Smart agricultural technology最新文献

{"title":"Duck egg embryonic development classification using transfer learning and CNN","authors":"Jinky J. Maglasang,&nbsp;Angelica C. Merced,&nbsp;Lyca B. Penales,&nbsp;Jennifer Joyce M. Montemayor,&nbsp;Renato V. Crisostomo,&nbsp;Haroun Al Raschid Christopher P. Macalisang,&nbsp;Malikey M. Maulana","doi":"10.1016/j.atech.2025.100932","DOIUrl":"10.1016/j.atech.2025.100932","url":null,"abstract":"<div><div>Duck eggs are a vital source of food and income for many Filipino households. However, in small to medium-sized poultry farms, farmers manually inspect eggs for quality during incubation, which can be laborious and prone to errors. This study aims to automate the classification process of duck eggs based on their stage of embryonic development (<em>fertilized</em>, <em>unfertilized</em>, or <em>rotten</em>) using image processing and deep learning techniques. A dataset of 9600 images of candled duck eggs were preprocessed using MPSO-CLAHE and applied uniform background transformation. The generated datasets were used to train CNN models based on AlexNet, VGG16, InceptionV3, ResNet50, and Xception. The VGG16 model exhibited superior performance with a training accuracy of 98.85%, validation accuracy of 98.81%, and testing accuracy of 97.40%. These initial results show the potential of this methodology to streamline production process and enhance the quality of duck egg products.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100932"},"PeriodicalIF":6.3,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143843322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated peanut defect detection using hyperspectral imaging and deep learning: A real-time approach for smart agriculture","authors":"Shih-Yu Chen ,&nbsp;Yu-Cheng Wu ,&nbsp;Yung-Ming Kuo ,&nbsp;Rui-Hong Zhang ,&nbsp;Tsai-Yi Cheng ,&nbsp;Yu-Chien Chen ,&nbsp;Po-Yu Chu ,&nbsp;Li-Wei Kang ,&nbsp;Chinsu Lin","doi":"10.1016/j.atech.2025.100943","DOIUrl":"10.1016/j.atech.2025.100943","url":null,"abstract":"<div><div>Manual visual inspection remains the prevailing approach for peanut quality classification; however, it is labor-intensive, prone to fatigue-induced errors, and often results in inconsistent outcomes. Peanut defects are typically categorized into four classes: healthy, underdeveloped, insect-damaged, and ruptured. This paper proposes an automated classification framework that integrates push-broom and snapshot hyperspectral imaging techniques with deep learning models for accurate and efficient peanut defect detection. A push-broom hyperspectral imaging system was employed to acquire a dataset of 1557 peanut samples, divided into a training set (477 samples: 237 healthy, 240 defective) and a test set (1080 samples). Spectral band selection was applied to reduce data dimensionality, followed by the development and evaluation of 1D, 2D, and 3D Convolutional Neural Network (CNN) models. Among them, the 3D-CNN architecture achieved the highest classification accuracy of 98 %. In addition, the snapshot imaging system enabled the construction of a lightweight CNN model for real-time defect detection. Principal Component Analysis (PCA) was utilized to identify five informative spectral bands, enabling efficient classification with an overall accuracy of 98.5 % and a Kappa coefficient of 97.3 %. The novelty of this study lies in the dual integration of push-broom and snapshot hyperspectral imaging with hybrid CNN architectures, enabling both high-accuracy offline analysis and lightweight real-time detection. The combination of spectral dimensionality reduction and attention-based modeling presents a scalable and computationally efficient solution for quality assessment. These findings represent a significant advancement in automated peanut grading, offering a robust, cost-effective, and scalable approach for deployment in smart agriculture and automated food quality control systems.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100943"},"PeriodicalIF":6.3,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143823840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Soil conservation and information technologies: A literature review","authors":"Jô Vinícius Barrozo Chaves ,&nbsp;Claudia Liliana Gutierrez Rosas ,&nbsp;Camila Porfirio Albuquerque Ferraz ,&nbsp;Luiz Henrique Freguglia Aiello ,&nbsp;Afonso Peche Filho ,&nbsp;Lia Toledo Moreira Mota ,&nbsp;Regina Márcia Longo ,&nbsp;Admilson Írio Ribeiro","doi":"10.1016/j.atech.2025.100935","DOIUrl":"10.1016/j.atech.2025.100935","url":null,"abstract":"<div><div>The evolution of real-time data technologies has significantly transformed