Computers and Electronics in Agriculture最新文献

{"title":"UAV multispectral remote sensing for the estimation of SPAD values at various growth stages of maize under different irrigation levels","authors":"Weitong Ma ,&nbsp;Wenting Han ,&nbsp;Huihui Zhang ,&nbsp;Xin Cui ,&nbsp;Xuedong Zhai ,&nbsp;Liyuan Zhang ,&nbsp;Guomin Shao ,&nbsp;Yaxiao Niu ,&nbsp;Shenjin Huang","doi":"10.1016/j.compag.2024.109566","DOIUrl":"10.1016/j.compag.2024.109566","url":null,"abstract":"<div><div>Chlorophyll is crucial for photosynthesis in plants and the readings by a SPAD meter (Soil and Plant Analyzer Development) can be used to represent leaf chlorophyll content for monitoring crop growth status and predicting grain yield. Remote sensing technology has shown potential in non-destructive monitoring of SPAD values over large areas, but current SPAD inversion models are limited in their ability to incorporate multiple principal components besides spectral parameters, adapt to other variables such as water stress, and predict SPAD only throughout the entire growth period. This two-year study used crop parameters (plant height and leaf area index) and vegetation indices (VI) derived from unmanned aerial vehicle (UAV) multispectral images to develop SPAD prediction models for maize under different irrigation levels in the 2018 and 2019 growing seasons in Inner Mongolia, China. Two nonlinear machine learning models, random forest (RF) and support vector regression (SVR), and a multiple statistical regression method (partial least squares regression (PLSR)) were used to modeling SPAD. The results showed that the VIs with a high correlation with SPAD varied at each growth stage and the accuracy of SPAD estimation model can be improved significantly by dividing different growth stages (R² increased by more than 104 %). PLSR performed better than RF and SVR for each growth period, especially at the reproductive R stage (R² = 0.79, RMSE = 2.25). LAI and PH did not always improve prediction accuracy, but adding crop parameters did increase the correlation coefficient between predicted values and biomass by 8.3 %. This study provides valuable insights into the estimation of SPAD at different growth stages of maize under varying water stress levels using UAV data and crop parameters, offering guidance for farmland management and yield prediction.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109566"},"PeriodicalIF":7.7,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526006","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}

{"title":"Detection of a vascular wilt disease in potato (‘Blackleg’) based on UAV hyperspectral imagery: Can structural features from LiDAR or SfM improve plant-wise classification accuracy?","authors":"Marston H.D. Franceschini ,&nbsp;Benjamin Brede ,&nbsp;Jan Kamp ,&nbsp;Harm Bartholomeus ,&nbsp;Lammert Kooistra","doi":"10.1016/j.compag.2024.109527","DOIUrl":"10.1016/j.compag.2024.109527","url":null,"abstract":"<div><div>Ensuring plant health is a key factor to maximize crop yield. Despite that, the current field scouting and disease monitoring approaches often rely on visual evaluations and are, therefore, subjective and time demanding. New methods to assist in disease detection and severity assessment are required to allow better crop management and higher throughput in field phenotyping studies. With this objective, techniques involving the use of multi- and hyperspectral imagery for retrieval of plant traits and assessment of general crop health status are increasingly being proposed as alternatives to conventional disease monitoring approaches. Conversely, research focusing on specific pathogens are still lacking in many cases, in particular studies investigating multi-source sensing approaches, which have the potential to improve retrieval/classification accuracy. In this study, hyperspectral imagery and point clouds obtained with LiDAR or through Structure from Motion algorithm (SfM) applied to high resolution RGB images were evaluated as possible alternatives to detect Blackleg (caused by bacteria of the genera <em>Pectobacterium</em> and <em>Dickeya</em>) in potato. It was demonstrated that all the different datasets have potential to discriminate healthy from diseased plants. The combination of Vegetation Indices (VIs) derived from hyperspectral images with structural features from LiDAR resulted in the best validation results (Balanced Accuracy – BA = 0.915). Small improvements were also achieved by combining VIs with SfM features (BA = 0.876) in comparison to VIs alone (BA = 0.846). Evaluation of feature importance for classification models derived from the different datasets indicated that after structural features derived from LiDAR or RGB imagery were added as predictor variables the relative importance of VIs for the predictions decreased, in particular for VIs related to LAI or other traits describing canopy properties. Finally, analysis of false negatives and positives indicated some limitations to the predictive potential of the different datasets, with diseased and healthy plants eventually presenting atypical structural and spectral characteristics in comparison to those expected for their classes. Therefore, multi-source sensing, including additional modalities (e.g., thermal or fluorescence), might be required to further improve detection of pathogens with complex symptoms, as those affecting roots, tubers and stems.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109527"},"PeriodicalIF":7.7,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526008","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}

