Biosystems Engineering最新文献_第6页

Recognition in the early stage of powdery mildew damage for cucurbits plants using spectral signatures 利用光谱特征识别瓜类植物白粉病的早期危害

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-13 DOI: 10.1016/j.biosystemseng.2025.03.001

Claudia Angélica Rivera-Romero , Elvia Ruth Palacios-Hernández , Jorge Ulises Muñoz-Minjares , Osbaldo Vite-Chávez , Roberto Olivera-Reyna , Iván Alfonso Reyes-Portillo

{"title":"Recognition in the early stage of powdery mildew damage for cucurbits plants using spectral signatures","authors":"Claudia Angélica Rivera-Romero , Elvia Ruth Palacios-Hernández , Jorge Ulises Muñoz-Minjares , Osbaldo Vite-Chávez , Roberto Olivera-Reyna , Iván Alfonso Reyes-Portillo","doi":"10.1016/j.biosystemseng.2025.03.001","DOIUrl":"10.1016/j.biosystemseng.2025.03.001","url":null,"abstract":"<div><div>One of the most significant diseases affecting cucurbit plants is powdery mildew, which causes substantial yield losses in both greenhouses and field crops, especially during the winter and summer periods. Therefore, early diagnosis and detection are essential for effective pathogen control. An advanced, non-invasive method was developed for remotely sensing this fungal disease and assessing damage levels using spectral reflectance. The primary objective of this study is to detect the onset of the disease before the first visible symptoms appear on the leaves through the use of vegetation indices. To achieve this, statistical analyses and multiple comparison tests were employed for feature selection, in combination with machine learning algorithms, such as a Support Vector Machine. The results demonstrated high reliability in distinguishing between healthy and infected cucurbit leaves with powdery mildew. By calculating vegetation indices (VIs), seven optimal features were identified, enabling the recognition of three damage levels with 98% accuracy and a Cohen's <span><math><mrow><mi>κ</mi></mrow></math></span> coefficient of up to 0.96. Spectral reflectance successfully differentiated powdery mildew damage levels in cucurbit plants, suggesting that this method could be recommended for crops with similar characteristics.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 144-158"},"PeriodicalIF":4.4,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143621020","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}

引用次数: 0

Evaluation of the cubicle hood sampler for monitoring methane production of dairy cows under barn conditions 牛舍条件下监测奶牛甲烷产量的隔间罩取样器的评价

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-09 DOI: 10.1016/j.biosystemseng.2025.02.008

Cécile M. Levrault , Peter W.G. Groot Koerkamp , Carel F.W. Peeters , Nico W.M. Ogink

{"title":"Evaluation of the cubicle hood sampler for monitoring methane production of dairy cows under barn conditions","authors":"Cécile M. Levrault , Peter W.G. Groot Koerkamp , Carel F.W. Peeters , Nico W.M. Ogink","doi":"10.1016/j.biosystemseng.2025.02.008","DOIUrl":"10.1016/j.biosystemseng.2025.02.008","url":null,"abstract":"<div><div>Monitoring methane production from individual cows is necessary to evaluate the success of greenhouse gas reduction strategies. However, monitoring methane production rates (<strong>MPR</strong>) under practical conditions remains challenging. In this paper, we investigate the performance of a potential solution to this challenge.</div><div>The cubicle hood sampler (<strong>CHS</strong>) is an on-barn monitoring device placed in cubicles that collects the air exhaled by the animals while they lie down. The MPR of 28 dairy cows were measured by four CHS devices and compared to the levels measured by climate respiration chambers (<strong>CRC</strong>). A linear regression showed no strong correlation between the two sets of estimates (<em>r</em> = 0.24). The estimates made by the CHS appeared to be inaccurate due to a sampling bias (insufficient breath recovery), which could not be corrected for. Using Bayesian modelling, information was pooled across individuals to model complete methane production curves and potentially improve the accuracy of the MPR estimates. However, the model was unable to compensate for the biased observations used for fitting, and accuracy levels did not improve. An under-recovery of the breath samples by the hood is suspected. These issues must be resolved. Nevertheless, the CHS ranked cows satisfactorily, with Kendall W values of 0.625 (<em>p</em> = 0.201) in the original dataset, and 0.659 (<em>p</em> = 0.214) after using the model. Resolving the bias issue is expected to have a simultaneous positive effect on the agreement between the two MPR rankings. We recommend to keep using the model to convert discrete measurements into methane production curves.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 115-125"},"PeriodicalIF":4.4,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143580521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Microcosm fabrication platform for live microscopy of plant-soil systems 植物-土壤系统活体显微镜显微制作平台