several sectors, including agriculture. Advances in sensors, transducers, and artificial intelligence (AI) have driven automation and optimization in agricultural production processes, enabling detailed analyses for soil conservation. However, intensive land use and climate change represent critical challenges, threatening biodiversity and water resource quality. Image processing and spatial data analysis tools support informed decision-making in precision agriculture. This study conducted a systematic review on the SCOPUS platform, emphasizing AI technologies applied to soil management, coverage, and classification. The optimal combination of search terms, including “Agriculture”, “Deep Learning”, and “Soil”, yielded 909 publications. We selected 190 publications for detailed analysis. The review underscored the importance of remote sensing in developing indexes and predictive models, despite existing limitations in the scale of analysis. The growing application of neural network algorithms to recognize soil and plant structures reflects advancements in Information and Communication Technologies (ICT). Since 2020, there has been a notable increase in AI-driven approaches to soil conservation, highlighting a shift toward sustainable and regenerative management practices.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100935"},"PeriodicalIF":6.3,"publicationDate":"2025-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143817322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic target spraying and field evaluation of unstructured orchard based on millimeter-wave radar","authors":"Xing Xu ,&nbsp;Jianying Li ,&nbsp;Dongying Shen ,&nbsp;Jieli Duan ,&nbsp;Zhou Yang ,&nbsp;Yinlong Jiang","doi":"10.1016/j.atech.2025.100937","DOIUrl":"10.1016/j.atech.2025.100937","url":null,"abstract":"<div><div>Pesticide precision spraying and efficient deposition is an important development direction of smart agriculture. Aiming at the problems of low pesticide spraying efficiency and severe pesticide loss in unstructured orchards in hilly and mountainous areas, this study proposes an automatic target spray control method. A tracked orchard sprayer based on millimeter-wave radar is designed to address these issues. The information transmission between millimeter wave radar, controller and sprayer are realized, and automatic target spray operation of \"Walking-Sensing-Spraying\" are realized. Based on the improved self-adaptive DBSCAN clustering algorithm, the improved self-adaptive Alpha_Shape algorithm (a surface reconstruction algorithm) and the least squares circle fitting, the three-dimensional reconstruction and parameter extraction of the target canopy were realized. The results showed that the average relative errors of plant height, canopy width and volume after correction were 1.51 %, 1.96 % and 3.24 %, respectively. The maximum absolute error is 9.59 cm, 5.96 cm and 0.22 m³. The millimeter-wave radar point cloud can effectively characterize the plant height, canopy width and volume information of the target canopy, and meet the detection accuracy requirements of target spraying. Field experiment results show that the spray coverage under t automatic target spray meets the needs of orchard pest control, the application of pesticides is reduced by 36.12 %, which achieves the purpose of increasing efficiency, reducing application and precise application. Meanwhile, it can also provide methodological reference for other research on automatic target operation and other fields of automatic target spray technology.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100937"},"PeriodicalIF":6.3,"publicationDate":"2025-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143807727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of an Internet of Things (IoT)-based disease forecaster to manage purple spot on asparagus fern","authors":"John R. Spafford ,&nbsp;Mary K. Hausbeck ,&nbsp;Benjamin P. Werling ,&nbsp;Stewart F. Tucker ,&nbsp;Younsuk Dong","doi":"10.1016/j.atech.2025.100941","DOIUrl":"10.1016/j.atech.2025.100941","url":null,"abstract":"<div><div><em>Stemphylium vesicarium</em> causes purple spot disease on asparagus spears rendering them unmarketable. The pathogen also infects the asparagus fern, causing premature defoliation, impacting subsequent yields. Foliar disease on the fern is managed with fungicides which can be applied according to TOMCAST (TOMato disease foreCASTing) based on disease severity values (DSV) or a calendar-based schedule. Leaf wetness sensors play an important role in generating DSV but are not standardized. We assessed disease control when fungicides (azoxystrobin