Pub Date : 2026-02-05DOI: 10.1016/j.compag.2026.111516
Xiaodong Zhang , Tingting Yu , Mohamed Farag Taha , Shenghan Zhou , Jin Zhou , Yixue Zhang , Yiqiu Zhao , Zongyao Cai , Jingjing Sun , Yuxiang Pan , Jianfeng Ping
Accurate acquisition of phenotypic characteristics in protected crops is a crucial prerequisite for intelligent control and digital breeding in greenhouses. To accurately assess the phenotypic traits of protected lettuce, a specialized in situ phenotypic detection method has been developed. The Multimodal Features and Attention Mechanism for Phenotype Detection Model (MFAMNet) was developed for protected lettuce, employing a segmented multi-source image dataset for synchronous regression testing. The results revealed that the predicted values generated by MFAMNet exhibited a strong correlation with the measured values, achieving coefficients of determination of 0.96, 0.92, 0.95, 0.94, and 0.95 for plant height, crown width, leaf area, fresh weight, and dry weight, respectively. Ablation tests demonstrated that the deep learning detection framework based on multi-modal feature fusion significantly outperformed single-feature detection models, highlighting the advantages of integrating diverse data modalities. In addition, the multi-modal feature attention mechanism (MMF) facilitates both inter-modality and intra-modality interactions by capturing the global correlations between modalities and employing dynamic sparse spatial attention. The effectiveness of MMF has been validated through comparative experiments, demonstrating its suitability for the phenotypic detection of artificially cultivated lettuce. In summary, the method proposed in this study facilitates real-time monitoring of facility crops, enabling precise control of environmental parameters in protected agriculture and optimizing resource allocation. This approach contributes to the development of a comprehensive intelligent agriculture system and establishes a foundation for unmanned farms.
{"title":"Non-destructive monitoring method for protected-lettuce yield using deep learning","authors":"Xiaodong Zhang , Tingting Yu , Mohamed Farag Taha , Shenghan Zhou , Jin Zhou , Yixue Zhang , Yiqiu Zhao , Zongyao Cai , Jingjing Sun , Yuxiang Pan , Jianfeng Ping","doi":"10.1016/j.compag.2026.111516","DOIUrl":"10.1016/j.compag.2026.111516","url":null,"abstract":"<div><div>Accurate acquisition of phenotypic characteristics in protected crops is a crucial prerequisite for intelligent control and digital breeding in greenhouses. To accurately assess the phenotypic traits of protected lettuce, a specialized in situ phenotypic detection method has been developed. The Multimodal Features and Attention Mechanism for Phenotype Detection Model (MFAMNet) was developed for protected lettuce, employing a segmented multi-source image dataset for synchronous regression testing. The results revealed that the predicted values generated by MFAMNet exhibited a strong correlation with the measured values, achieving coefficients of determination of 0.96, 0.92, 0.95, 0.94, and 0.95 for plant height, crown width, leaf area, fresh weight, and dry weight, respectively. Ablation tests demonstrated that the deep learning detection framework based on multi-modal feature fusion significantly outperformed single-feature detection models, highlighting the advantages of integrating diverse data modalities. In addition, the multi-modal feature attention mechanism (MMF) facilitates both inter-modality and intra-modality interactions by capturing the global correlations between modalities and employing dynamic sparse spatial attention. The effectiveness of MMF has been validated through comparative experiments, demonstrating its suitability for the phenotypic detection of artificially cultivated lettuce. In summary, the method proposed in this study facilitates real-time monitoring of facility crops, enabling precise control of environmental parameters in protected agriculture and optimizing resource allocation. This approach contributes to the development of a comprehensive intelligent agriculture system and establishes a foundation for unmanned farms.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111516"},"PeriodicalIF":8.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-05DOI: 10.1016/j.compag.2026.111515
Leena Priya, Pradip Kar
The nitrogen cycle in the environment is negatively impacts the biosphere through the accumulation of nitrogenous pollutants, including nitrite, nitrate, and ammonium ions, in aquatic systems. There is a real need to adapt/further develop easy and reliable real-time sensing of total nitrogen, as well as to monitor nitrogen in aqueous medium. The aim was to develop a chemiresistive sensor based on a hybrid of poly(meta-aminophenol) (PmAP), with 4-quinonimine functionalized silver nanoparticles for sensitive and selective detection of total nitrogen as nitrogenous contaminants in aqueous media. First, 4-quinonimine functionalized material was synthesized, followed by the successful preparation of a core–shell hybrid with PmAP. The prepared hybrid was confirmed to have a polymer shell measuring 20–40 nm surrounding a functionalized silver core, which has an average diameter of 23 nm and varies between 5 to 50 nm. A two-probe sensor layer was fabricated on a screen-printed carbon electrode for the efficient detection of total nitrogen as nitrite, nitrate, and ammonium ions in aqueous media. The sensor exhibited a selective sensing response toward nitrogen, with a sensitivity of 10 µA mM−1, a limit of detection (LOD) of 0.01 mM and a sensor resolution of 0.012 mM within the linearity range of 0.01–2 mM for the nitrogen in terms of those nitrogenous contaminants. The fabricated sensor shows strong applicability for real-time nitrogen monitoring in aqueous media, particularly for agricultural applications and smart agriculture systems. An average 96 ± 4 % accuracy was also verified for nitrogen sensing in agricultural river water samples.