{"title":"MRSU2Net: A novel method for semantic segmentation of group lettuce from individual Objectives to group Objectives","authors":"Pan Zhang,&nbsp;Daoliang Li","doi":"10.1016/j.compag.2024.109560","DOIUrl":"10.1016/j.compag.2024.109560","url":null,"abstract":"<div><div>Semantic segmentation methods have played an important role in a wide range of applications, as they contribute to more accurate phenotypic information extraction in the field of plant phenotype. However, the high annotation cost of semantic segmentation datasets remains a major challenge, and most of them are constructed and validated on training and testing datasets with similar scales. Most studies overlook its effectiveness on multi-scale datasets, especially on low resolution datasets. Although some semantic segmentation methods extract and learn multi-scale features from datasets through methods such as multi-scale feature fusion modules and attention mechanisms, the model’s scale down compatibility, i.e. the segmentation reliability of the model on low resolution datasets, has not yet been verified. To address this challenge, this study proposes for the first time a new approach to plant object oriented semantic segmentation, which involves modeling individual target datasets and validating group target datasets. This modeling approach can significantly reduce the annotation cost of datasets to some extent. On this basis, we propose a multi-scale feature fusion module (MSFAF-M) for multi-level feature relationship exploration and a multi receptive field feature fusion module (MRFFF-S) for single-layer feature relationship exploration. By applying MSFAF-M and MRFFF-S to U2Net, an upgraded semantic segmentation method MRSU2Net is proposed, which can fully extract global and local feature information of target objects at multiple scales, and improve the segmentation reliability of semantic segmentation models based on individual target datasets on multi-scale group target datasets. Due to the fact that the construction approach of the semantic segmentation model proposed in this study is different from traditional semantic segmentation methods, we validated the scale down compatibility of MRSU2Net on the target dataset of lettuce populations collected at the seedling stage. When MRSU2Net is applied to group target images with the same resolution (2992 × 2992), the MIoU is 0.9719 and the inference-time is 0.3550. When MRSU2Net is applied to group target images of the same input size (224 × 224), the MIoU can reach 0.7346 and the inference time is 0.0219. The results demonstrate that the segmentation performance of the MRSU2Net constructed in this study is significantly superior to other classic semantic segmentation methods in low resolution images.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109560"},"PeriodicalIF":7.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526000","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}

{"title":"Optimizing soil resistance and disturbance of bionic furrow opener for paddy field based on badger claw using the CFD-DEM method","authors":"Zhenyu Tang ,&nbsp;Zhiwei Zeng ,&nbsp;Shuanglong Wu ,&nbsp;Dengbin Fu ,&nbsp;Jihan He ,&nbsp;Yinghu Cai ,&nbsp;Ying Chen ,&nbsp;Hao Gong ,&nbsp;Long Qi","doi":"10.1016/j.compag.2024.109549","DOIUrl":"10.1016/j.compag.2024.109549","url":null,"abstract":"<div><div>Paddy field fertilizer banding requires the use of high-performance soil furrow openers. The bionic design method has been identified as an effective approach to obtain desired results of a soil-engaging tool. This study utilized this approach in designing a bionic furrow opener for fertilization in paddy fields. The bionic furrow opener was designed based on the main physical characteristics of the North American badger claws, with the fitted curve of the badger claw enlarged eight times. A paddy field soil-opener interaction model was developed to evaluate the performance of different design alternatives with various combinations of curvature radii (<em>R</em>) and width (<em>W_o</em>). The opener performance (including soil resistance, soil disturbance characteristics, and time for completing soil backfilling) of the different design alternatives was monitored in simulations and analyzed. Results showed that the designed bionic furrow opener achieved the best performance with a combination of <em>R</em> and <em>W_o</em> being 25 mm and 30 mm, respectively. Force measurement experiments and soil disturbance measurement experiments were conducted in field and laboratory conditions, respectively to validate the soil-opener interaction model and the performance of optimized opener. The validation results showed relative errors of 16.64 % and 13.9 % for horizontal soil resistance force and vertical soil resistance respectively, 4.62 % for soil disturbance width, and 11.64 % for soil backfilling height between the experiment and simulation. These findings provide a theoretical foundation to enhance working efficiency and optimized design of furrow openers for fertilizer application in paddy fields, thereby contributing to improved agricultural productivity in rice cultivation.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109549"},"PeriodicalIF":7.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526007","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}

