Image deblurring aims to restore high-quality images from blurry ones. Although existing methods have achieved remarkable progress in restoration performance, most approaches still mainly rely on designing complex modules in the spatial domain, leading to the gradual saturation of performance improvements. This paper proposes a novel CNN–Transformer hybrid architecture for image deblurring, dubbed DSCTNet, by analyzing the differences between blurry and sharp images in the fourier and wavelet domains. Specifically, DSCTNet comprises two stages. The first stage is the local restoration stage, which progressively removes different levels of blur through multiple iterations. During this process, the multi-scale features generated at each iteration are fused into enhanced prior features, which are then forwarded along with the restored outputs to the second stage. The second stage is the global restoration stage, which employs a dual-branch design in the spatial and wavelet domains and incorporates a Transformer-based self-attention mechanism to capture long-range dependencies. By leveraging the prior features from the first stage, this stage effectively compensates for the Transformer’s limitations in local modeling, further improving structural restoration and detail reconstruction based on the preceding stage’s results. Additionally, a dual-domain feature enhancement module is designed to extract features at different frequencies in the fourier domain and features with multiple receptive fields in the spatial domain, subsequently fusing these features to enhance representational richness and diversity. Extensive experiments demonstrate that the proposed DSCTNet achieves superior image restoration performance compared with existing state-of-the-art methods.
Single-event transients (SETs) of the power supply circuit are investigated by conducting heavy-ion tests. The most sensitive device in the circuit whose SET response dominates the circuit functional degradation is identified. Then the impact of load currents and load types on the SET cross section of the sensitive device is analyzed. A combination of circuit characteristics tests and SPICE simulations is used to illustrate the mechanism of the influence on device SETs response under different load conditions. The research results may provide valuable assistance for power supply circuit design and device application.
To address the challenges of low computational efficiency in existing convolutional network models and their application to rotating machinery fault detection under non-stationary operating conditions, an integrated bearing fault detection method is proposed that synthesizes time-domain, frequency-domain, and time-frequency domain features through adversarial graph convolutional learning. The core innovation lies in combining multi-domain features with adversarial GCN learning to enhance the model’s recognition and training effectiveness as well as the data’s representation capability. This method extracts time-domain and frequency-domain features of bearing signals using feature fusion technology, converts one-dimensional vibration signals into two-dimensional images via Gramian angular fields, kurtosis spectrum analysis, and chirplet transform, and integrates these images into new feature images to achieve data augmentation, thereby better participating in the model’s training tasks. Confusion matrices, T-SNE, and U-MAP are used for visual classification. Experimental results show that under three different rotational speed conditions, the average accuracy of this method in fan bearing fault diagnosis can reach 98.4
Abstract Distributed Denial of Service attacks (DDoS) are a common and influential network malicious behavior. The timely and accurate detection of Distributed Denial-of-Service (DDoS) attacks constitutes a critically significant research imperative in cyber security. Most current research focuses on classification based on statistical characteristics of network traffic, but less considers the significance of packet payload feature for DDoS attack identification. This paper proposes an adaptive DDoS detection framework integrating machine learning with payload feature engineering. The methodology comprises three phases: 1) constructing a heterogeneous task classification system based on packet metadata analysis, 2) establishing a hierarchical keyword lexicon through payload decomposition and feature pattern mining, followed by feature vector transformation via numerical encoding, and 3) implementing supervised learning algorithms for discriminative model training and feature validity verification. This multilevel feature engineering approach demonstrates enhanced adaptability in DDoS attack pattern recognition compared to conventional detection paradigms. Test results on the public datasets CIC-DDoS-2019, ISCX-SlowDoS-2016 and DoS/DDoS-MQTT-IoT show that the average detection rate of the method in this paper reaches 98.9% for attack behaviors, and the false alarm rate is only 0.1%.
This study systematically investigates the influence of the Y2O3/Ta2O5 doping on the phase structure, microstructure, and dielectric properties of Ba(Zn0.70Co0.25)1/3(Nb0.95W0.075)2/3Oδ ceramics for 5G/6G applications. The results demonstrate that Ta5+ substitutes for Nb5, while Y3+ replaces Co2+/Co3+, with the accompanying vacancies and defects enhancing ion migration and diffusion. This process promotes densification and lowers the optimal sintering temperature. Optimal doping concentrations of 0.4 wt