{"title":"Detection of total nitrogen contaminants in agricultural water using a poly(m-aminophenol) and silver nanoparticle hybrid sensor","authors":"Leena Priya, Pradip Kar","doi":"10.1016/j.compag.2026.111515","DOIUrl":"10.1016/j.compag.2026.111515","url":null,"abstract":"<div><div>The nitrogen cycle in the environment is negatively impacts the biosphere through the accumulation of nitrogenous pollutants, including nitrite, nitrate, and ammonium ions, in aquatic systems. There is a real need to adapt/further develop easy and reliable real-time sensing of total nitrogen, as well as to monitor nitrogen in aqueous medium. The aim was to develop a chemiresistive sensor based on a hybrid of poly(<em>meta</em>-aminophenol) (PmAP), with 4-quinonimine functionalized silver nanoparticles for sensitive and selective detection of total nitrogen as nitrogenous contaminants in aqueous media. First, 4-quinonimine functionalized material was synthesized, followed by the successful preparation of a core–shell hybrid with PmAP. The prepared hybrid was confirmed to have a polymer shell measuring 20–40 nm surrounding a functionalized silver core, which has an average diameter of 23 nm and varies between 5 to 50 nm. A two-probe sensor layer was fabricated on a screen-printed carbon electrode for the efficient detection of total nitrogen as nitrite, nitrate, and ammonium ions in aqueous media. The sensor exhibited a selective sensing response toward nitrogen, with a sensitivity of 10 µA mM<sup>−1</sup>, a limit of detection (LOD) of 0.01 mM and a sensor resolution of 0.012 mM within the linearity range of 0.01–2 mM for the nitrogen in terms of those nitrogenous contaminants. The fabricated sensor shows strong applicability for real-time nitrogen monitoring in aqueous media, particularly for agricultural applications and smart agriculture systems. An average 96 ± 4 % accuracy was also verified for nitrogen sensing in agricultural river water samples.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111515"},"PeriodicalIF":8.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-05DOI: 10.1016/j.compag.2026.111501
A. Mishra , A. Krief , M.M. Sahoo , A. Schachter , I. Gonda , N. Dudai , T. Trigano , I. Herrmann
Rosemary extracts, including carnosic and rosmarinic acids (CA and RA, respectively), are known for their antimicrobial and antioxidant capabilities. Traditional quantification methods of CA and RA (later on called selected phytochemicals) are often destructive and time-consuming. This study presents a spectral, non-destructive, and time-efficient approach for estimating selected phytochemicals in pre- and post-harvest stages. We acquired spectral data from field-grown rosemary plants, dry leaves, and powder as well as UAV-borne hyperspectral imagery. The analysis included a transformation sequence (second derivative, Yeo–Johnson, and standardization), followed by partial least squares regression (PLSR). To mimic real-life scenarios, we investigated a training–testing strategy denoted by “leave-one-day-out”, systematically excluding each day’s data from training. For CA estimation, the PLSR model achieved a coefficient of determination () of 0.75 with a relative root mean square error (RRMSE) of 10.42% at the canopy level, 0.80 (RRMSE: 8.91%) for dry leaves, and 0.76 (RRMSE: 9.09%) for powder. RA estimation was challenging at the canopy level with an of 0.52 (RRMSE: 13.42%), but improved in post-harvest samples, reaching of 0.79 (RRMSE: 10.0%) for dry leaves and 0.75 (RRMSE: 9.78%) for powder. These results demonstrated the efficiency of the proposed approach. It offers a reliable alternative to traditional methods, with potential applications in agriculture and post-harvest industry.