alternated with chlorothalonil) were applied according to TOMCAST using SpecConnect or the Internet of Things (IoT)-based LOCO-DM (Low-Cost sensor monitoring system for Disease Management) at two thresholds (15 or 20 DSV) or every 10 days. Weather data to determine the DSV were generated and compared using SpecConnect and LOCO-DM. The METER Group PHYTOS 31 sensor used in LOCO-DM provided more accurate results compared to the SpecConnect. In 2022, the SpecConnect model and LOCO-DM generated a season total of 113 and 109 DSV, respectively. In 2023, the 10-day treatment received 8 applications, the SpecConnect TOMCAST 15 and 20 DSV treatment received 6 and 4 applications, respectively. The LOCO-DM TOMCAST 15 and 20 DSV received 6 and 5 applications, respectively. Only the 10-day and LOCO-DM 15 DSV had a significantly lower final disease assessment than the non-treated control. Area under disease progress curve (AUDPC) data indicated that all treatments limited disease compared to the non-treated control. The final disease assessment and AUDPC values were similar between intervals applied using SpecConnect and LOCO-DM. The IoT based LOCO-DM can be used as an accessible way to advance disease forecasting so that fungicides are applied only when the risk of crop infection is high which may reduce disease management costs and environmental exposure without sacrificing control.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100941"},"PeriodicalIF":6.3,"publicationDate":"2025-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143800642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improved crop row detection by employing attention-based vision transformers and convolutional neural networks with integrated depth modeling for precise spatial accuracy","authors":"Hassan Afzaal ,&nbsp;Derek Rude ,&nbsp;Aitazaz A. Farooque ,&nbsp;Gurjit S. Randhawa ,&nbsp;Arnold W. Schumann ,&nbsp;Nicholas Krouglicof","doi":"10.1016/j.atech.2025.100934","DOIUrl":"10.1016/j.atech.2025.100934","url":null,"abstract":"<div><div>Precision agriculture has emerged as a revolutionary technology for tackling global food security issues by optimizing crop yield and resource management. Incorporating artificial intelligence (AI) within agricultural practices has fundamentally transformed the discipline by facilitating sophisticated data analysis, predictive modeling, and automation. This research presents a novel framework that integrates deep learning, precision agriculture, and depth modeling to detect crop rows and spatial information accurately. The proposed framework employs the latest attention and convolution-based encoders, such as ConvFormer, CAFormer, Swin Transformer, and ConvNextV2, in precisely identifying crop rows across varied and challenging agricultural environments. The binary segmentation models were trained using a high-resolution soybean crop dataset (733 images), which consisted of data from fifteen distinct locations in Canada, collected during different growth phases. LabelMe and albumentation tools were used to generate a segmentation dataset, followed by data augmentation techniques to enhance data generalization and robustness. With training (∼70 %, 513 images), validation (∼15 %, 109 images), and test (∼15 %, 111 images) splits, the models learned to differentiate crop rows from background noise, achieving notable accuracy across multiple metrics, including Precision, Recall, F1 Score, and Dice Score.</div><div>An essential element of this pipeline is incorporating the Depth Pro model for precise computation of Ground Sampling Distance (GSD) by estimating images' absolute height and depth maps. The depth maps were analyzed to examine GSD variability across fifteen clusters of field images, revealing a spectrum of GSD values ranging from 0.5 to 2.0 mm/pixel for most clusters. The proposed model demonstrates superior performance in crop row segmentation tasks, achieving an F1 Score of 0.8012, Precision of 0.8512, Recall of 0.7584, and Accuracy of 0.8477 on the validation set. In comparative analysis with state-of-the-art (SOTA) models, ConvFormer outperformed alternatives such as ConvNextv2, CAFormer, and Swin S3 across multiple metrics. Notably, ConvFormer achieves a higher balance of precision and recall than ResNet models, which exhibit lower metrics (e.g., F1 Score of 0.7307 and Recall of 0.6551), underscoring its effectiveness in complex agricultural scenarios. Furthermore, classic machine vision methods were tested for extracting line information from binary segmentation masks, which can be useful for plant analytics, autonomous driving, and other various applications. The proposed workflow offers a robust solution for automating field operations, optimizing resource efficiency, and improving crop productivity.