• 学术搜索
  • 科研智能体
    • Research Labs
    • AI 阅读
    • AI 文库
    • 深度研究
    • 学者亮点
  • 学术资源
    • AI2000
    • 期刊/会议
    • 学者库
    • 学术API
      MCP&Skill
    • 溯源树
    • 数据集
  • 知识沉淀
    • 学术空间
订阅小程序
旧版功能
aminer vip
开通会员低至0.73元/天
一次搞定AI科研
立即登录
  • English
  • 联系方式
    AGRIENGINEERING

    AGRIENGINEERING

    JournalISSN 2624-7402eISSN 2624-7402中科院 农林科学 3区

    年发文量

    研究主题

    论文(1529)

    排序
    1Explainable AI for Smart Characterization of Dairy-Lagoon Wastewater Using Farm-Held-Out Validation
    Abdullah Al Rawi,Prachi Pandey, Behrooz Pakzadeh,Pramod Pandey

    Commercial dairy farms are essential to milk production, food security, and rural economies, but they also generate large volumes of manure and wastewater requiring reliable characterization. Measuring solids, organic matter, and dissolved constituents is important for pollution control, nutrient management, and resource recovery. Artificial intelligence may reduce analytical burden, although its usefulness depends on demonstrating that predictions generalize across farms and that model explanations are scientifically credible. This study evaluated 205 lagoon-manure samples collected between May and August 2021, within a four-year project, from 18 commercial dairies in California’s Central Valley. Sample availability varied among measurements and modeling tasks because of missing observations. Gravimetric and chemical measurements were analyzed using polynomial ridge regression, random forest, and XGBoost with farm-held-out cross-validation; validated models were interpreted using SHAP and LIME. All 205 samples had complete gravimetric records, and chemical measurements were available for 155 samples. Samples averaged 96.02% moisture, 3.98% total solids, and 3.06% organic matter. Organic matter was predicted from the combined gravimetric measurements (R2 = 0.975, RMSE = 0.97 percentage points, MAE = 0.39, and RPD = 6.32). Total solids alone produced similar pooled performance (R2 = 0.979; RPD = 6.86), and a linear regression on total solids performed comparably (R2 = 0.980). However, prediction errors were strongly concentration dependent. For the linear benchmark, mean absolute errors represented approximately 56%, 21%, and 11% of mean organic-matter concentrations in samples containing <1%, 1–10%, and >10% total solids, respectively. These findings indicate that strong pooled performance does not establish reliable screening accuracy for highly dilute samples. In contrast, nitrate, sodium, salinity, calcium, and pH could not be predicted reliably (R2 = −0.431 to 0.091; RPD = 0.84–1.05). SHAP identified moisture–solids as the dominant contributor to organic-matter predictions, whereas LIME supported sample-level interpretation. These findings apply to flush-based dairies in California’s Central Valley, and this approach can support low-cost screening of dairy-lagoon manure for organic matter while directing chemically governed properties to laboratory measurement.

    2026
    引用
    AI阅读
    加入学术空间
    2Comparative Analysis of Explainable Spectroscopic-Fusion Models for Avocado Leaf Nitrogen Prediction
    Md. Abu Johab, Md Mahmudul Hasan,Md. Niaz Imtiaz,Most. Mira Khatun,Marcus Vinicius da Silva Ferreira,Md Wadud Ahmed,Md. Toukir Ahmed

    Precise nutrient management in avocado orchards requires accurate and interpretable predictions of leaf nitrogen content. In this study, explainable machine-learning models integrating spectral reflectance (350–2500 nm) and multispectral spatial features were comparatively evaluated for non-destructive estimation of avocado leaf nitrogen content. Partial Least Squares Regression (PLSR) and one-dimensional Convolutional Neural Network (1D-CNN) models were first developed using only spectral reflectance data as a baseline. To incorporate spatial–spectral information, multispectral images were arranged into cubes and processed with a 3D-CNN to extract deep features. These features were subsequently fused with the original spectral data to form a combined dataset. Both PLSR and 1D-CNN models were retrained using the fused inputs. The CNN hyperparameters were fine-tuned using Bayesian optimization, achieving a prediction R2 of 0.73 and an RMSE of 0.92 mg/g, which was comparable to the conventional PLSR model. The fusion approach showed numerically better performance than models based solely on spectra, suggesting spatial context may provide complementary information relevant to nitrogen estimation. Explainable artificial intelligence techniques, including SHapley Additive exPlanations and Integrated Gradients, identified 745 nm (red-edge), 994 nm (near-infrared), and 2459 nm (shortwave infrared), together with deep spatial features extracted from multispectral image cubes, as the most important predictors of nitrogen content in avocado leaves.

