Pub Date : 2024-03-18DOI: 10.1109/TAFE.2024.3369995
Muhammad Hamza Asad;Babalola Ekunayo-oluwabami Oreoluwa;Abdul Bais
Soil Organic Matter (SOM) is a key component for sustainable agriculture planning and soil management. Nutrient analysis, spectroscopy and digital soil imaging are commonly used to estimate SOM in a controlled lab setting. These methods are accurate, but the controlled lab setting is not scalable. For scalability, high-resolution satellite imagery is widely employed. However, special conditions of the Canadian Prairies, like harsh weather and crop residue cover, pose significant challenges in getting the spectral signatures of bare soil. To overcome these challenges, this paper presents a novel methodology that explores the prospects of using high-resolution ground images acquired under Uncontrolled Field Conditions (UFC) for SOM estimation. The developed methodology first extracts bare soil from images using deep learning methods. As the image samples are acquired under UFC, variable ambient illumination influences soil colour. To counter this, in the second step, we propose unsupervised colour constancy to mitigate the effects of variable ambient lighting conditions. In the third step, colour space and texture features are extracted to estimate SOM. We compare our proposed method with the state-of-the-art (SOTA) SOM estimation methods. We also performed an ablation study to compare the results of the SOTA with and without the addition of the colour constancy block. With the developed methodology, our bare soil segmentation model achieves a mean intersection over union value of 0.8134. Similarly, with the colour constancy methods applied on bare soil segmented images, our proposed method improves the $R^{2}$ score by more than 30% with respect to the SOTA.
土壤有机质（SOM）是可持续农业规划和土壤管理的关键组成部分。养分分析、光谱分析和数字土壤成像通常用于在受控实验室环境中估算土壤有机质。这些方法都很精确，但受控实验室环境无法扩展。为提高可扩展性，高分辨率卫星图像被广泛采用。然而，加拿大草原的特殊条件，如恶劣的天气和作物残茬覆盖，给获取裸露土壤的光谱特征带来了巨大挑战。为了克服这些挑战，本文提出了一种新颖的方法，探索使用在非受控野外条件（UFC）下获取的高分辨率地面图像进行 SOM 估算的前景。所开发的方法首先使用深度学习方法从图像中提取裸露土壤。由于图像样本是在 UFC 条件下获取的，不同的环境光照会影响土壤颜色。为此，在第二步中，我们提出了无监督色彩恒定法，以减轻环境光照条件变化的影响。第三步，提取色彩空间和纹理特征来估计 SOM。我们将我们提出的方法与最先进的（SOTA）SOM 估算方法进行了比较。我们还进行了一项消融研究，以比较添加和未添加色彩恒定块的 SOTA 结果。利用所开发的方法，我们的裸土分割模型达到了 0.8134 的平均交集大于联合值。同样，在裸土分割图像上应用色彩恒定方法后，我们提出的方法比 SOTA 提高了 30% 以上的 R^{2}$ 分数。
{"title":"Mapping Soil Organic Matter Under Field Conditions","authors":"Muhammad Hamza Asad;Babalola Ekunayo-oluwabami Oreoluwa;Abdul Bais","doi":"10.1109/TAFE.2024.3369995","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3369995","url":null,"abstract":"Soil Organic Matter (SOM) is a key component for sustainable agriculture planning and soil management. Nutrient analysis, spectroscopy and digital soil imaging are commonly used to estimate SOM in a controlled lab setting. These methods are accurate, but the controlled lab setting is not scalable. For scalability, high-resolution satellite imagery is widely employed. However, special conditions of the Canadian Prairies, like harsh weather and crop residue cover, pose significant challenges in getting the spectral signatures of bare soil. To overcome these challenges, this paper presents a novel methodology that explores the prospects of using high-resolution ground images acquired under Uncontrolled Field Conditions (UFC) for SOM estimation. The developed methodology first extracts bare soil from images using deep learning methods. As the image samples are acquired under UFC, variable ambient illumination influences soil colour. To counter this, in the second step, we propose unsupervised colour constancy to mitigate the effects of variable ambient lighting conditions. In the third step, colour space and texture features are extracted to estimate SOM. We compare our proposed method with the state-of-the-art (SOTA) SOM estimation methods. We also performed an ablation study to compare the results of the SOTA with and without the addition of the colour constancy block. With the developed methodology, our bare soil segmentation model achieves a mean intersection over union value of 0.8134. Similarly, with the colour constancy methods applied on bare soil segmented images, our proposed method improves the \u0000<inline-formula><tex-math>$R^{2}$</tex-math></inline-formula>\u0000 score by more than 30% with respect to the SOTA.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 1","pages":"138-150"},"PeriodicalIF":0.0,"publicationDate":"2024-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140544190","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}