{"title":"Pre- and post-harvest spectral estimation of carnosic acid and rosmarinic acid in rosemary","authors":"A. Mishra , A. Krief , M.M. Sahoo , A. Schachter , I. Gonda , N. Dudai , T. Trigano , I. Herrmann","doi":"10.1016/j.compag.2026.111501","DOIUrl":"10.1016/j.compag.2026.111501","url":null,"abstract":"<div><div>Rosemary extracts, including carnosic and rosmarinic acids (CA and RA, respectively), are known for their antimicrobial and antioxidant capabilities. Traditional quantification methods of CA and RA (later on called selected phytochemicals) are often destructive and time-consuming. This study presents a spectral, non-destructive, and time-efficient approach for estimating selected phytochemicals in pre- and post-harvest stages. We acquired spectral data from field-grown rosemary plants, dry leaves, and powder as well as UAV-borne hyperspectral imagery. The analysis included a transformation sequence (second derivative, Yeo–Johnson, and standardization), followed by partial least squares regression (PLSR). To mimic real-life scenarios, we investigated a training–testing strategy denoted by “leave-one-day-out”, systematically excluding each day’s data from training. For CA estimation, the PLSR model achieved a coefficient of determination (<span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span>) of 0.75 with a relative root mean square error (RRMSE) of 10.42% at the canopy level, 0.80 (RRMSE: 8.91%) for dry leaves, and 0.76 (RRMSE: 9.09%) for powder. RA estimation was challenging at the canopy level with an <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> of 0.52 (RRMSE: 13.42%), but improved in post-harvest samples, reaching <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> of 0.79 (RRMSE: 10.0%) for dry leaves and 0.75 (RRMSE: 9.78%) for powder. These results demonstrated the efficiency of the proposed approach. It offers a reliable alternative to traditional methods, with potential applications in agriculture and post-harvest industry.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111501"},"PeriodicalIF":8.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-05DOI: 10.1016/j.compag.2026.111460
Tao Zhou , Xuan Zhang , Jishu Wang , Linfeng Liu , Jianping Yang , Linyu Li , Tong Li
In complex agricultural environments such as mountain orchards, small autonomous equipment is widely used for multi-objective agricultural operation tasks (e.g., inspection, picking, spraying) due to its high mobility and adaptability. However, unstructured terrain and limited energy significantly increase the complexity of path planning and task scheduling. For this reason, this paper proposes a bi-level planning framework for multi-task path optimization problems in mountain orchards. In the first level, the framework adopts the Improved A*-History (IA*-H) algorithm, to solve the problem of paths between fruit trees or from a warehouse to a fruit tree in a mountain orchard, where terrain ups and downs. In the second level, a new Multi-Strategy Discrete Grey Wolf Optimizer (MSD-GWO) algorithm is proposed, to solve the path problem for multi-task scheduling throughout the orchard. After two level execution the optimal sequence and path for multi-tasks is determined. The experiment utilized typical mountainous orchard terrain data (Chu orange base in Longling County, Baoshan City, Yunnan Province), and the experimental results showed that the planning time of our method was reduced by 94.5% compared to Dijkstra and 85.1% compared to Z*. And Compared to the greedy scheduling strategy, our approach reduces path length by 24.6% and energy consumption by 20.6%. which verified the effectiveness and feasibility of our proposed bi-level framework. Our code can be found at https://github.com/zhoutao2333/ABLFPP.
{"title":"A bi-level planning framework for solving multi-task schedule path planning problem in mountain orchard","authors":"Tao Zhou , Xuan Zhang , Jishu Wang , Linfeng Liu , Jianping Yang , Linyu Li , Tong Li","doi":"10.1016/j.compag.2026.111460","DOIUrl":"10.1016/j.compag.2026.111460","url":null,"abstract":"<div><div>In complex agricultural environments such as mountain orchards, small autonomous equipment is widely used for multi-objective agricultural operation tasks (e.g., inspection, picking, spraying) due to its high mobility and adaptability. However, unstructured terrain and limited energy significantly increase the complexity of path planning and task scheduling. For this reason, this paper proposes a bi-level planning framework for multi-task path optimization problems in mountain orchards. In the first level, the framework adopts the Improved A*-History (IA*-H) algorithm, to solve the problem of paths between fruit trees or from a warehouse to a fruit tree in a mountain orchard, where terrain ups and downs. In the second level, a new Multi-Strategy Discrete Grey Wolf Optimizer (MSD-GWO) algorithm is proposed, to solve the path problem for multi-task scheduling throughout the orchard. After two level execution the optimal sequence and path for multi-tasks is determined. The experiment utilized typical mountainous orchard terrain data (Chu orange base in Longling County, Baoshan City, Yunnan Province), and the experimental results showed that the planning time of our method was reduced by 94.5% compared to Dijkstra and 85.1% compared to Z*. And Compared to the greedy scheduling strategy, our approach reduces path length by 24.6% and energy consumption by 20.6%. which verified the effectiveness and feasibility of our proposed bi-level framework. Our code can be found at <span><span>https://github.com/zhoutao2333/ABLFPP</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111460"},"PeriodicalIF":8.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-05DOI: 10.1016/j.compag.2026.111439
Baoxian Liang , Lihong Xu , Yu Su , Jianwei Du , Zhichao Deng
Efficient and scalable multi-task allocation presents a fundamental challenge in multi-machine cooperative operation for unmanned farming. Conventional approaches often assume static attributes and fixed-scale instances, thereby facing significant challenges in adapting to the dynamic characteristics of agricultural production processes. To address the dynamic multi-task allocation problem with time windows (DMAPTW), we propose a novel RL framework that automatically learns high-quality scheduling policies. A scale-agnostic representation mechanism is designed to accurately reflect the current system status, ensuring that the derived policy network is scale-agnostic. To enhance adaptability across diverse production environments, a combination method integrating problem-specific dispatching rules is implemented. Concurrently, a dense reward mechanism is proposed to directly associate the optimization objective. Numerical experiments conducted on a comprehensive set of synthetic instances demonstrate that the proposed algorithm exhibits robust flexibility in handling varying production configurations. Furthermore, comparative analyses reveal that this algorithm consistently outperforms meta-heuristic baselines by 28%–40%, indicating superior computational efficiency and robustness.
