to enhance the quality of fused images. Extensive experiments demonstrate that REFusion surpasses nine state-of-the-art methods in both visual quality and quantitative metrics, particularly in reproducing scenes obscured by smoke.

In the textile manufacturing industry, fabric defect detection is essential for ensuring product quality. Traditional approaches based on Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) often encounter scalability issues, particularly due to the high computational complexity of the self-attention mechanism in ViTs. To address these limitations, this study introduces FabricMamba, a real-time defect detection framework built on the You Only Look Once version 8 (YOLOv8) CNN architecture. The model enhances detection precision and efficiency for complex fabric defects in high-resolution images while minimizing computational cost. YOLOv8 was selected as the base model due to its strong balance between accuracy and inference speed, which is critical in fast-paced textile production settings. FabricMamba extends YOLOv8 with several innovations: the Parallel Large Separable Kernel Attention (P-LSKA) mechanism for multi-scale perception, the Visual State Space Module (MVSS) for long-range dependency modeling, the lightweight DySample module for reduced resource usage, and Programmable Gradient Information (PGI) to optimize training without increasing inference complexity. Extensive evaluations were conducted using a proprietary industrial fabric defect dataset and two public benchmarks, TILDA Textile Texture Database and FDDS Object Detection Dataset. FabricMamba achieved a mean Average Precision (mAP) of 90.0 %, 97.7 %, and 39.1 % on the respective datasets, outperforming the YOLOv8 baseline by 1.8 %, 2.3 %, and 2.0 %. Compared to Mamba-YOLO, FabricMamba reduced model size and computational requirements by 36.7 % and 33.1 %, respectively, with recall improving by 2.9 %, 1.4 %, and 4.0 %. These results confirm the model effectiveness and practical potential for industrial fabric inspection tasks.

LoRa backscatter achieves long-range communication at the cost of only tens of micro-watts of power when implemented in integrated circuits (ICs), which makes it a potential enabler for massive IoT. However, despite the above advantages, we find that the current tag design wastes approximately 72% of the chip functional area due to the use of large-size inductors in the impedance loads array. This inefficiency significantly increases the cost per chip during mass production. To address this die area issue, we propose OsTAG, a novel LoRa backscatter design that eliminates all inductors in the impedance loads array while maintaining the same quantization resolution. The key innovation lies in the creation of virtual impedance using oversampling. Realizing such design requires overcoming the technical challenges of mitigating approximation error and managing oversampling clock imperfections. To validate our design, we develop prototype and an IC, demonstrating promising results in terms of both performance and efficiency. OsTAG is expected to cost only 28% of the existing chip area while achieving comparable hundred-meter level communication range.

We propose a simple but effective approach to detect salient objects by exploring both patch-level and object-level cues under the framework of absorbing Markov chain. Saliency detection is carried out in a two-stage scheme. In the first stage, we conduct random walk on absorbing Markov chain with coarsely selected background seeds in the boundary. The result is integrated with a objectness map which is generated by finding potential object candidates to boost the saliency detection for object completeness. And in the second stage, we use the refined background seeds computed by the first stage as absorbing nodes for the absorbing Markov chain to obtain the final saliency map. Experimental results on four publicly available datasets demonstrate the robustness and efficiency of our proposed approach against 8 state-of-the-art methods in terms of five performance criterions.

ABSTRACT Energy‐efficient routing becomes one of the most critical technologies for sustaining the overall network lifetime of wireless sensor networks. In this paper, we propose a novel data transmission scheme between a number of specified source nodes and the single sink, which can efficiently restrict the usage frequency of each relay node, measured by the number of source nodes using it for data transmission. On the basis of the importance of source nodes that is closely related to deployed location, they form a descending sequence such that each node finds the minimum energy path earlier than the succeeding one. Then, the energy‐efficient multiple path algorithm with the computational complexity ofO(n3) is developed for deriving the minimum energy paths, wherenis the number of nodes in the network. Also, a polynomial algorithm is presented for deriving the range of the feasible values ofN0serving as the threshold of the usage frequency of relay nodes, in which each can guarantee the existence of the solution. Further, we theoretically investigate the existence of the solution and the tree‐structured solution usingm‐ary tree. Extensive simulation results show that our proposed scheme can achieve significant performance enhancement. Copyright © 2012 John Wiley & Sons, Ltd.

In this study, we propose a module-level Evolvable Hardware (EHW) approach to design synchronous sequential circuits and minimize the circuit complexity (the number of logic gates and wires used). Firstly, we use Genetic Algorithm (GA) to implement state simplification and obtain near-optimal state assignment. Then, in the pre-evolution stage, EHW evolves a set of high performing circuits and uses data mining method to find frequently evolved blocks from these circuits. The frequently evolved block would be re-used as function or terminal for evolving better circuits in the re-evolution stage. EHW has a faster convergence so that the circuit with small complexity could be evolved. Auto starting ability of circuits would also be test by the fitness function of EHW. Finally, sequence detectors, modulon counters, and ISCAS'89 circuit are used as the proof for our evolutionary design approach. Simulation results are given, and our evolutionary algorithm is shown to be better than other methods in terms of convergence time, success rate, and maximum fitness across generations.

In wireless sensor networks (WSNs), the nodes are usually powered by batteries with limited capacities, which has become one of the main challenges to the wide deployment of WSN-based applications. The overall network lifetime, which is usually determined by the node with the shortest lifetime because of the unbalanced energy consumption, becomes a major issue. Aiming at this, we consider the optimization problem of how to maximize the data amount that can be transmitted from the source node to the sink node, given an energy-limited wireless sensor network. Specifically, we first analyze the problem on the basis of the max-flow min-cut theorem to obtain the maximum relay capacity in terms of the number of successfully transmitted packets from the source node to the sink node, from the theoretical point of view. Then, we propose our algorithms to achieve the optimal solution. In this way, the expected network lifetime can be maximized. Further, we also prove the correctness of the proposed algorithms. Extensive simulations show that significant performance enhancement can be achieved by using our proposed algorithms.