Pub Date : 2024-03-10DOI: 10.1109/TAFE.2024.3392985
Valerio F. Annese;Valerio Galli;Giulia Coco;Mario Caironi
Traditional energy sources, such as alkaline batteries, cannot be in direct contact with food due to health hazards. However, a recently developed battery, constituted only by food-grade materials, overcomes this limitation as it can be used in direct contact with food without any contamination risk. In this work, we assess the feasibility of adopting an edible battery to power up traditional electronics for sensors-on-food or sensors-in-package applications. The feasibility study is divided into three main analyses. First, the lifetime of the battery against multiple charge–discharge cycles is assessed. In our experiments, the battery maintains a capacity of ∼ 6 μA·h after 100 cycles. Then, the study progressed to resistive sensors. As a test case, we demonstrated that data obtained from thermistors and photoresistors powered up by the battery have a cross-correlation coefficient > 0.99 with respect to using a traditional power supply as the energy source. Finally, the edible battery is successfully used to power operational-amplifier -based circuits performing amplification and filtering. This study indicates that, although more research is necessary to enhance the battery's performance, edible batteries represent a feasible alternative for supplying power for a limited time to basic low-power circuits.
{"title":"Organic Nontoxic Rechargeable Batteries in Food Packaging: A Feasibility Study","authors":"Valerio F. Annese;Valerio Galli;Giulia Coco;Mario Caironi","doi":"10.1109/TAFE.2024.3392985","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3392985","url":null,"abstract":"Traditional energy sources, such as alkaline batteries, cannot be in direct contact with food due to health hazards. However, a recently developed battery, constituted only by food-grade materials, overcomes this limitation as it can be used in direct contact with food without any contamination risk. In this work, we assess the feasibility of adopting an edible battery to power up traditional electronics for sensors-on-food or sensors-in-package applications. The feasibility study is divided into three main analyses. First, the lifetime of the battery against multiple charge–discharge cycles is assessed. In our experiments, the battery maintains a capacity of ∼ 6 μA·h after 100 cycles. Then, the study progressed to resistive sensors. As a test case, we demonstrated that data obtained from thermistors and photoresistors powered up by the battery have a cross-correlation coefficient > 0.99 with respect to using a traditional power supply as the energy source. Finally, the edible battery is successfully used to power operational-amplifier -based circuits performing amplification and filtering. This study indicates that, although more research is necessary to enhance the battery's performance, edible batteries represent a feasible alternative for supplying power for a limited time to basic low-power circuits.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 2","pages":"252-258"},"PeriodicalIF":0.0,"publicationDate":"2024-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142430873","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}
Pub Date : 2024-03-06DOI: 10.1109/TAFE.2024.3385610
Vamsee Krishna Bodasingi;Bakul Rao;Harish K. Pillai
Soil moisture and electrical conductivity (EC) measurements are vital for improving water usage efficiency and monitoring soil salinization. Water stress and groundwater salinity are the twin problems associated with excessive groundwater extraction and irrigation with saline water. Small and marginal farmers are the majority in India and require low-cost sensors for measuring soil moisture and EC. This work evaluates the performance of a low-cost capacitance-resistive sensor for soil moisture and EC measurement in different soil types. The results show that bulk EC and bulk density of the soils moderately affect the sensor output (oscillation frequency) for moisture measurement. The multilinear regression models developed using experimental data for moisture have an R 2 of 0.93, which is at par with the commercial sensors reviewed by researchers. Bulk EC variation with moisture showed significant variation with soil type due to differences in the salinity levels of soil samples. Therefore, the field devices require standard EC thresholds for salinity similar to laboratory standards.