{"title":"Deep reinforcement learning for unmanned farming dynamic multi-task allocation problem","authors":"Baoxian Liang , Lihong Xu , Yu Su , Jianwei Du , Zhichao Deng","doi":"10.1016/j.compag.2026.111439","DOIUrl":"10.1016/j.compag.2026.111439","url":null,"abstract":"<div><div>Efficient and scalable multi-task allocation presents a fundamental challenge in multi-machine cooperative operation for unmanned farming. Conventional approaches often assume static attributes and fixed-scale instances, thereby facing significant challenges in adapting to the dynamic characteristics of agricultural production processes. To address the dynamic multi-task allocation problem with time windows (DMAPTW), we propose a novel RL framework that automatically learns high-quality scheduling policies. A scale-agnostic representation mechanism is designed to accurately reflect the current system status, ensuring that the derived policy network is scale-agnostic. To enhance adaptability across diverse production environments, a combination method integrating problem-specific dispatching rules is implemented. Concurrently, a dense reward mechanism is proposed to directly associate the optimization objective. Numerical experiments conducted on a comprehensive set of synthetic instances demonstrate that the proposed algorithm exhibits robust flexibility in handling varying production configurations. Furthermore, comparative analyses reveal that this algorithm consistently outperforms meta-heuristic baselines by 28%–40%, indicating superior computational efficiency and robustness.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111439"},"PeriodicalIF":8.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-04DOI: 10.1016/j.compag.2026.111526
Zijing Huang , Won Suk Lee , Yiannis Ampatzidis , Shinsuke Agehara , Natalia A Peres
Accurate yield forecasting is crucial in optimizing resource management and decision-making processes in agriculture, particularly in crops such as strawberries, which require precise predictions due to their rapid and continuous ripening cycles. This study introduces PheMuT, a novel phenology-informed, multi-modal time-series model that integrates visual and meteorological data streams to enhance strawberry yield forecasting. The proposed method employs advanced computer vision techniques, including two YOLOv11 detectors, an optimized ByteTrack tracker, Segment Anything (SAM), and Depth Anything v2 (DAv2), for precise fruit detection, canopy, and volume estimation. Concurrently, high-frequency weather data are processed using a self-supervised autoregressive Temporal Convolutional Network (TCN), resulting in concise and informative weather embeddings. These visual and weather features are fused within an LSTM-based model to produce weekly yield forecasts. PheMuT was validated using two strawberry cultivars at a Florida research facility over two consecutive seasons. Results indicated that PheMuT improved forecasting accuracy, reducing mean absolute error (MAE) by 10.7%, root mean squared error (RMSE) by 12.5%, and mean absolute percentage error (MAPE) by 18.6% compared to baseline manual methods. Additionally, the model exhibited a notable improvement of 17.2% in the coefficient of determination (R2). PheMuT offers an efficient, automated framework for yield forecasting. Code and data are available athttps://github.com/Sycamorers/PheMuT. The full datasets used in this study are available from the authors upon request.
{"title":"PheMuT: A phenology-informed, multi-modal time-series model for strawberry yield forecasting","authors":"Zijing Huang , Won Suk Lee , Yiannis Ampatzidis , Shinsuke Agehara , Natalia A Peres","doi":"10.1016/j.compag.2026.111526","DOIUrl":"10.1016/j.compag.2026.111526","url":null,"abstract":"<div><div><em>Accurate yield forecasting is crucial in optimizing resource management and decision-making processes in agriculture, particularly in crops such as strawberries, which require precise predictions due to their rapid and continuous ripening cycles. This study introduces PheMuT, a novel phenology-informed, multi-modal time-series model that integrates visual and meteorological data streams to enhance strawberry yield forecasting. The proposed method employs advanced computer vision techniques, including two YOLOv11 detectors, an optimized ByteTrack tracker, Segment Anything (SAM), and Depth Anything v2 (DAv2), for precise fruit detection, canopy, and volume estimation. Concurrently, high-frequency weather data are processed using a self-supervised autoregressive Temporal Convolutional Network (TCN), resulting in concise and informative weather embeddings. These visual and weather features are fused within an LSTM-based