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-08 DOI: 10.1016/j.biosystemseng.2025.02.006

Yangminghao Liu , Daniel Patko , Alberto Lora de la Mata , Xingshui Dong , Emma Gomez Peral , Xinhua He , Bruno Ameduri , Vincent Ladmiral , Michael P. MacDonald , Lionel X. Dupuy

{"title":"Microcosm fabrication platform for live microscopy of plant-soil systems","authors":"Yangminghao Liu , Daniel Patko , Alberto Lora de la Mata , Xingshui Dong , Emma Gomez Peral , Xinhua He , Bruno Ameduri , Vincent Ladmiral , Michael P. MacDonald , Lionel X. Dupuy","doi":"10.1016/j.biosystemseng.2025.02.006","DOIUrl":"10.1016/j.biosystemseng.2025.02.006","url":null,"abstract":"<div><div>Our ability to fully understand how plants acquire water and nutrients from the soil is constrained by the limitations of current technologies. Soil structures and properties are complex, dynamic, and profoundly modified by root and microbial secretions. Detailed descriptions of soil properties are rarely available to the researcher because natural soil is opaque, making direct observations challenging. To address these experimental difficulties, microcosm systems dedicated to live imaging of rhizosphere processes in highly controlled environmental conditions were developed. The system is based on fluorinated granular materials with low refractive indices, termed transparent soils. Microcosm chambers were assembled using poly(dimethyl siloxane) parts (PDMS) fabricated by injection moulding and subsequently joined to glass slides. The control of liquid fluxes in the microcosm was achieved by syringes passing through the PDMS parts or through custom made PDMS sponges. The platform was tested for live imaging experiments using Light Sheet Fluorescence microscopy. Results demonstrated that the platform is suitable for a diverse range of experiments, including live observation of plant roots, split-soil systems and investigations into the effects of soil heterogeneity, controlled water content experiments, and dye tracer monitoring. The technique was used to quantify the increase in infiltration rate due to the presence of roots in soil. This study demonstrates the potential of combining new materials and microfabrication techniques to overcome current limitations on plant-soil interaction research.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 105-114"},"PeriodicalIF":4.4,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143580520","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}

引用次数: 0

Deep learning-based model to classify mastitis in Holstein dairy cows 基于深度学习的荷斯坦奶牛乳腺炎分类模型

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-06 DOI: 10.1016/j.biosystemseng.2025.02.013

Mengyuan Chu , Yongsheng Si , Qian Li , Xiaowen Liu , Gang Liu

{"title":"Deep learning-based model to classify mastitis in Holstein dairy cows","authors":"Mengyuan Chu , Yongsheng Si , Qian Li , Xiaowen Liu , Gang Liu","doi":"10.1016/j.biosystemseng.2025.02.013","DOIUrl":"10.1016/j.biosystemseng.2025.02.013","url":null,"abstract":"<div><div>The occurrence and prevalence of dairy cow mastitis has brought significant challenges to animal welfare and economy. To overcome the complexities and accumulated errors present in previous detection methods, a rapid and accurate mastitis detection approach is developed based on image processing and deep learning, leveraging thermal infrared imaging. Image processing techniques, including the Hough transform and morphological operations, are used to classify affected cows from thermal images. An image pyramid is constructed based on upsampling to tackle motion blur induced by the cows' rapid movement. The multi-scale convolution and the spatial and channel Squeeze & Excitation (scSE) block were integrated into the DenseNet-201 architecture to enhance the feature extraction process. This enabled the network to adaptively recalibrate channel-wise feature responses and strengthening the discriminative power of the learned representations. For mastitis detection, a deep learning model, the multi-scale scSE-DenseNet-201 (MS-scSE-DenseNet-201) architecture, is refined to predict the severity of mastitis. The framework takes images of both sides of the cow's udder as input, and outputs one of three mastitis severity levels: negative (N), subclinical mastitis (SCM), or clinical mastitis (CM). To assess the model's performance in detecting mastitis, a dataset comprising 5000 thermal images from 802 cows, was used. The model achieved accuracy, precision, and recall of 90.18%, 92.16%, and 88.38%, respectively, showing notable improvement over previous methods. This work integrated object segmentation and blind deblurring to strengthen the MS-scSE-DenseNet-201 in the automatic detection of cow mastitis, which will open a promising application horizon for other animal disease diagnostics.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 92-104"},"PeriodicalIF":4.4,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551919","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}