{"title":"Laboratory Evaluation of a Low-Cost Soil Moisture and EC Sensor in Different Soil Types","authors":"Vamsee Krishna Bodasingi;Bakul Rao;Harish K. Pillai","doi":"10.1109/TAFE.2024.3385610","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3385610","url":null,"abstract":"Soil moisture and electrical conductivity (EC) measurements are vital for improving water usage efficiency and monitoring soil salinization. Water stress and groundwater salinity are the twin problems associated with excessive groundwater extraction and irrigation with saline water. Small and marginal farmers are the majority in India and require low-cost sensors for measuring soil moisture and EC. This work evaluates the performance of a low-cost capacitance-resistive sensor for soil moisture and EC measurement in different soil types. The results show that bulk EC and bulk density of the soils moderately affect the sensor output (oscillation frequency) for moisture measurement. The multilinear regression models developed using experimental data for moisture have an R\u0000<sup>2</sup>\u0000 of 0.93, which is at par with the commercial sensors reviewed by researchers. Bulk EC variation with moisture showed significant variation with soil type due to differences in the salinity levels of soil samples. Therefore, the field devices require standard EC thresholds for salinity similar to laboratory standards.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 2","pages":"276-283"},"PeriodicalIF":0.0,"publicationDate":"2024-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142430800","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}
Pub Date : 2024-03-05DOI: 10.1109/TAFE.2024.3366335
Rekha Raja
Designing software architectures for autonomous robots for agricultural contexts is a demanding and difficult job due to the requirement to monitor numerous sensors and actuators, as well as autonomous decision-making in unpredictable, unexpected scenarios. Depending on the essential requirements of a robotic device for agricultural usage, robot software architecture is created differently. Since no single software architecture exists for all applications, extensive knowledge of the various software architectures for robots is needed when creating your own robotic architecture or selecting one from a number of existing architectures. As a result, this article provides a comprehensive history of software architecture and its application in the agricultural domain along with a chronology of how software design has evolved over time. We provide several case studies to understand the importance of application of software architecture in agriculture and food industry and how to choose the best architecture for agricultural tasks. Finally, this article discusses the open obstacles and difficulties that must be addressed in order to ensure more advancements in the development of robot architecture for agricultural applications.
{"title":"Software Architecture for Agricultural Robots: Systems, Requirements, Challenges, Case Studies, and Future Perspectives","authors":"Rekha Raja","doi":"10.1109/TAFE.2024.3366335","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3366335","url":null,"abstract":"Designing software architectures for autonomous robots for agricultural contexts is a demanding and difficult job due to the requirement to monitor numerous sensors and actuators, as well as autonomous decision-making in unpredictable, unexpected scenarios. Depending on the essential requirements of a robotic device for agricultural usage, robot software architecture is created differently. Since no single software architecture exists for all applications, extensive knowledge of the various software architectures for robots is needed when creating your own robotic architecture or selecting one from a number of existing architectures. As a result, this article provides a comprehensive history of software architecture and its application in the agricultural domain along with a chronology of how software design has evolved over time. We provide several case studies to understand the importance of application of software architecture in agriculture and food industry and how to choose the best architecture for agricultural tasks. Finally, this article discusses the open obstacles and difficulties that must be addressed in order to ensure more advancements in the development of robot architecture for agricultural applications.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 1","pages":"125-137"},"PeriodicalIF":0.0,"publicationDate":"2024-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140544156","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}
Pub Date : 2024-03-05DOI: 10.1109/TAFE.2024.3365202
Dipak Hrishi Das;Sourav Dey Roy;Priya Saha;Mrinal Kanti Bhowmik
With the blooming interest in computer vision-based technologies for future automation of food producers, there is a need for incorporating an automatic bruise detection module in robotic apple harvesting because of decreasing accessibility and growing labor costs. Although numerous studies have been published for automatic quality inspection of fruit and other agricultural products, there is a lack of publicly available image-based datasets for quality inspection/automatic detection of bruises. Toward the aim of developing a bruise detection system for apple harvesting, especially at night time, this article describes the designing issues (i.e., protocol) and creation of a new infrared imaging-based dataset titled “TU-IR Apple Image Dataset,” which contains 1375 infrared images of apples defining four major categories of bruises (i.e., fresh, slight, moderate, and severe). Along with the infrared images, ground truths (in the form of binary masks) and measurements of suspicious bruised regions are defined. This study also investigates the efficiency of infrared imaging technology for automatic bruise detection in apples by performing an analysis of temperature-based, intensity-based, texture-based, shape-based, and deep convolutional neural network-based features. The classification performance was evaluated using eight different feature sets. Based on the experimental results, considering the most outer-performed classifier, deep convolutional neural networks as a fixed feature extraction method were found to provide the highest prediction performance for discriminating between fresh and three categories of bruises in apples with an average accuracy, specificity, and sensitivity of 93.87%, 80.57%, and 92.02%, respectively.