model to produce weekly yield forecasts. PheMuT was validated using two strawberry cultivars at a Florida research facility over two consecutive seasons. Results indicated that PheMuT improved forecasting accuracy, reducing mean absolute error (MAE) by 10.7%, root mean squared error (RMSE) by 12.5%, and mean absolute percentage error (MAPE) by 18.6% compared to baseline manual methods. Additionally, the model exhibited a notable improvement of 17.2% in the coefficient of determination (R<sup>2</sup>). PheMuT offers an efficient, automated framework for yield forecasting. Code and data are available at</em> <span><span><em>https://github.com/Sycamorers/PheMuT</em></span><svg><path></path></svg></span><em>.</em> The full datasets used in this study are available from the authors upon request.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111526"},"PeriodicalIF":8.9,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-04DOI: 10.1016/j.compag.2026.111522
Sally Deborah Pereira da Silva , Vinicius Richter , Norton Borges Junior , Regiane Aparecida Ferreira , Gustavo Vedooto Ferreira , Telmo Jorge Carneiro Amado , Luan Pierre Pott , Lucio de Paula Amaral
Root malformation disorder (RMD) is an abiotic stress that compromises the early development of Eucalyptus saligna plantations in Brazil, reducing growth, nutrient uptake, and canopy vigor. Despite its operational relevance, scalable tools for objective field detection remain limited. Remote sensing using unmanned aerial vehicles (UAVs) combined with deep learning offers a promising alternative for identifying early physiological stress. The objectives of this work were: (i) to characterize the biophysical attributes of plants affected by RMD; and (ii) to evaluate the feasibility of a deep learning–based approach to map different plant health conditions of E. saligna at the stand scale. Multispectral data from a RedEdge-MX sensor (Blue, Green, Red, Red-edge, NIR) were collected over an 11 ha, six-month-old E. saligna stand in Southern Brazil. Field measurements included plant height, diameter at breast height (DBH), chlorophyll content, and leaf nutrient concentrations. Ten vegetation indices (VIs) were computed, and Random Forest (Gini importance) identified two key predictors: the Canopy Chlorophyll Content Index (CCCI) and the Plant Senescing Reflectance Index (PSRI). U-Net++ was trained to classify four classes: healthy, unhealthy, dead plants, and soil/residues. RMD-affected trees showed significant reductions in height, DBH, chlorophyll content, and nutrient concentrations. The combination CCCI + PSRI yielded the best discrimination of unhealthy plants (precision = 98.75%, recall = 92.94%, F1-score = 95.76%), with an overall accuracy of 98.77%. Applied to the full stand, 93.41% of trees were classified as healthy, 3.70% as unhealthy, and 2.90% as dead. These findings demonstrate that UAV multispectral imagery integrated with U-Net++ enables accurate, low-cost detection of RMD-related stress, supporting early silvicultural decision-making and routine plantation monitoring.
{"title":"Remote detection of root malformation disorder in Eucalyptus saligna using UAV multispectral imagery and U-Net++","authors":"Sally Deborah Pereira da Silva , Vinicius Richter , Norton Borges Junior , Regiane Aparecida Ferreira , Gustavo Vedooto Ferreira , Telmo Jorge Carneiro Amado , Luan Pierre Pott , Lucio de Paula Amaral","doi":"10.1016/j.compag.2026.111522","DOIUrl":"10.1016/j.compag.2026.111522","url":null,"abstract":"<div><div>Root malformation disorder (RMD) is an abiotic stress that compromises the early development of <em>Eucalyptus saligna</em> plantations in Brazil, reducing growth, nutrient uptake, and canopy vigor. Despite its operational relevance, scalable tools for objective field detection remain limited. Remote sensing using unmanned aerial vehicles (UAVs) combined with deep learning offers a promising alternative for identifying early physiological stress. The objectives of this work were: (i) to characterize the biophysical attributes of plants affected by RMD; and (ii) to evaluate the feasibility of a deep learning–based approach to map different plant health conditions of <em>E. saligna</em> at the stand scale. Multispectral data from a RedEdge-MX sensor (Blue, Green, Red, Red-edge, NIR) were collected over an 11 ha, six-month-old <em>E. saligna</em> stand in Southern Brazil. Field measurements included plant height, diameter at breast height (DBH), chlorophyll content, and leaf nutrient concentrations. Ten vegetation indices (VIs) were computed, and Random Forest (Gini importance) identified two key predictors: the Canopy Chlorophyll Content Index (CCCI) and the Plant Senescing Reflectance