{"title":"A data-driven approach to the processing of sniffer-based gas emissions data from dairy cattle","authors":"Peter Løvendahl ,&nbsp;Viktor Milkevych ,&nbsp;Rikke Krogh Nielsen ,&nbsp;Martin Bjerring ,&nbsp;Coralia Manzanilla-Pech ,&nbsp;Kresten Johansen ,&nbsp;Gareth F Difford ,&nbsp;Trine M Villumsen","doi":"10.1016/j.compag.2024.109559","DOIUrl":"10.1016/j.compag.2024.109559","url":null,"abstract":"<div><div>“Sniffers” record methane (CH₄) emissions from cows visiting milking robots, providing gas concentration data. These instruments have infrared carbon dioxide (CO₂) and CH₄ sensors, an air pump, and a data logger. In this study, a process for the synchronization of sniffer emissions data with cow identification (ID) data and records from automatic milking systems (AMSs) was developed. The process enables the extraction of gas phenotypes for genetic analysis. It involves the calculation of intermediate control variables to account for time drift in data loggers, sensor calibration drift, and background concentration fluctuations, and the condensation of data from each milking visit into a single datapoint. The process was developed and assessed with research station data from three groups of approximately 70 cows, each accessing one AMS unit over a 2-month period. Raw emissions data, including clock times, from CH₄ and CO₂ channels were recorded every second. They were synchronized with the AMS data using specific events occurring in the CH₄ or CO₂ channel at the beginning or end of each milking event. The synchronized data were divided into non-milking (baseline, ambient gas concentrations) and cow ID–linked milking (cow emissions) sets. The non-milking periods varied in duration from a few seconds to hours, and some were interrupted by unrecorded events. Baseline values were extracted after the filtering of non-milking period data against unrecorded events (e.g., washing, feed-only sessions) and the use of a small fractile as the baseline estimate. At the beginning of each milking event, 30–45 s were required for the CH₄ and CO₂ concentrations to reach stable high levels, and most events lasted at least 5 min. Accordingly, a restricted recording window of 30–300 s, which excluded the initial unstable period while retaining data from the majority of milking events, was established. Gas concentrations significantly exceeding the baseline were selected as responses to ensure that only data obtained when the cows’ heads were sufficiently close to the sniffer air inlets were included. The mean value of the selected records was used as the response phenotype for each milking event. The concentration phenotypes showed moderate to high repeatability, but the CH₄:CO₂ ratio had only moderate repeatability. The pipeline developed in this study enables the effective extraction of baseline-adjusted emissions phenotypes from sniffer data obtained in milking robots.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109559"},"PeriodicalIF":7.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526003","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}

{"title":"YOLO-CG-HS: A lightweight spore detection method for wheat airborne fungal pathogens","authors":"Tao Cheng ,&nbsp;Dongyan Zhang ,&nbsp;Chunyan Gu ,&nbsp;Xin-Gen Zhou ,&nbsp;Hongbo Qiao ,&nbsp;Wei Guo ,&nbsp;Zhen Niu ,&nbsp;Jiyuan Xie ,&nbsp;Xue Yang","doi":"10.1016/j.compag.2024.109544","DOIUrl":"10.1016/j.compag.2024.109544","url":null,"abstract":"<div><div>The rapid, accurate and real-time online detection of spore concentration of various airborne pathogens in field crops is of great significance in guiding agricultural producers scientifically, enabling them to forecast disease development and implement timely preventive and control measures. This study presents a quantitative spore detection method for two prevalent wheat airborne fungal diseases using the YOLO-CG-HS lightweight model. Initially, the lightweight Context Guided module (CG) is integrated into the original Backbone of YOLOv5s to enhance the capture of global and edge information in spore images. Subsequently, the High-level Screening-feature Pyramid Networks (HS-FPN) module is incorporated into the Head to better integrate multi-scale feature information of spores, thereby improving the model’s detection performance and ability to capture spore micro-targets. The model’s robustness is then tested across various scenarios, including different shapes, densities, and complex backgrounds. Results indicate that the inclusion of both the CG module and the HS-FPN module into the original baseline model significantly reduces the number of model parameters to only 1.21 M. The model’s average precision (mAP) stands at 95.9 %, with an FPS of 152.5, maintaining performance levels similar to the original model. Moreover, the designed model effectively addresses the challenge of identifying difficult and missed cases resulting from spore adhesion and overlap in various airborne wheat diseases. The YOLO-CG-HS lightweight model developed in this study accurately detects various types of pathogen spores while balancing parameters, efficiency, and accuracy. This offers crucial technical support for the model migration and application of low-cost and high-precision embedded field spore capture instruments.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109544"},"PeriodicalIF":7.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526893","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}