    2026
    引用
    AI阅读
    加入学术空间
    3A Physics-Informed Multimodal Pipeline Coupling Convolutional Quality Classification, Q10 Spoilage Kinetics, and a Lightweight Language Model for Smartphone-Based Fruit Shelf-Life Estimation
    Halleluyah Oluwatobi Aworinde, Abidemi Emmanuel Adeniyi, Oluwasubomi Oluwatoforiti Adenekan, Oluwarotimi Randle

    The waste left over between the production and consumption of fruits and vegetables ranges from 40% to 50%, and a large share of domestic wastage results from people’s inability to assess the shelf life of the produce currently available to them. Machine-vision systems capable of non-destructive food classification have seen rapid development, but they mostly result in a simple label of the product’s freshness, neglecting the storage environment that defines the rate of spoilage and leaving no rationale understandable to a layman. In this paper, the study describes, implements, and characterizes an end-to-end multimodal pipeline that takes a smartphone photograph of a fruit, along with its storage conditions and lightweight additional information supplied by the user, and predicts the number of days remaining to preserve it. The four-class quality classifier (Unripe, Ripe, Overripe, Rotten) was trained using a two-stage transfer learning procedure and a pre-trained ImageNet EfficientNetB3 architecture, processing 7440 curated images of apples, bananas, and oranges from five publicly available datasets. The training corpus size was (7440 images; 6000/960/480 train/val/test split), while the chance-baseline comparison was (25% under four-way balance, i.e., a 3.7-fold improvement). The storage conditions were derived from the freely available weather service, without dedicated hardware, and modeled using a storage mode to reflect actual indoor conditions. The kinetics of postharvest food preservation were incorporated into the model not as a post hoc correction but as part of the engineering features: a Q10-adjusted total shelf life and an elapsed-life fraction, computed with reference storage data. A gradient-boosted regressor accepts the concatenated categorical, environmental, and physics-derived feature vectors and produces the prediction of the days left. The quality propagation between the photographs occurs analytically according to the decay law, not via recurrent neural network classification. An 8B-parameter instruction-tuned language model, complemented by a rule-based deterministic fallback based on reference composition tables, converts numeric predictions into an explanation and storage recommendations. It was implemented using FastAPI and as a cross-platform mobile application. The classifier achieves an accuracy of 92.92% and 0.929 macro-F1 score on a balanced test set (480 images), where the per-class F1 score varies between 0.877 (Overripe) and 0.979 (Unripe), while the remainder of the error falls almost entirely in the vicinity of the Overripe/Rotten classification boundary. The gradient-boosted surrogate model approximates the analytical kinetic model with a mean absolute error of 1.8 days, RMSE of 2.4 days, and R2 = 0.82. Combining visual quality estimation with an explicit kinetic model, rather than training a network to learn the physics of deterioration from images, yields an actionable, interpretable shelf-life prediction for stored products that requires only mobile phone camera sensor input. The study reports the regression metrics explicitly as surrogate-approximation fidelity to an analytical kinetic model, rather than as predictive validation accuracy for stored produce.

    2026
    引用
    AI阅读
    加入学术空间
    4Moisture-Dependent Airflow Resistance and Predictive Modeling of Bulk Camelina Seeds for Drying and Aeration Design
    Wilber Akatuhurira,Clairmont L. Clementson, Kenneth Hellevang

