Detailed spatial distribution of soil organic matter (SOM) in arable land is essential for agricultural management and decision making. Based on digital soil mapping (DSM) theory, much attention has been focused on the selection of environmental covariates. However, the importance of human activity factors in SOM prediction has not received enough attention, especially in arable soil. Moreover, due to the insufficient amount of soil sampling data used to train and validate the DSM model, the prediction results may be questionable, and some even contradictory. This paper explores the effectiveness of the human footprint, amount of fertilizer application, agronomic management level, crop planting type, and irrigation guarantee degree in SOM mapping of arable land in Heilongjiang Province. The results show that the model only including environmental covariates accounts for 41% of the variation in SOM distribution. The model combining the five human activity factors increases the SOM spatial prediction by 39% in terms of R² (coefficient of determination), 12% in terms of RMSE (root mean square error), 15% in terms of MAE (mean absolute error), and 11% in terms of LCCC (Lin’s concordance correlation coefficient), showing better prediction accuracy and performance. This indicates that human activity factors play a crucial role in determining SOM distribution in arable land. In the SOM prediction, soil moisture is the most important environmental covariate, and the amount of fertilizer application with a relative importance of 11.36% (ranking 3rd) is the most important human activity factor, higher than the annual average precipitation and elevation. From a spatial point of view, the Sanjiang Plain is a difficult area for prediction.

Accurate organ-level segmentation is essential for achieving high-throughput, non-destructive, and automated plant phenotyping. To address the challenge of intelligent acquisition of phenotypic parameters in tomato plants, we propose TSINet, an end-to-end dual-task segmentation network designed for effective and precise semantic labeling and instance recognition of tomato point clouds, based on the Pheno4D dataset. TSINet adopts an encoder–decoder architecture, where a shared encoder incorporates four Geometry-Aware Adaptive Feature Extraction Blocks (GAFEBs) to effectively capture local structures and geometric relationships in raw point clouds. Two parallel decoder branches are employed to independently decode shared high-level features for the respective segmentation tasks. Additionally, a Dual Attention-Based Feature Enhancement Module (DAFEM) is introduced to further enrich feature representations. The experimental results demonstrate that TSINet achieves superior performance in both semantic and instance segmentation, particularly excelling in challenging categories such as stems and large-scale instances. Specifically, TSINet achieves 97.00% mean precision, 96.17% recall, 96.57% F1-score, and 93.43% IoU in semantic segmentation and 81.54% mPrec, 81.69% mRec, 81.60% mCov, and 86.40% mWCov in instance segmentation. Compared with state-of-the-art methods, TSINet achieves balanced improvements across all metrics, significantly reducing false positives and false negatives while enhancing spatial completeness and segmentation accuracy. Furthermore, we conducted ablation studies and generalization tests to systematically validate the effectiveness of each TSINet component and the overall robustness of the model. This study provides an effective technological approach for high-throughput automated phenotyping of tomato plants, contributing to the advancement of intelligent agricultural management.

Accurately and effectively detecting the growth position and contour size of apple fruits is crucial for achieving intelligent picking and yield predictions. Thus, an effective fruit edge detection algorithm is necessary. In this study, a fusion edge detection model (RED) based on a convolutional neural network and rough sets was proposed. The Faster-RCNN was used to segment multiple apple images into a single apple image for edge detection, greatly reducing the surrounding noise of the target. Moreover, the K-means clustering algorithm was used to segment the target of a single apple image for further noise reduction. Considering the influence of illumination, complex backgrounds and dense occlusions, rough set was applied to obtain the edge image of the target for the upper and lower approximation images, and the results were compared with those of relevant algorithms in this field. The experimental results showed that the RED model in this paper had high accuracy and robustness, and its detection accuracy and stability were significantly improved compared to those of traditional operators, especially under the influence of illumination and complex backgrounds. The RED model is expected to provide a promising basis for intelligent fruit picking and yield prediction.

With the explosive growth of information on the internet, personalized recommendation technology has become an important tool for helping users efficiently acquire information. However, existing spreading-based recommendation algorithms only consider user choices and fail to fully leverage the potential relationships between users and items. Additionally, the incomplete utilization of user and item information limits their application potential and applicable scenarios, resulting in suboptimal recommendation performance in practical applications. To address this issue, we propose a Light Graph Convolutional Recommendation Algorithm Based on Hybrid Spreading (LGCNHS). This algorithm first optimizes the embeddings of users and items using their respective feature matrix, then learns the latent embedding representations of users and items through a lightweight graph convolutional network. Finally, the latent embedding representations are incorporated as key parameters into the hybrid spreading recommendation algorithm to generate recommendations. Comparative experiments on two publicly available datasets, MovieLens and Douban, demonstrate that LGCNHS achieves improved accuracy and diversity in recommendations compared to related methods. The algorithm code is available on github.

The precise estimation of protein inter-residue distances is essential for high-accuracy protein structure modeling. Currently, the prediction methods are predominantly based on MSA-derived coevolutionary features or language model-based sequence features. To effectively leverage the strengths of both methods, this study developed MF-ProtDisMap (Multi-Feature Protein Distance Map), an integrative framework that effectively combines both feature types to achieve superior real-valued distance prediction. Briefly, MSA Transformer is employed to extract the coevolutionary features from protein multiple sequence alignments, whereas ESM2 is used to capture long-range interactions and sequence-level features. To reduce the computational cost while maximumly represent fused feature information, we implement group pooling for feature dimensionality reduction and introduce Diff-former—a novel module combining a diffusion model with a triangular attention mechanism to enhance representation learning. MF-ProtDisMap achieved a MAE of 2.20 Å and an RMSE of 3.40 Å in the protein real-valued distance prediction task. The predicted distances can be converted into contact results, achieving ROC and PR values of 84.56% and 81.01%, respectively. These results demonstrate that MF-ProtDisMap outperforms state-of-the-art real-valued protein distance methods.