引用次数: 0

Citrus fruit diameter estimation in the field using monocular camera 柑桔果实直径的单目相机估算

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-01 DOI: 10.1016/j.biosystemseng.2025.02.012

Hongchun Qu , Haitong Du , Xiaoming Tang , Shidong Zhai

{"title":"Citrus fruit diameter estimation in the field using monocular camera","authors":"Hongchun Qu , Haitong Du , Xiaoming Tang , Shidong Zhai","doi":"10.1016/j.biosystemseng.2025.02.012","DOIUrl":"10.1016/j.biosystemseng.2025.02.012","url":null,"abstract":"<div><div>Accurate and efficient measurement of citrus fruit size is essential for managing tree form and estimating yields. Conventional manual methods are reliable but highly labour-intensive, while existing machine vision solutions often require specialised setups (e.g., distance calibration or 3D sensors). In this study, a low-cost, monocular-based framework that uses mature and healthy leaves as natural reference objects was proposed, eliminating the need for manual markers or complex camera parameter calibration. By compiling an offline leaf-size distribution from multiple citrus varieties, this method automatically converts fruit pixels to real-world diameters using the largest near-frontal leaf in each image. Further, the work integrates the deformable convolution (DNCv2) and shuffle attention (SA) into a YOLOv8 detector to improve occlusion handling, ensuring robust detection even when fruits are partially obscured by foliage. Extensive validation on three different citrus cultivars shows that leaf-size variability contributes less than 3.2% relative error in diameter estimation, while the overall approach achieves 93.14% accuracy and <em>R</em><sup>2</sup> = 0.76. Key contributions include: (1) a novel monocular technique leveraging inherent orchard elements (leaves) as references, (2) advanced detection modules to tackle partial occlusion, (3) cross-variety validation demonstrating consistent performance, and (4) a fast, user-friendly workflow suitable for real-world orchard applications. Future work will explore multi-frame or multi-view strategies to further refine diameter measurement under heavy occlusion.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 47-60"},"PeriodicalIF":4.4,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143510032","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}

引用次数: 0

Detection and tracking of oestrus dairy cows based on improved YOLOv8n and TransT models 基于改进的 YOLOv8n 和 TransT 模型的发情奶牛检测与跟踪

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-01 DOI: 10.1016/j.biosystemseng.2025.02.005

Zheng Wang , Hongxing Deng , Shujin Zhang , Xingshi Xu , Yuchen Wen , Huaibo Song

{"title":"Detection and tracking of oestrus dairy cows based on improved YOLOv8n and TransT models","authors":"Zheng Wang , Hongxing Deng , Shujin Zhang , Xingshi Xu , Yuchen Wen , Huaibo Song","doi":"10.1016/j.biosystemseng.2025.02.005","DOIUrl":"10.1016/j.biosystemseng.2025.02.005","url":null,"abstract":"<div><div>Real-time monitoring of oestrus cows in dairy farming is labour-intensive and time-consuming. To achieve accurate detection and real-time positioning of oestrus cows in natural scenes, a model named YOLO-TransT, integrating the improved YOLOv8n and Transformer Tracking (TransT) models, was proposed for oestrus cow detection and tracking. Firstly, the Context Augmentation Module (CAM) was incorporated into YOLOv8n to enhance the model's focus on the oestrus cow by associating with mounting behaviour; Secondly, the Squeeze-and-Excitation (SE) module was introduced to boost the network's learning ability and suppress redundant features; Thirdly, the improved YOLOv8n and TransT were integrated to obtain the YOLO-TransT model, which realised the detection and tracking of oestrus cow; Finally, based on YOLO-TransT, a cow oestrus monitoring and warning system was designed. The experimental results showed that in the detection part of the YOLO-TransT, the improved YOLOv8n achieved a 92.60% Average Precision of oestrus (AP<sub>oestrus</sub>), 92.00% F1-score, with 3.14 M parameters, 9.70 G Floating-point Operations (FLOPs), and a 7.0 ms/frame detection speed. Compared to the original YOLOv8n, the improved YOLOv8n had increased AP<sub>oestrus</sub> by 4.10% and F1-score by 3.25%, while keeping the parameters, FLOPs, and detection speed essentially unchanged; In the tracking part, the TransT model had a tracking success rate of 70.3%, a precision value of 85.5%, and an Area under Curve (AUC) value of 71.4%. In conclusion, the YOLO-TransT could accurately detect and track oestrus cows in natural scenes, laying the foundation for intelligent livestock breeding. The dataset and code were released on GitHub (<span><span>https://github.com/XingshiXu/ZhengWang_YOLO-TransT</span><svg><path></path></svg></span>).</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 61-76"},"PeriodicalIF":4.4,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143527127","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}