Index (PSRI). U-Net++ was trained to classify four classes: healthy, unhealthy, dead plants, and soil/residues. RMD-affected trees showed significant reductions in height, DBH, chlorophyll content, and nutrient concentrations. The combination CCCI + PSRI yielded the best discrimination of unhealthy plants (precision = 98.75%, recall = 92.94%, F1-score = 95.76%), with an overall accuracy of 98.77%. Applied to the full stand, 93.41% of trees were classified as healthy, 3.70% as unhealthy, and 2.90% as dead. These findings demonstrate that UAV multispectral imagery integrated with U-Net++ enables accurate, low-cost detection of RMD-related stress, supporting early silvicultural decision-making and routine plantation monitoring.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111522"},"PeriodicalIF":8.9,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-04DOI: 10.1016/j.compag.2026.111520
Abdul Qadeer , Qi Wang , Rizwan Azim , Xiaole Zhao , Wen Ma , Ibrahim Awuku , Fujia Li , Qinglin Liu , Yanping Liu , Bing Liu , Xuchun Li , Muhammad Sanaullah , Abdul Wakeel , Safiya Bibi
Soil degradation, water scarcity, and plastic residue accumulation pose significant challenges to sainfoin (Onobrychis viciifolia L.) production under ridge-furrow rainwater harvesting (RFRH) in semiarid region. This study was aimed to optimize ridge width and straw length under novel RFRH integrated with straw-soil crust improving carbon sequestration and sainfoin production. A Three-year field experiment was carried out using a randomized block design comprising 10 treatments and 3 replications. Treatments were 3 ridge widths (30, 45, and 60 cm) 3 mulching materials (ridges integrated with soil crust (SC), short-chopped straw-soil crust (SSC, 2 cm), and long-chopped straw-soil crust (LSC, 10 cm)), and conventional flat planting (FP) as a control. The RFRH integrated with chopped straw-soil crust increased runoff, soil water storage (SWS), soil organic carbon (SOC), fodder yield, and water use efficiency (WUE) of sainfoin. Runoff coefficient for the ridge widths of 30, 45, and 60 cm was 0.23, 025, and 0.28, respectively, while for SC, SSC, and LSC was 0.21, 0.25, and 0.30, over three years. Compared to FP, the increase in SWS for the ridge widths of 30, 45, and 60 cm was 15.2, 23.5, and 32.6 mm, respectively, while for SC, SSC, and LSC was 14.4, 22.6, and 34.3 mm. The increase in SOC for the ridge widths was 20.1%, 33.5%, and 44.7%, respectively, while for straw lengths was 24%, 31.5%, and 42.8%. The increase in fodder yield for the ridge widths was 8.5%, 16.8%, and 28.9%, respectively, while for straw lengths was 13.4%, 17.4%, and 23.5%. The increase in WUE of sainfoin for the ridge widths was 2.0, 3.2, and 5.4 kg ha−1 mm−1, respectively, while for straw lengths was 2.3, 3.2, and 5 kg ha−1 mm−1. Structural equation modeling revealed that ridge width showed direct positive (standardized path coefficients = 0.56***) effect on SOC and indirect positive (standardized path coefficients = 0.15*) effect on WUE of sainfoin, while straw length demonstrated direct positive effect on SOC (standardized path coefficients = 0.41***) and WUE of sainfoin (standardized path coefficients = 0.15*). The Runoff coefficient, SWS, SOC, fodder yield, and WUE of sainfoin increased as the ridge width and straw length increased. In RFRH, wide ridges (60 cm) integrated with long-chopped straw-soil crust (10 cm) enhanced carbon sequestration and sainfoin production, offering viable replacement to plastic film mulching in semiarid region.
在半干旱区，土壤退化、水资源短缺和塑料残留物积累对垄沟雨水集雨生产红豆（Onobrychis viciifolia L.）构成了重大挑战。本研究旨在优化秸秆-土壤结皮复合RFRH下的垄宽和秸秆长度，以提高固碳和红豆素产量。采用随机区组设计，10个处理，3个重复，进行了为期3年的田间试验。3种垄沟宽度（30、45和60 cm） × 3种覆盖材料（垄沟与土壤结皮（SC）、短切秸秆-土壤结皮（SSC, 2 cm）和长切秸秆-土壤结皮（LSC, 10 cm）），以常规平栽（FP）为对照。秸秆-土壤结皮复合可提高红豆的径流量、土壤储水量、土壤有机碳、饲料产量和水分利用效率。30、45和60 cm的径流系数分别为0.23、025和0.28，而SC、SSC和LSC的径流系数分别为0.21、0.25和0.30。与FP相比，脊宽为30、45和60 cm时，SWS分别增加了15.2、23.5和32.6 mm，而SC、SSC和LSC分别增加了14.4、22.6和34.3 mm。垄沟宽度和秸秆长度分别增加了20.1%、33.5%和44.7%，秸秆长度分别增加了24%、31.5%和42.8%。垄沟宽度和秸秆长度分别提高了8.5%、16.8%和28.9%，秸秆长度分别提高了13.4%、17.4%和23.5%。垄沟宽度对红豆素水分利用效率的影响分别为2.0、3.2和5.4 kg ha - 1 mm - 1，而秸秆长度对红豆素水分利用效率的影响分别为2.3、3.2和5 kg ha - 1 mm - 1。结构方程模型表明，秸秆宽度对红豆的有机碳（标准化路径系数= 0.56***）和水分利用效率有直接的正影响（标准化路径系数= 0.15* *），秸秆长度对红豆的有机碳（标准化路径系数= 0.41***）和水分利用效率有直接的正影响（标准化路径系数= 0.15* *）。径流量系数、SWS、SOC、饲料产量和水分利用效率随垄宽和秸秆长度的增加而增加。在RFRH中，宽垄（60厘米）与长切秸秆-土壤结皮（10厘米）相结合，增强了碳固存和红豆素的生产，为半干旱地区的塑料薄膜覆盖提供了可行的替代方案。
