As Internet of Things (IoT) technology sees extensive adoption in smart agriculture, smart healthcare, and smart cities, emerging systems are increasingly confronted with complex and dynamic security threats. Intrusion Detection Systems (IDS), a key technology in network security, effectively enhance IoT system safety by detecting and monitoring anomalous activities. Nevertheless, IDS relying on traditional Machine Learning (ML) technologies demonstrate limited efficacy in classifying malicious traffic. In recent years, approaches that convert network security data into image sets and leverage Deep Transfer Learning (DTL) for classification have gained rapid popularity. While these methods substantially improve detection accuracy, they also lead to increased time and resource consumption during training. To balance high detection accuracy with reduced time consumption, this study introduces an efficient intrusion detection approach based on the Vision Transformer (ViT), utilizing its powerful feature extraction capabilities to enhance performance. The proposed High-performance ViT Intrusion Detection System (HiViT-IDS) begins by transforming one-dimensional network traffic data into RGB images and leverages the ViT model’s exceptional representational power for efficient classification. Experimental results on the ToN-IoT and Edge-IIoTset datasets reveal classification accuracies of 99.70% and 100%, respectively. In comparison to existing mainstream DTL approaches, the proposed model achieves considerable reductions in training time while sustaining high performance. The findings suggest that the HiViT-IDS offers superior potential and a competitive edge in adapting to complex and dynamic network environments.

Mine ecological restoration is a critical process for promoting the sustainable development of resource-dependent regions, yet existing monitoring methods remain limited in accuracy and adaptability. To address challenges such as small-object recognition, insufficient multi-scale feature fusion, and blurred boundaries in UAV-based remote sensing imagery, this paper proposes an enhanced semantic segmentation model based on Segformer. Specifically, a multi-scale feature-enhanced feature pyramid network (MSFE-FPN) is introduced between the encoder and decoder to strengthen cross-level feature interaction. Additionally, a selective feature aggregation pyramid pooling module (SFA-PPM) is integrated into the deepest feature layer to improve global semantic perception, while an efficient local attention (ELA) module is embedded into lateral connections to enhance sensitivity to edge structures and small-scale targets. A high-resolution UAV image dataset, named the HUNAN Mine UAV Dataset (HNMUD), is constructed to evaluate model performance, and further validation is conducted on the public Aeroscapes dataset. Experimental results demonstrated that the proposed method exhibited strong performance in terms of segmentation accuracy and generalization ability, effectively supporting the image analysis needs of mine restoration scenes.

In view of the current problems of complex models and insufficient data processing in ultra-short-term prediction of photovoltaic power generation, this paper proposes a photovoltaic power ultra-short-term prediction model named HPO-KNN-SRU, based on a Simple Recurrent Unit (SRU), K-Nearest Neighbors (KNN), and Hunter–Prey Optimization (HPO). Firstly, the sliding time window is determined by using the autocorrelation function (ACF), partial correlation function (PACF), and model training. The Pearson correlation coefficient method is used to filter the principal meteorological factors that affect photovoltaic power. Then, the K-Nearest Neighbors (KNN) algorithm is utilized for effective outlier detection and processing to ensure the quality of input data for the prediction model, and the Hunter–Prey Optimization (HPO) algorithm is applied to optimize the parameters of the KNN algorithm. Finally, the efficient Simple Recurrent Unit (SRU) model is used for training and prediction, with the Hunter–Prey Optimization (HPO) algorithm applied to optimize the parameters of the SRU model. Simulation experiments and extensive ablation studies using photovoltaic data from the Desert Knowledge Australia Solar Centre (DKASC) in Alice Springs, Australia, validate the effectiveness of the integrated model, the KNN outlier handling, and the HPO algorithm. Compared to the Support Vector Regression (SVR), Long Short-Term Memory (LSTM), Temporal Convolutional Network (TCN), and Simple Recurrent Unit (SRU) models, this model exhibits an average reduction of 19.63% in Mean Square Error (RMSE), 27.54% in Mean Absolute Error (MAE), and an average increase of 1.96% in coefficient of determination ($R^2$) values.

Pruning and compression of models are practical approaches for deploying and applying deep convolutional neural networks in scenarios with limited memory and computational resources. To mitigate the impact of pruning on model accuracy and enhance the stability of pruning (defined as the negligible drop in test accuracy immediately following pruning), an algorithm for reward-penalty decoupling is introduced in this study to achieve automated sparse training and channel pruning. During sparse training, the influence of unimportant channels is automatically identified and reduced,thereby preserving the ability of the important channels for feature recognition. First, by utilizing the gradient information learned through network backpropagation, the feature scaling factors of the batch normalization layers are combined with the gradient to determine the importance threshold for the network channels. Subsequently, a two-stage sparse training algorithm is proposed based on the reward-penalty decoupling strategy, applying different loss function strategies to the feature scaling factors of "important" and "unimportant" channels during decoupled sparse training. This approach has been experimentally validated across various tasks, baselines, and datasets, demonstrating its superiority over the previous state-of-the-art methods. The results indicate that the effect of pruning on model accuracy is significantly alleviated by the proposed method, and pruned models require only limited fine-tuning to achieve excellent performance.

Existing diffusion models outperform generative models like Generative Adversarial Networks in image synthesis and editing. However, they struggle with high-precision image editing while preserving image details and the accuracy of editing instructions. To address these challenges, we propose a dual attention control method to achieve high-precision image editing. Our approach includes two key attention control modules: (1) cross-attention control module, which combines the cross-attention maps of the original and edited images through weighted parameters, ensures that the synthesized edited image retains the structure of the input image. (2) Self-attention control module, which varies based on the editing task, applied at “coarse” and “fine” layers, since the coarse layers help maintain input image details and the fine layers are better suited for style transformations. Experimental evaluations have demonstrated that our approach achieves excellent results in detail preservation, content consistency, visual realism, and semantic understanding, making it especially suitable for tasks requiring high-precision editing. Specifically, compared to the editing outcomes under no control conditions, the introduction of dual visual attention control has led to an increase of 6.19% in CLIP scores, a reduction of 29.3% in LPIPS, and a decrease of 24.7% in FID. These significant improvements not only validate the effectiveness of the dual attention control but also attest to the method’s substantial flexibility and adaptability across different scenarios. Notably, our approach is a zero-shot solution, requiring no user optimization or fine-tuning, facilitating real-world applications.