This study proposes an improved YOLOv11 model to address the limitations of traditional tomato recognition algorithms in complex agricultural environments, such as lighting changes, occlusion, scale variations, and complex backgrounds. These factors often hinder accurate feature extraction, leading to recognition errors and reduced computational efficiency. To overcome these challenges, the model integrates several architectural enhancements. First, the UniRepLKNet block replaces the C3k2 module in the standard network, improving computational efficiency, expanding the receptive field, and enhancing multi-scale target recognition. Second, the RFCBAMConv module in the neck integrates channel and spatial attention mechanisms, boosting small-object detection and robustness under varying lighting conditions. Finally, the TADDH module optimizes the detection head by balancing classification and regression tasks through task alignment strategies, further improving detection accuracy across different target scales. Ablation experiments confirm the contribution of each module to overall performance improvement. Our experimental results demonstrate that the proposed model exhibits enhanced stability under special conditions, such as similar backgrounds, lighting variations, and object occlusion, while significantly improving both accuracy and computational efficiency. The model achieves an accuracy of 85.4%, recall of 80.3%, and $mAP@50$ of 87.3%. Compared to the baseline YOLOv11, the improved model increases $mAP@50$ by 2.2% while reducing parameters to 2.16 M, making it well-suited for real-time applications in resource-constrained environments. This study provides an efficient and practical solution for intelligent agriculture, enhancing real-time tomato detection and laying a solid foundation for future crop monitoring systems.

With the continuous advancement of information technology in the agricultural field, a large amount of unstructured agricultural textual information has been generated. This information is crucial for supporting the development of smart agriculture, making the application of named entity recognition in the agricultural field more urgent. In order to enhance the accuracy of agricultural entity recognition, this study utilizes the pre-trained BERT-wwm model for word embedding into the text. Additionally, a channel attention mechanism (CA) is introduced in the BILSTM-CRF downstream feature extraction network to comprehensively capture the contextual features of the text. Experimental results demonstrate that the proposed method significantly improves the performance of named entity recognition, with increased accuracy, recall, and F1 value. The successful implementation of this method provides reliable support for downstream tasks such as agricultural knowledge graph construction and question and answer systems and establishes a foundation for better understanding and utilization of agricultural textual information.

Current research on tea primarily focuses on foundational studies of phenotypic characteristics, with insufficient exploration of the relationship between tea varieties and suitable populations. To address this issue, this paper proposes a link prediction model based on Graph Neural Networks (GNN) and tensor decomposition, named GNN-FTuckER, designed to predict the “tea suitability” relationships within the tea knowledge graph. This model integrates the SE-GNN structural encoder with an improved TuckER model decoder. The SE-GNN encoder enhances the modeling capability of the global graph structure by explicitly modeling relations, entities, and triples, thereby obtaining embedding vectors through aggregation, updating, and iterative operations. The improved TuckER model enhances the capture of complex semantics between entities and relations by introducing nonlinear activation functions. To support our research, we constructed a tea dataset, TeaPle. In comparative experiments, GNN-FTuckER achieved superior performance on both public datasets (WN18RR, FB15k-237) and the TeaPle dataset. Ablation studies indicate that the model improved H@10 by 4.3% on the WN18RR dataset and by 1.5% on the FB15k-237 dataset, with a 1.3% increase in MRR. In the TeaPle dataset, H@3 improved by 4.7% and H@10 increased by 3.1%. This research provides significant insights for further exploring the potential of tea varieties and evaluating the health benefits of tea consumption.

This study proposes an improved link prediction model for predicting the “suitable for people” relationship within the knowledge graph of tea. The relationships between various types of tea and suitable target groups have yet to be fully explored, and the existing InteractE model still does not adequately capture a portion of the complex information around the interactions between entities and relationships. In this study, we integrate SENet into the feature layer of the InteractE model to enhance the capturing of helpful information in the feature channels. Additionally, the GCN layer is employed as the encoder, and the SENet-integrated InteractE model is used as the decoder to further capture the neighbour node information in the knowledge graph. Furthermore, our proposed improved model demonstrates significant improvements compared to several standard models, including the original model from public datasets (WN18RR, Kinship). Finally, we construct a tea dataset comprising 6698 records, including 330 types of tea and 29 relationship types. We predict the “suitable for people” relationship in the tea dataset through transfer learning. When comparing our model with the original model, we observed an improvement of 1.4% in H@10 for the WN18RR dataset, a 7.6% improvement in H@1 for the Kinship dataset, and a 5.2% improvement in MRR. Regarding the tea dataset, we achieved a 4.1% increase in H@3 and a 2.5% increase in H@10. This study will help to fully exploit the value potential of tea varieties and provide a reference for studies assessing healthy tea drinking.

Fine-grained maturity assessment of Sichuan pepper (Zanthoxylum bungeanum cv. Hanyuan) is challenging: CNNs are data-hungry and opaque, and vision-language models (VLMs) achieve only chance-level accuracy ($16.67\%$) on this six-way task when used end-to-end. We present KG-RAG, a retrieval-augmented classification framework that combines CNN visual features with a 25,881-triplet knowledge graph built from VLM-extracted structured attributes. KG-RAG introduces three key components: (i) a hierarchical attribute consistency score (HACS) that generalizes Jaccard re-ranking via mutual-information weighting and family-level regularization; (ii) a confidence-guided retrieval gate for adaptive parametric/non-parametric fusion, with calibration as a secondary benefit; and (iii) a two-stage VLM curriculum that repurposes a VLM—inaccurate as an end-to-end classifier but reliable as an attribute extractor—into a structured knowledge provider. On 2114 expert-annotated images (inter-annotator Cohen’s $\kappa =0.89$), $10\times 5$-fold cross-validation shows consistent gains across four backbones (+1.25 to +4.65 percentage points), with the largest gain in the low-data regime (+14.89 pp at 10% training data). Statistical significance is assessed via paired t-tests with repeated-CV caveats.

Tea pest and disease named entity recognition (NER) faces challenges resulting from dense domain terminology, multi-granularity entity structures, and long-distance semantic dependencies. This paper proposes E-BERT-wwm-BiGRU-RAT-CRF, integrating whole-word masking E-BERT with three innovations—a trainable relative position bias matrix, a cross-layer hierarchical mask matrix, and a heterogeneous multi-head attention mechanism—followed by bidirectional gated recurrent units (BiGRU), residual attention (RAT), and conditional random fields (CRF). On a self-constructed tea pest and disease corpus of over 300,000 characters across seven entity categories, the model achieves 93.67% precision, 93.07% recall, and 93.37% F1-score, outperforming the baseline by 2.73 percentage points in F1-score. Ablation experiments confirm the contribution of each module.