{"title":"Recent advances on highly sensitive plasmonic nanomaterial enabled sensors for the detection of agrotoxins: Current progress and future perspective","authors":"Amruta Shelar ,&nbsp;Sanyukta Salve ,&nbsp;Harshali Shende ,&nbsp;Deepak Mehta ,&nbsp;Manohar Chaskar ,&nbsp;Shivraj Hariram Nile ,&nbsp;Rajendra Patil","doi":"10.1016/j.compag.2024.109545","DOIUrl":"10.1016/j.compag.2024.109545","url":null,"abstract":"<div><div>Agrotoxins, such as agrochemical residues and mycotoxins, pollute the environment and have potentially adverse effects on human life and ecosystem. Excessive use of agrochemicals such as pesticides and fertilizers may increase crop production, however, leads to toxic residue accumulation on farms. Concomitantly, the emergence of various fungus-associated toxins in fields can have adverse effects on human health. Currently, there are few methods available for monitoring these harmful substances, making it imperative to develop a sensor that can detect and quantify agrotoxin residues in the environment efficiently, rapidly, and on-site. Designing sensors based on nanotechnology approaches is a recommended approach for detect and quantify agrotoxin residues in the environment. NMs enhanced surface plasmon resonance (SPR) sensors have been developed and employed as useful tools to sense difficult-to-detect compounds in various fields. Therefore, nanosensors based on SPR will be a promising approach to detect agrotoxin residues because they are easy to use, don’t require labelling, are miniaturized, have high specificity and sensitivity, and are available with real-time measurement capabilities. However, currently, there is a gap in the literature and a lack of comprehensive reviews on recent advances in NMs for the design of SPR nanosensors to detect of agrotoxin residues in soil. The objective of this review is to provide an overview of NMs for SPR nanosensors used in detecting agrotoxins. The review also discusses the challenges associated with agrotoxin detection and the future prospects of using nanosensors for this purpose.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109545"},"PeriodicalIF":7.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526752","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}

{"title":"Numerical simulation analysis and experimental research on liquid sloshing in herbicide tank of the plant protection UAV","authors":"Xiaobo Sun ,&nbsp;Mengchen Cai ,&nbsp;Longhui Niu ,&nbsp;Qi Wang ,&nbsp;Wenqi Zhou ,&nbsp;Han Tang ,&nbsp;Jinwu Wang","doi":"10.1016/j.compag.2024.109532","DOIUrl":"10.1016/j.compag.2024.109532","url":null,"abstract":"<div><div>The sloshing of the liquid inside the herbicide tank of a plant-protection unmanned aerial vehicle (UAV) is a key factor affecting the stability and safety of UAV operations. To address this issue, this study analyzed the sloshing behavior of the liquid by employing flight tests and simulations to reveal the intrinsic characteristics of the sloshing liquid. An attitude monitoring system for the herbicide tank was constructed to collect acceleration signals during the plant protection UAV operation, and a corresponding model was developed to conduct simulations. The results indicate that the simulations based on the finite volume method were consistent with the experimental results, with the correlation coefficients of the captured free-surface curves exceeding 0.89. During the acceleration stage, the liquid exhibited periodic swaying. In the sudden stop stage, the free surface overturned, generating a large number of bubbles and vortices inside the liquid. This changed the direction of liquid flow, sloshing force, and sloshing mode of the free liquid surface, with the accumulation and collision of the liquid at the wall and corners becoming more significant. Under the turning condition, the liquid transitioned from counterclockwise swirling to swaying, but the amplitude of the liquid sloshing was reduced compared with the sudden stop condition. In summary, this study explored the sloshing process in the herbicide tank of a plant protection UAV based on sensor technology and computer numerical simulations, providing insights for the optimization and upgrading of equipment for plant protection UAV.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109532"},"PeriodicalIF":7.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526002","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}