引用次数: 0

Mechanism of rice bran removal at individual grain and population levels in abrasive rice mill 磨料碾米机在籽粒和群体水平上去除米糠的机理

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-01 DOI: 10.1016/j.biosystemseng.2025.02.010

Ze Sun , Xinlei Wang , Anqi Li , Jiaming Fei , Wenyu Feng , Dan Zhao , Yanlong Han , Fuguo Jia , Hao Li , Shouyu Ji , Zhuozhuang Li

{"title":"Mechanism of rice bran removal at individual grain and population levels in abrasive rice mill","authors":"Ze Sun , Xinlei Wang , Anqi Li , Jiaming Fei , Wenyu Feng , Dan Zhao , Yanlong Han , Fuguo Jia , Hao Li , Shouyu Ji , Zhuozhuang Li","doi":"10.1016/j.biosystemseng.2025.02.010","DOIUrl":"10.1016/j.biosystemseng.2025.02.010","url":null,"abstract":"<div><div>In the process of rice bran layer removal using abrasive rice mills, over-milling will result in nutritional loss, while under-milling will result in poor palatability. However, achieving moderate milling with an abrasive rice mill can be challenging due to the rice bran layer removal mechanism. This study investigates the mechanism of bran layer removal in abrasive rice mills by analysing the wear and structural characteristics on the rice surface, as well as the motion of rice grains in the milling chamber. The results showed that surface wear due to the contact of the rice grains with the grit was the main reason for debranning. At the individual grain level, the process of removing the bran layer in the abrasive rice mill is phased, synchronised, and orderly. The removal process can be divided into three stages depending on the morphology of the residual bran layer and the wear mechanism. The rotation motion leads to the synchronous removal of the bran layer in different regions of the rice grains. The bran layer in different regions is removed sequentially due to the varying number of depressions. At the rice population level, the axial and radial positions exchange of the rice grains in the milling chamber ensures overall uniformity in removing the rice bran layer. These findings are valuable for optimising the design of the abrasive mills and guiding the mill uniformity in similar types of mills.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 32-46"},"PeriodicalIF":4.4,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143510031","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}

引用次数: 0

Three-dimensional dynamic simulation of the rice root system under different phosphorus concentrations 不同磷浓度下水稻根系三维动态模拟

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-03-01 DOI: 10.1016/j.biosystemseng.2025.02.011

Le Yang, Lan Long, Shirong Ai, Qiangqiang Zhou, Wenhui Li, Ting Liu

{"title":"Three-dimensional dynamic simulation of the rice root system under different phosphorus concentrations","authors":"Le Yang, Lan Long, Shirong Ai, Qiangqiang Zhou, Wenhui Li, Ting Liu","doi":"10.1016/j.biosystemseng.2025.02.011","DOIUrl":"10.1016/j.biosystemseng.2025.02.011","url":null,"abstract":"<div><div>Phosphorus is a vital element for plant growth, and it exerts a significant influence on the growth, development, and yield of rice. In order to study the dynamic growth status of rice roots under different phosphorus concentrations, this paper takes the roots of two rice varieties, HuaJing (HJ) and Metzam (MTZ), as the research objects, simulate the growth of the root system under phosphorus deficient (LP), 50% phosphorus (MP), and normal phosphorus (HP) conditions, and proposes an L-system-based three-dimensional rice root-environmental growth model DRoots, which employs the idea of the operator-splitting approach to the extended movement of soil phosphorus and the root growth of the two processes are modelled separately, and then they are coupled by iteration. Root growth is simulated in a rectangular geometry planter during root growth modelling, and collisions between roots and between roots and the inner wall of the planter are considered. The root growth of two varieties of rice at different phosphorus concentrations was compared by three-dimensional dynamic simulation studies, which analysed the existence of differences in the phosphorus demand and response of the root systems of the two varieties of rice from different perspectives, as well as the correlation between the indicators of each parameter. The results showed that these parameters were significantly correlated with each other, and thus the model could simulate the growth of rice root system more accurately. This study provides a reference for other crop root systems to explore the mechanism of phosphorus effect on rice root growth.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 77-91"},"PeriodicalIF":4.4,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551917","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}