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100934"},"PeriodicalIF":6.3,"publicationDate":"2025-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143806975","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the contribution of mango orchards to carbon sequestration using field derived data and satellite remote sensing techniques in Shibganj, Bangladesh","authors":"Md Asaduzzaman ,&nbsp;M Golam Mahboob ,&nbsp;Sharmin Sultana ,&nbsp;Nazmun Naher","doi":"10.1016/j.atech.2025.100931","DOIUrl":"10.1016/j.atech.2025.100931","url":null,"abstract":"<div><div>Carbon sequestration mitigates global warming by capturing and storing atmospheric CO₂. The robust growth, wide adaptability and long lifespan traits of mango plantations make them significant contributors to carbon sequestration. However, evaluating their carbon sequestration potential is challenging, as it requires accurate biomass estimation over a large area. This study aims to mapping mango orchards, investigate relationships between diameter at breast height (DBH) and the Normalized Difference Vegetation Index (NDVI) and evaluate the carbon sequestration potential using Landsat 8 OLI satellite imagery. Integrating field derived data with satellite remote sensing techniques are capable of estimating biomass carbon at a spatial scale with reasonable cost while maintaining acceptable accuracy. Using these integrated techniques, we have determined the role of mango orchards in carbon sequestration. Overall, 61 plots, each measuring 2500 m², were established to represent samples from different land cover types. The relationship between DBH and NDVI was scrutinized using simple linear regression analysis yielding R² value of 0.85, 0.96 and 0.77, for DBH classes &lt; 25 cm, &gt;25 cm to &lt;65 cm and &gt;65 cm, respectively. Using allometric equations with field derived data, the mean biomass carbon stock density was found 3.85 Mg ha⁻¹. However, when combining field derived data with remote sensing approaches, we found biomass carbon stock densities ranging from a maximum 8 Mg ha⁻¹ to a minimum 0.15 Mg ha⁻¹. The methodology adopted in this study will aid in large scale biomass carbon estimation for species-specific area, but its application to ecosystems with diverse species may require further investigation.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100931"},"PeriodicalIF":6.3,"publicationDate":"2025-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143817319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AIoT based soil nutrient analysis and recommendation system for crops using machine learning","authors":"Sehrish Munawar Cheema ,&nbsp;Ivan Miguel Pires","doi":"10.1016/j.atech.2025.100924","DOIUrl":"10.1016/j.atech.2025.100924","url":null,"abstract":"<div><div>Agriculture is indispensable to the global economy, and its growth is vital to any country's economic success. Menace changing climate, soil erosion, salinity, depletion in carrying capacity of the soil, and other environmental factors have challenged sustainable agriculture vis-a-vis the agronomic response of crops. Predicting the suitability of a crop for specific land is a challenging task as it depends on diverse climate, environmental, and soil factors. We proposed the solution to measure and analyze soil and environmental factors such as pH level, macro nutrients potassium (K), Nitrogen (N), Phosphorus (P) and humidity (h), temperature (t) and average rainfall. We utilized crop recommendation dataset from Kaggle consisting 22 crops. We build a prediction model using machine learning techniques. The models were trained on individual dataset of 20 major crops of Punjab Pakistan, using Decision Tree with AdaBoost, K-Nearest Neighbor (KNN), Decision Tree (DT), Random Forest (RF) and Support Vector Machine (SVM). The developed system compares and evaluates real-time data collected from implanted IoT-based sensors with a training dataset located in a cloud repository. Comparing the five ML models, Decision Tree with AdaBoost demonstrated the highest performance (AC: 98%). The system enables data-driven decision-making for selecting suitable crops for cultivation at specific sites through a user-friendly interface for farmers. Proposed system is non-intrusive for producing crop recommendations under diverse environmental regions and conditions, provides farmers with data-driven and valuable insights. The proposed system enables timely interventions to prevent crop loss, increasing global food security and contribute in sustainable agriculture.