(1) With the development of artificial intelligence, people expect to use modern information technology to solve the critical problems encountered in agriculture. How to identify sunflower diseases as early and quickly as possible and take corresponding measures has become a key issue for increasing crop production and farmers’ income. Sunflowers, as an important oil crop, are vulnerable to infections by various diseases, such as downy mildew, leaf scar, gray mold, etc. (2) In order to select a better lightweight model that can be embedded into mobile devices or embedded devices for sunflower disease detection, we compared five lightweight deep learning models in this study, including SqueezeNet, ShuffleNetV2, MnasNet-A1, MobileNetV3-Small, and EfficientNetV2-Small. The dataset used to train and test the models included 1892 images. These images were divided into four categories, namely, downy mildew, gray mold, leaf scar, and fresh leaves. (3) By evaluating the accuracy, precision, recall, and F1 score of each model, we found that EfficeintNetV2-Small exhibited the highest performance with an accuracy of 90.19%. Whereas the other models, SqueezeNet, ShuffleNetV2, MnasNet-A1, and MobileNetV3-Small, achieved accuracies of 84.08%, 79.31%, 88.59%, and 84.08%, respectively. To address the problem of poor generalization ability of models caused by small datasets, we adopted the transfer learning technique. After doing that, the recognition accuracies of the five models, SqueezeNet, ShuffleNetV2, MnasNet-A1, MobileNetV3-Small, and EfficeintNetV2-Small, reached 96.02%, 95.23%, 94.96%, 96.92%, and 99.20%, respectively. The accuracies of these five models were improved by 14.2%, 20%, 7.2%, 15.2%, and 10%. Based on the comparative results, we found EfficeintNetV2-Small was an optimal choice for sunflower disease identification due to its high detection accuracy.

Air quality level has a complex nonlinear relationship with air pollutant and meteorological conditions, including multiple factors, overlapping information, and difficulty solving equations. In order to identify significant factors, remove correlations, reduce data dimensionality, and simplify the model structure, a BP neural network model for air quality level prediction optimized by stepwise discriminant analysis (STEPDISC) and principal component analysis (PCA) is proposed with 12 factors of historical daily meteorology and air pollutants in Bayannur city as samples. The results showed that, at the significance level of 0.01, the STEPDISC method retained 9 significant impact factors. The PCA method made an orthogonal linear combination of the 9 factors to form the principal components, and the contribution of the top 5 principal components were 37.6%, 19.2%, 15.3%, 8.8%, and 7.7%. At a contribution threshold of 0.85, the top 5 principal component scores were used as input nodes to construct the STEPDISC-PCA-BP model, which had a prediction accuracy of 85.5%.Compared with the PCA-BP and BP model, which had a prediction accuracy of 61.8% and 56.7%, respectively, the STEPDISC-PCA-BP model has a higher prediction accuracy, shorter time, and lower complexity of structure and data dimensionality, and can provide the necessary technical support for the local air quality improvement.

Aiming at the differences in spatial spectral attributes between forested land and cultivated land on remote sensing images, and the deficiencies of existing remote sensing change detection methods that are difficult to capture fine edge structures and distinguish pseudo changes, this paper introduces an Edge Enhanced Hierarchical Dual Gated Change Detection (EHDCD) model for forested land and cultivated land, aiming to meet the demand for representing the complex features of these two land types. The model designs an Edge Enhanced Channel Attention Module (EECA) to strengthen the edge recognition ability and suppress the noise interference; proposes a High-Low Level Dynamic Adaptation Strategy (HiLo) to realize the balanced expression of detail information and semantic features; and constructs a Dual Gated Feature Compensation Module (DGFM) to effectively reduce the misdetection rate of change detection. Experiments show that the F1 scores of the model on the self-constructed forest and agricultural dataset FC-CD and public datasets CLCD and SYSU-CD reach 89.06%, 83.37%, and 85.06%, respectively, which can more accurately support the dynamic monitoring applications of forest land and cropland.

Introduction: Accurate identification of crop diseases is crucial for ensuring crop quality and yield. However, existing deep learning models for crop disease identification lack robustness in complex field environments and suffer from large model parameter sizes, which makes them difficult to deploy on resource-constrained devices. This gap between laboratory models and practical field applications necessitates the development of a lightweight and robust identification model. Methods: To address these challenges, this paper proposes a lightweight YOLO-CGA model for sunflower disease identification and deploys it on a Raspberry Pi for field application. The model incorporates three key improvements based on YOLOv8n-cls: (1) A CBAM_ADown module is designed, which integrates attention mechanisms with asymmetric downsampling to enhance feature extraction and noise suppression in complex image backgrounds; (2) The C2f module of YOLOv8n-cls is replaced with the C3Ghost module, which utilizes ghost convolution to reduce parameter count while preserving fine-grained features; (3) An AFC_SPPF module is constructed, which aggregates multi-scale disease features through a multi-branch adaptive fusion structure to improve recognition performance for diverse lesions. Results: Experimental results on three major datasets show that the proposed YOLO-CGA model achieves high identification accuracy: 98.48% on the BARI-Sunflower dataset, 98.32% on the Cotton Disease Dataset, and 91.11% on the FGVC8 dataset. Meanwhile, the model maintains a lightweight property with only 0.92M parameters, which is significantly fewer than that of other comparative models. Discussion: The deployment of the YOLO-CGA model on the Raspberry Pi end device effectively bridges the gap between laboratory models and field applications, fulfilling the demand for real-time and on-site crop disease identification. The integration of attention mechanisms, ghost convolution, and multi-scale feature fusion enables the model to balance accuracy, robustness, and lightweight performance, making it suitable for resource-limited field scenarios.

To address the challenge of fragmented, unstructured knowledge in the cultural relic domain, where existing entity and relation extraction models suffer from boundary confusion and feature degradation for long entities and overlapping triples, this paper proposes a subject-guided two-stage joint entity and relation extraction model tailored to cultural relic texts, and constructs a Cultural Relic Knowledge Graph System. Building on CasRel’s cascaded labeling framework, we design a Multi-Head Self-Attention Decoder Enhanced by Relative Position Encoding (MHSA-RPE) to explicitly model inter-entity positional relations and alleviate boundary confusion. We further propose a Boundary–Global Dual-Branch Subject Fusion Module (BGDSFM) to encode local boundary and global contextual features in parallel, alleviating feature degradation from simple average pooling. Experiments on DuIE2.0 and our self-built Palace Museum Cultural Relic Entity–Relation Dataset (PM-CRER) show that the proposed model achieves F1-scores of 79.4% and 75.9% respectively. It outperforms mainstream baselines, surpassing its prototype CasRel by 3.6 percentage points on PM-CRER and the latest cascaded state-of-the-art CECRel by 2.6 percentage points on DuIE2.0. Based on this model, a Chinese Cultural Relic Knowledge Graph System supporting the multimodal display of cultural relic images is constructed, providing technical references for the digital protection, dissemination and utilization of cultural relic knowledge.