{"title":"Structural equation modeling revealed optimized ridge-furrow configuration integrated with straw-soil crust enhancing carbon sequestration and sainfoin yield in semiarid agroecosystems","authors":"Abdul Qadeer , Qi Wang , Rizwan Azim , Xiaole Zhao , Wen Ma , Ibrahim Awuku , Fujia Li , Qinglin Liu , Yanping Liu , Bing Liu , Xuchun Li , Muhammad Sanaullah , Abdul Wakeel , Safiya Bibi","doi":"10.1016/j.compag.2026.111520","DOIUrl":"10.1016/j.compag.2026.111520","url":null,"abstract":"<div><div>Soil degradation, water scarcity, and plastic residue accumulation pose significant challenges to sainfoin (<em>Onobrychis viciifolia</em> L.) production under ridge-furrow rainwater harvesting (RFRH) in semiarid region. This study was aimed to optimize ridge width and straw length under novel RFRH integrated with straw-soil crust improving carbon sequestration and sainfoin production. A Three-year field experiment was carried out using a randomized block design comprising 10 treatments and 3 replications. Treatments were 3 ridge widths (30, 45, and 60 cm) <span><math><mo>×</mo></math></span> 3 mulching materials (ridges integrated with soil crust (SC), short-chopped straw-soil crust (SSC, 2 cm), and long-chopped straw-soil crust (LSC, 10 cm)), and conventional flat planting (FP) as a control. The RFRH integrated with chopped straw-soil crust increased runoff, soil water storage (SWS), soil organic carbon (SOC), fodder yield, and water use efficiency (WUE) of sainfoin. Runoff coefficient for the ridge widths of 30, 45, and 60 cm was 0.23, 025, and 0.28, respectively, while for SC, SSC, and LSC was 0.21, 0.25, and 0.30, over three years. Compared to FP, the increase in SWS for the ridge widths of 30, 45, and 60 cm was 15.2, 23.5, and 32.6 mm, respectively, while for SC, SSC, and LSC was 14.4, 22.6, and 34.3 mm. The increase in SOC for the ridge widths was 20.1%, 33.5%, and 44.7%, respectively, while for straw lengths was 24%, 31.5%, and 42.8%. The increase in fodder yield for the ridge widths was 8.5%, 16.8%, and 28.9%, respectively, while for straw lengths was 13.4%, 17.4%, and 23.5%. The increase in WUE of sainfoin for the ridge widths was 2.0, 3.2, and 5.4 kg ha<sup>−1</sup> mm<sup>−1</sup>, respectively, while for straw lengths was 2.3, 3.2, and 5 kg ha<sup>−1</sup> mm<sup>−1</sup>. Structural equation modeling revealed that ridge width showed direct positive (standardized path coefficients = 0.56***) effect on SOC and indirect positive (standardized path coefficients = 0.15*) effect on WUE of sainfoin, while straw length demonstrated direct positive effect on SOC (standardized path coefficients = 0.41***) and WUE of sainfoin (standardized path coefficients = 0.15*). The Runoff coefficient, SWS, SOC, fodder yield, and WUE of sainfoin increased as the ridge width and straw length increased. In RFRH, wide ridges (60 cm) integrated with long-chopped straw-soil crust (10 cm) enhanced carbon sequestration and sainfoin production, offering viable replacement to plastic film mulching in semiarid region.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111520"},"PeriodicalIF":8.9,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-04DOI: 10.1016/j.compag.2026.111531
Ambra Tosto , Alejandro Morales , Niels P.R. Anten , Pieter A. Zuidema , Jochem B. Evers
Pruning affects tree functioning by removing biomass and triggering compensatory responses. Functional-structural plant (FSP) models, combining three-dimensional plant architecture with physiological processes, are suitable tools to study pruning effects. We present and evaluate the first FSP model for cocoa trees and we simulate pruning impact on young cocoa tree functioning.
We performed two experiments: a parametrization experiment, assessing branching responses to pruning treatments (heading and thinning); and an evaluation experiment measuring the pruning effects on stem radius, leaf number and crown diameter of cocoa trees.
We developed an FSP model that simulates tree growth as a result of the interaction between physiological processes, tree architecture and pruning-induced changes in branching patterns. Bud break is simulated stochastically, based on bud position and pruning interventions and was parameterized with field observations. The evaluation experiment was replicated in silico to evaluate model predictions and quantify the effect of pruning on tree functioning.
Our model captured the immediate effects of pruning on tree structure and partially simulated the compensatory response in leaf production observed in the experiment. In the simulations, pruning reduced total light interception. The simulated mean light interception per unit leaf area was increased in one treatment. However, this advantage was quickly lost due to induced branch production.
Our model is a novel tool to study the impact of pruning, as it explicitly simulates tree architecture and pruning-induced responses. Our results highlight the necessity of dynamic simulations to understand pruning impact.