引用次数: 0

Early detection of downy mildew in vineyards using deep neural networks for semantic segmentation 基于深度神经网络语义分割的葡萄园霜霉病早期检测

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-02-25 DOI: 10.1016/j.biosystemseng.2025.02.007

Inés Hernández , Rui Silva , Pedro Melo-Pinto , Salvador Gutiérrez , Javier Tardaguila

{"title":"Early detection of downy mildew in vineyards using deep neural networks for semantic segmentation","authors":"Inés Hernández , Rui Silva , Pedro Melo-Pinto , Salvador Gutiérrez , Javier Tardaguila","doi":"10.1016/j.biosystemseng.2025.02.007","DOIUrl":"10.1016/j.biosystemseng.2025.02.007","url":null,"abstract":"<div><div>Downy mildew is a critical disease in viticulture, typically identified through manual inspection of individual leaves in the field by experts. The combination of artificial intelligence techniques with mobile platforms can optimise non-invasive detection. This work focused on employing semantic segmentation deep neural networks to detect visual symptoms of downy mildew in high-resolution grapevine images under field conditions. Vineyard canopy images were collected from 14 plots using both manual and mobile platform methods. The study compared six architectures and six encoders using transfer learning, as well as two SegNet AdHoc architectures. To address imbalance problems, simple data augmentation, MixUp, oversampling, and undersampling techniques were employed. The results were adjusted through test-time augmentation. The study found that the U-Net architecture, using the MobileVit-S encoder and the Dice loss function, was particularly efficient. The U-Net architecture with light-weight encoders exhibited potential for real-time applications. The robustness of the model was improved by combining oversampling and undersampling with simple data augmentation during training. The classification of areas with and without disease symptoms achieved an accuracy of 86% and an f1-score of 82%. Additionally, the number of symptoms in grapevine canopy images was detected with an NRMSE of 12%. In conclusion, the proposed methodology shows promise for efficiently early assessing grapevine downy mildew under field conditions. This approach could be applied to other crop diseases and pests, taking advantage of the complexity of the dataset to strengthen the robustness of the model in real-world scenarios.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 15-31"},"PeriodicalIF":4.4,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143487901","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}

引用次数: 0

Acceleration of pipeline analysis for irrigation networks through parallelisation in Graphic Processing Units 通过图形处理单元的并行化加速灌溉管网的管道分析

IF 4.4 1区农林科学

Biosystems Engineering Pub Date : 2025-02-21 DOI: 10.1016/j.biosystemseng.2025.02.004

Fernández-Pato J, Zapata N, Latorre B, Playán E

{"title":"Acceleration of pipeline analysis for irrigation networks through parallelisation in Graphic Processing Units","authors":"Fernández-Pato J, Zapata N, Latorre B, Playán E","doi":"10.1016/j.biosystemseng.2025.02.004","DOIUrl":"10.1016/j.biosystemseng.2025.02.004","url":null,"abstract":"<div><div>This paper reports on the development of a Farming irrigation network Analysis and Simulation Tool (FAST) on Graphic Processing Units (GPU). The tool is oriented to accelerate the optimisation of pressurised hydraulic networks equipped with hydrants and/or sprinklers, which may require millions of hydraulic simulations to converge to the optimal solution. GPU devices contain a large number of processors working in parallel and are capable of applying the same computational algorithms over different simulation parameters. Collective and on-farm pressurised irrigation networks typically have a branched structure, without flow recirculation. This permits to implement massive parallelisation of hydraulic calculations. The efficiency of the proposed code was compared to the EPANET hydraulic software, which is widely used worldwide for this type of problems. Results show efficiency gains larger than 6,000x with respect to simulations performed using the EPANET developer's toolkit. An evaluation of the efficiency scalability in terms of the network size was also assessed. Results showed a dramatic performance improvement as the network size increased. FAST-GPU leverages the massive parallelisation capabilities of GPUs to achieve a staggering speedup compared to traditional CPU-bound simulations. This paradigm shift opens the doors for complex irrigation network analysis previously considered computationally prohibitive. This is particularly necessary for the optimisation of network design and management processes.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"252 ","pages":"Pages 1-14"},"PeriodicalIF":4.4,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143453942","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}

引用次数: 0