{"title":"Online optimization of adjustable settings for agricultural machinery assisted by digital twin","authors":"Dafang Guo ,&nbsp;Linze Wang ,&nbsp;Yuefeng Du ,&nbsp;Zhikang Wu ,&nbsp;Weiran Zhang ,&nbsp;Qiao Zhi ,&nbsp;Ruofei Ma","doi":"10.1016/j.compag.2024.109504","DOIUrl":"10.1016/j.compag.2024.109504","url":null,"abstract":"<div><div>To address the complex and variable agricultural production, the adjustable settings on agricultural machinery have become increasingly numerous. However, determining the most applicable set of settings from thousands of possible combinations has emerged as a new challenge, one that is difficult to achieve through traditional experience-based decision-making and feedback control. This study analyzed the characteristics of the online optimization problem for adjustable settings in agricultural machinery, framing it as a single-objective cost optimization problem with a continuous feasible region and multi-modality. By introducing Digital Twin (DT) technology, a DT-assisted online optimization method (DTAOO) is proposed to search for the optimal set of setting. Specifically, DTAOO consists of two parts. One part involves the building of the DT, using an ensemble modeling combined with data augmentation to quickly establish and reconstruct the DT based on small sample data collected from physical space. The other part is the DT-assisted evolutionary algorithm (DTAEA), which employs the DT to predictively evaluate candidate solutions in a virtual space. This assists the evolutionary algorithm in searching for the most promising candidate solutions. In numerical experiments, the performance of DTAOO was evaluated through a series of benchmark problems and compared with other representative peer algorithms. Experimental results show that DTAOO achieved better results than peer algorithms on some complex benchmark problems. On multi-peak benchmark tests with uncertainty, DTAOO demonstrated a significant advantage. By applying DTAOO to optimize the settings related to the threshing process of a corn combine harvester, the grain breakage rate was reduced and working efficiency was improved, demonstrating the practical applicability of DTAOO. This study contributes to searching the optimal set of adjustable settings for agricultural machinery in complex production environments, offering the potential to improve production quality and efficiency without additional costs, and providing a reference for the operation, optimization and control of intelligent agricultural production systems.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109504"},"PeriodicalIF":7.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526004","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}

{"title":"High-throughput spike detection and refined segmentation for wheat Fusarium Head Blight in complex field environments","authors":"Qiong Zhou ,&nbsp;Ziliang Huang ,&nbsp;Liu Liu ,&nbsp;Fenmei Wang ,&nbsp;Yue Teng ,&nbsp;Haiyun Liu ,&nbsp;Youhua Zhang ,&nbsp;Rujing Wang","doi":"10.1016/j.compag.2024.109552","DOIUrl":"10.1016/j.compag.2024.109552","url":null,"abstract":"<div><div>Fusarium Head Blight (FHB) is a devastating disease of wheat worldwide. It is an explosive epidemic disease that can severely reduce or even fail wheat production. Estimating the disease ear rate and disease severity is crucial for effective plant protection. Manual assessment is labor-intensive and time-consuming. Accurately and quickly segmenting wheat ears and areas affected by Fusarium head blight (FHB) in complex field environments is essential for quantitative assessment of wheat trait phenotypes and FHB in wheat plants. This paper presents DeepFHB, an automated method for efficiently detecting, locating, and segmenting dense wheat spikes and diseased areas in digital images captured under natural field conditions. The experiment consists of three steps:Firstly, the process begins by generating initial coarse-grained mask predictions at lower resolutions to provide a rough segmentation. Secondly, a quadtree-based method is employed to identify and refine multi-scale inconsistent regions. Finally, a transformer-based refinement network is introduced to predict highly accurate instance segmentation masks. The results demonstrate that the DeepFHB algorithm outperforms traditional methods in detecting and segmenting diseased areas. Our DeepFHB model achieves state-of-the-art single-model results of 64.408 box AP and 64.966 mask AP on the FHB-SA dataset. This study is capable of rapidly and accurately segmenting wheat spikes and wheat scab lesions in agricultural scenarios with high field density, high crop occlusion, and high background interference. This provides a foundation for subsequent targeted research to assist agricultural workers in assessing the severity of wheat diseases.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109552"},"PeriodicalIF":7.7,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142446195","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}