{"title":"Quantifying the impact of pruning on young cocoa trees using a functional-structural plant model","authors":"Ambra Tosto , Alejandro Morales , Niels P.R. Anten , Pieter A. Zuidema , Jochem B. Evers","doi":"10.1016/j.compag.2026.111531","DOIUrl":"10.1016/j.compag.2026.111531","url":null,"abstract":"<div><div>Pruning affects tree functioning by removing biomass and triggering compensatory responses. Functional-structural plant (FSP) models, combining three-dimensional plant architecture with physiological processes, are suitable tools to study pruning effects. We present and evaluate the first FSP model for cocoa trees and we simulate pruning impact on young cocoa tree functioning.</div><div>We performed two experiments: a parametrization experiment, assessing branching responses to pruning treatments (heading and thinning); and an evaluation experiment measuring the pruning effects on stem radius, leaf number and crown diameter of cocoa trees.</div><div>We developed an FSP model that simulates tree growth as a result of the interaction between physiological processes, tree architecture and pruning-induced changes in branching patterns. Bud break is simulated stochastically, based on bud position and pruning interventions and was parameterized with field observations. The evaluation experiment was replicated <em>in silico</em> to evaluate model predictions and quantify the effect of pruning on tree functioning.</div><div>Our model captured the immediate effects of pruning on tree structure and partially simulated the compensatory response in leaf production observed in the experiment. In the simulations, pruning reduced total light interception. The simulated mean light interception per unit leaf area was increased in one treatment. However, this advantage was quickly lost due to induced branch production.</div><div>Our model is a novel tool to study the impact of pruning, as it explicitly simulates tree architecture and pruning-induced responses. Our results highlight the necessity of dynamic simulations to understand pruning impact.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111531"},"PeriodicalIF":8.9,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-04DOI: 10.1016/j.compag.2026.111514
Zhiyan Liang , Luhan Wang , Hexiang Wang , Baohua Zhang , Chengliang Liu
Autonomous navigation of robots primarily relies on environment mapping, localization, path planning, and obstacle avoidance. However, when operating in large-scale and complex orchard environments over extended periods, robots often suffer from mapping drift and accumulated localization errors, posing significant challenges to perception and path planning. This study presents a multi-sensor fusion hardware platform specifically designed for agricultural orchard settings. Based on this platform, an enhanced FAST-LIO2 framework is proposed, incorporating loop closure detection and factor graph optimization to reduce point cloud matching errors and obtain a more accurate prior map. Building on the improved FAST-LIO2, a relocalization module based on the Normal Distributions Transform (NDT) point cloud matching algorithm is introduced to ensure more precise pose estimation. The 3D point cloud map is then processed using methods such as Statistical Outlier Removal (SOR) filtering and pass-through filtering before being projected into a 2D grid map. Path planning is subsequently performed using the RRT* and Timed Elastic Band (TEB) algorithms, leveraging the 2D map and real-time relocalization data. The proposed autonomous navigation system is evaluated in various orchard environments. The integration of backend optimization and relocalization significantly enhanced system performance, reducing point cloud matching errors by up to 93% in large-scale uneven terrains, with a root mean square error (RMSE) as low as 0.77 m. Moreover, the global planner RRT* and local planner TEB demonstrated the ability to generate safer and smoother trajectories. The results validate the safety and robustness of the proposed method, highlighting its promising application in autonomous navigation for orchard scenarios.
{"title":"Autonomous obstacle avoidance and path planning for mobile robots in orchard environments combining with map construction and positioning methods","authors":"Zhiyan Liang , Luhan Wang , Hexiang Wang , Baohua Zhang , Chengliang Liu","doi":"10.1016/j.compag.2026.111514","DOIUrl":"10.1016/j.compag.2026.111514","url":null,"abstract":"<div><div>Autonomous navigation of robots primarily relies on environment mapping, localization, path planning, and obstacle avoidance. However, when operating in large-scale and complex orchard environments over extended periods, robots often suffer from mapping drift and accumulated localization errors, posing significant challenges to perception and path planning. This study presents a multi-sensor fusion hardware platform specifically designed for agricultural orchard settings. Based on this platform, an enhanced FAST-LIO2 framework is proposed, incorporating loop closure detection and factor graph optimization to reduce point cloud matching errors and obtain a more accurate prior map. Building on the improved FAST-LIO2, a relocalization module based on the Normal Distributions Transform (NDT) point cloud matching algorithm is introduced to ensure more precise pose estimation. The 3D point cloud map is then processed using methods such as Statistical Outlier Removal (SOR) filtering and pass-through filtering before being projected into a 2D grid map. Path planning is subsequently performed using the RRT* and Timed Elastic Band (TEB) algorithms, leveraging the 2D map and real-time relocalization data. The proposed autonomous navigation system is evaluated in various orchard environments. The integration of backend optimization and relocalization significantly enhanced system performance, reducing point cloud matching errors by up to 93% in large-scale uneven terrains, with a root mean square error (RMSE) as low as 0.77 m. Moreover, the global planner RRT* and local planner TEB demonstrated the ability to generate safer and smoother trajectories. The results validate the safety and robustness of the proposed method, highlighting its promising application in autonomous navigation for orchard scenarios.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"244 ","pages":"Article 111514"},"PeriodicalIF":8.9,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146173941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号