Timely and accurate detection of agricultural disasters is crucial for ensuring food security and enhancing post-disaster response efficiency. This paper proposes a deployable UAV-based multimodal agricultural disaster detection framework that integrates multispectral and RGB imagery to simultaneously capture the spectral responses and spatial structural features of affected crop regions. To this end, we design an innovative stride–cross-attention mechanism, in which stride attention is utilized for efficient spatial feature extraction, while cross-attention facilitates semantic fusion between heterogeneous modalities. The experimental data were collected from representative wheat and maize fields in Inner Mongolia, using UAVs equipped with synchronized multispectral (red, green, blue, red edge, near-infrared) and high-resolution RGB sensors. Through a combination of image preprocessing, geometric correction, and various augmentation strategies (e.g., MixUp, CutMix, GridMask, RandAugment), the quality and diversity of the training samples were significantly enhanced. The model trained on the constructed dataset achieved an accuracy of 93.2%, an F1 score of 92.7%, a precision of 93.5%, and a recall of 92.4%, substantially outperforming mainstream models such as ResNet50, EfficientNet-B0, and ViT across multiple evaluation metrics. Ablation studies further validated the critical role of the stride attention and cross-attention modules in performance improvement. This study demonstrates that the integration of lightweight attention mechanisms with multimodal UAV remote sensing imagery enables efficient, accurate, and scalable agricultural disaster detection under complex field conditions.

This paper proposes a disease detection model based on the maxmin-diffusion mechanism, aimed at improving the accuracy and robustness of disease detection tasks in the agricultural field. With the development of smart agriculture, automated disease detection has become one of the key tasks driving agricultural modernization. Traditional disease detection models often suffer from significant accuracy loss and robustness issues when dealing with complex disease types and dynamically changing time-series data. To address these problems, this paper introduces the maxmin-diffusion mechanism, which dynamically adjusts attention weights to enhance the model’s focus on key disease regions while suppressing interference from irrelevant areas, significantly improving the segmentation accuracy of disease regions. Through a series of experiments, the proposed model demonstrates outstanding performance across various disease detection tasks. For bacterial spot disease detection, the model achieves a precision of 0.98, recall of 0.95, accuracy of 0.96, and mIoU of 0.96, indicating that it can efficiently and accurately identify disease regions even in complex backgrounds. Compared to traditional self-attention and CBAM mechanisms, the maxmin-diffusion mechanism shows significant advantages in fine-grained feature extraction and time-series data processing, particularly in the recognition of dynamically changing disease regions, where it exhibits higher detection accuracy and robustness. Furthermore, the model underwent lightweight optimization, enabling the proposed disease detection model to not only achieve high-precision detection but also run efficiently on resource-constrained mobile devices. This provides strong technical support for the application of smart agriculture.

Banking data centers generate heterogeneous IT operational streams from servers, network devices, edge nodes, and business terminals. Existing detectors are vulnerable to asynchronous sampling and unreliable channels, while separate root-cause and language-model modules cannot preserve an end-to-end evidence trail. This study presents a cloud–edge–end framework that couples sensor-quality-aware temporal encoding, topology- and chronology-constrained event-graph reasoning, and evidence-constrained language generation. At the edge, dynamic reliability scores suppress missing, noisy, delayed, drifting, and frozen channels. At the cloud layer, the model ranks root causes and identifies propagation relations and onset times from device topology, logs, alarms, and maintenance records. The language model then generates traceable incident interpretations from the resulting evidence package and rejects claims when support is insufficient or conflicting. Experiments use IT operational data collected from the headquarters and disaster-recovery data centers of a commercial bank between 2024 and 2025; the dataset contains no customer financial or transaction records. The proposed method achieves $95.38\%$ Accuracy, $95.06\%$ Precision, $94.71\%$ Recall, a $94.88\%$ F1-score, and a $97.41\%$ AUC for hardware anomaly detection. Root-cause localization accuracy, Top-3 Accuracy, event-relation F1-score, and temporal localization error are $91.84\%$, $97.26\%$, $90.73\%$, and $1.82$ s, respectively. The complete model attains a $94.68\%$ evidence support rate and a $4.26\%$ unsupported statement rate. These results show that reliability is propagated from sensing through system-level diagnosis to evidence-grounded interpretation.