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100924"},"PeriodicalIF":6.3,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient detection and counting method for maize seedling plots","authors":"Feiyun Wang ,&nbsp;Hanlu Jiang ,&nbsp;Jincan Wu ,&nbsp;Fupeng Li ,&nbsp;Bo Zhao ,&nbsp;Wenhua Mao ,&nbsp;Chengxu Lv ,&nbsp;Liming Zhou ,&nbsp;Qingzhong Xu","doi":"10.1016/j.atech.2025.100914","DOIUrl":"10.1016/j.atech.2025.100914","url":null,"abstract":"<div><div>To efficiently detect and count maize seedlings in complex field conditions, this study first developed a sample dataset under diverse backgrounds and lighting scenarios and introduced a data augmentation technique called “M_AUG.” YOLOv5s was selected as the base model, enhanced with the Swin Transformer (Swin TR)to improve feature extraction across various scales and complex environments. The model also incorporated multi-scale attention (EMA)to enhance the representation of small samples and positive/negative samples, along with the Asymptotic Feature Pyramid Network (AFPN)to integrate seedling features at different levels. The results showed that the proposed SEA-YOLOv5 achieved mAP_0.5 of 98.6 %, mAP_0.5–0.95 of 73.2 %. and F1 of 97.1 %, with the parameters count of 5.55 million and a weight size of 11.7 MB. Compared to YOLOv5, SEA-YOLOv5 improved mAP_0.5 by 5.8 %, mAP_0.5–0.95 by 9.9 %, and F1 by 5.4 %, while reducing the parameter count by 1.46 million and weight size by 2.7 MB. SEA-YOLOv5 was compared with YOLOv7, YOLOv8s, Faster R-CNN, RetinaNet, YOLOv10s, DNE-YOLO, and YOLOv11s, and the results indicated that SEA-YOLOv5 outperformed the comparison models in overall performance. Upon deploying SEA-YOLOv5 on the Jetson Orin NX and conducting seedling detection and counting trials across eight plots, the model achieved a miss rate of just 0.63 % and a frame rate of 74.6 FPS. Thus, it can be concluded that the SEA-YOLOv5 model developed in this study provides high accuracy, a compact design, and strong portability, making it well-suited for real-time detection and counting applications in the field.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100914"},"PeriodicalIF":6.3,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143792797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of flow field visualization technology in analysing the influence of nutrient solution flow on hydroponic lettuce growth","authors":"Bateer Baiyin ,&nbsp;Yue Xiang ,&nbsp;Yang Shao ,&nbsp;Jung Eek Son ,&nbsp;Kotaro Tagawa ,&nbsp;Satoshi Yamada ,&nbsp;Mina Yamada ,&nbsp;Qichang Yang","doi":"10.1016/j.atech.2025.100933","DOIUrl":"10.1016/j.atech.2025.100933","url":null,"abstract":"<div><div>Nutrient solution flow is important for the growth and root morphology of lettuce in hydroponics, requiring precise regulation to optimise yield and quality. However, the mechanisms involved remain poorly understood. We examined the influence of varying nutrient solution flow rates on lettuce growth, root morphology, and nitrogen uptake. We assessed lettuce performance at five growth stages, measuring shoot and root dry and fresh weights, root morphology, and nitrogen uptake. Particle image velocimetry was employed to visualise the flow field, providing a deeper understanding of how flow patterns impact the root environment. In the early growth stage, lettuce under no flow conditions exhibited higher shoot and root biomass. However, moderate flow consistently outperformed other conditions as growth progressed, demonstrating significantly higher fresh and dry weights. High flow initially suppressed growth, highlighting the detrimental effects of excessively fast flow rates. No flow initially promoted root development, while moderate flow enhanced root growth later in the lifecycle. Nitrogen uptake analysis showed that moderate flow achieved the highest efficiency, while high flow increased nitrogen uptake flux in later stages. PIV visualisation revealed that moderate flow delivered uniform flow vectors and moderate velocity, enhancing nutrient ion contact with roots and uptake efficiency. In contrast, high flow resulted in chaotic flow vectors, high vorticity, and potential root damage, reducing uptake efficiency. Under no flow conditions, nutrient ion transport relied solely on diffusion, limiting nutrient availability during rapid growth and maturation. In conclusion, moderate flow was optimal for promoting lettuce growth and root development.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100933"},"PeriodicalIF":6.3,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143823841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}