    Reliable airflow-resistance data are essential for designing efficient drying, aeration, and storage systems for bulk seeds. However, airflow resistance of camelina remains largely unknown despite its growing importance as an oilseed crop for biofuel production. This study evaluated the effects of airflow rate, bed depth, filling method, and moisture content on pressure drop through bulk camelina seeds and assessed established airflow-resistance models. Seeds were conditioned to 6, 8, 11, and 14% moisture content (w.b.) and tested at airflow rates of 0.02–0.50 m3 s−1 m−2 across five bed depths under loose- and dense-fill conditions. Ergun, Hukill and Ives, and Shedd models were applied to the experimental data. Results indicated that pressure drop increased significantly (p < 0.05) with airflow rate and bed depth, was greater under dense-fill conditions, and decreased with increasing moisture content. All models adequately described the experimental data (R2 = 0.981–0.989), with the Shedd model providing the best overall performance. A moisture-dependent modified Shedd model was developed to incorporate moisture effects and described the observed pressure-drop trends (R2 = 0.975; RMSE = 54.87 Pa m−1). This model was calibrated and evaluated using the same experimental dataset and has not yet been validated with independent data. The results provide crop-specific parameters for predicting airflow resistance and support fan sizing, energy-efficient aeration, and drying-system design for camelina seeds.

    2026
    引用
    AI阅读
    加入学术空间
    5Improved P2PNet with Multi-Scale Feature Fusion and Point-Relation Refinement for Image-Based Maize Kernel Counting
    Renfeng Zhang, Gaopeng Zhang, Xiaotao Fu,Yingchun Kuang

    Accurate counting of visible maize kernels from RGB images is important for maize phenotyping and yield-related trait assessment, but remains challenging because kernels are densely distributed and frequently affected by occlusion and local appearance similarity. Point-based counting provides a promising alternative to bounding-box-based detection by reducing annotation requirements while retaining object-level localization. However, the original P2PNet relies on a single prediction feature with limited direct interaction between high-resolution spatial information and high-level semantic features. To address this limitation, a bidirectional feature pyramid network (BiFPN) was incorporated into P2PNet to enable bidirectional multi-scale feature fusion among C3, C4, and C5 while retaining P4 as the prediction level. A Point Relation Refinement Module (PRM) was further investigated to examine the effects of candidate-point relational modeling on initial predictions through multi-head self-attention and residual refinement. Experiments were conducted on a maize kernel dataset containing 452 RGB images and 92,314 point annotations. Across three random seeds, the combined BiFPN+PRM model achieved the lowest mean absolute error (MAE) of 26.33±1.78 and mean relative error (MRE) of 10.50±0.48% among the four main P2PNet ablation configurations, compared with 27.70±1.92 and 11.54±0.83% for the original P2PNet. BiFPN alone did not consistently improve counting accuracy but reduced computational complexity from 117.474 to 105.159 GFLOPs and increased inference throughput from 279.98 to 288.90 FPS. PRM exhibited sample-dependent behavior and substantial computational overhead. However, top-512 candidate selection reduced inference latency by 74.44% compared with global attention under the actual-resolution evaluation protocol, with slightly lower mean MAE and RMSE but a higher mean MRE (10.50% to 11.01%). These findings demonstrate the potential of the proposed framework for non-destructive image-based kernel quantification in maize phenotyping.

    2026
    引用
    AI阅读
    加入学术空间
    立即登录,查看全部 1529 篇论文

    高被引作者

    作者引用发文
    Sawal Hamid Md Ali2681
    Muhammad Salman Khan2681
    Muhammad E. H. Chowdhury2681
    Mohamed Arselene Ayari2681
    Tawsifur Rahman2681
    Mamun Bin Ibne Reaz2681
    Al-Emadi, N.2681
    Amith Khandakar2681
    Mohammad Tariqul Islam2681
    Sumbwanyambe, M.2132

    高产作者

    作者引用发文
    Paulo Eduardo Teodoro6318
    Larissa Pereira Ribeiro Teodoro6117
    Gabriel Araujo e Silva Ferraz6516
    Marcos Vinicius Silva3415
    Fabio Henrique Rojo Baio3715
    Mathew G. Pelletier9314
    John D. Wanjura9513
    Dthenifer Cordeiro Cordeiro3113
    Irenilza A. Naas9412
    Lilong Chai8412

    相关期刊

    Biosystems Engineering
    Precision Agriculture
    Journal of Agricultural Engineering
    International Journal of Agricultural and Biological Engineering
    Applied Engineering in Agriculture
    INMATEH - Agricultural Engineering
    Engenharia Agricola
    Information Processing in Agriculture
    Journal of Biosystems Engineering
    Artificial Intelligence in Agriculture