{"title":"TU-IR Apple Image Dataset: Benchmarking, Challenges, and Asymmetric Characterization for Bruise Detection in Application of Automatic Harvesting","authors":"Dipak Hrishi Das;Sourav Dey Roy;Priya Saha;Mrinal Kanti Bhowmik","doi":"10.1109/TAFE.2024.3365202","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3365202","url":null,"abstract":"With the blooming interest in computer vision-based technologies for future automation of food producers, there is a need for incorporating an automatic bruise detection module in robotic apple harvesting because of decreasing accessibility and growing labor costs. Although numerous studies have been published for automatic quality inspection of fruit and other agricultural products, there is a lack of publicly available image-based datasets for quality inspection/automatic detection of bruises. Toward the aim of developing a bruise detection system for apple harvesting, especially at night time, this article describes the designing issues (i.e., protocol) and creation of a new infrared imaging-based dataset titled “TU-IR Apple Image Dataset,” which contains 1375 infrared images of apples defining four major categories of bruises (i.e., fresh, slight, moderate, and severe). Along with the infrared images, ground truths (in the form of binary masks) and measurements of suspicious bruised regions are defined. This study also investigates the efficiency of infrared imaging technology for automatic bruise detection in apples by performing an analysis of temperature-based, intensity-based, texture-based, shape-based, and deep convolutional neural network-based features. The classification performance was evaluated using eight different feature sets. Based on the experimental results, considering the most outer-performed classifier, deep convolutional neural networks as a fixed feature extraction method were found to provide the highest prediction performance for discriminating between fresh and three categories of bruises in apples with an average accuracy, specificity, and sensitivity of 93.87%, 80.57%, and 92.02%, respectively.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 1","pages":"105-124"},"PeriodicalIF":0.0,"publicationDate":"2024-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140544191","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}
Consumer behaviors and habits in food choices impact their physical health and have implications for climate change and global warming. Efficient food image recognition can assist individuals in making more environmentally friendly and healthier dietary choices using end devices, such as smartphones. Simultaneously, it can enhance the efficiency of server-side training, thereby reducing carbon emissions. We propose a lightweight deep neural network named Global Shuffle Net (GSNet) that can efficiently recognize food images. In GSNet, we develop a novel convolution method called global shuffle convolution, which captures the dependence between long-range pixels. Merging global shuffle convolution with classic local convolution yields a framework that works as the backbone of GSNet. Through GSNet's ability to capture the dependence between long-range pixels at the start of the network, by restricting the number of layers in the middle and rear, the parameters and floating operation operations (FLOPs) can be minimized without compromising the performance, thus permitting a lightweight goal to be achieved. Experimental results on four popular food recognition datasets demonstrate that our approach achieves state-of-the-art performance with higher accuracy and fewer FLOPs and parameters. For example, in comparison to the current state-of-the-art model of MobileViTv2, GSNet achieved 87.9% accuracy of the top-1 level on the Eidgenössische Technische Hochschule Zürich (ETHZ) Food-101 dataset with 28% reduction in the parameters, 37% reduction in the FLOPs, but a 0.7% more accuracy.
{"title":"Lightweight Food Image Recognition With Global Shuffle Convolution","authors":"Guorui Sheng;Weiqing Min;Tao Yao;Jingru Song;Yancun Yang;Lili Wang;Shuqiang Jiang","doi":"10.1109/TAFE.2024.3386713","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3386713","url":null,"abstract":"Consumer behaviors and habits in food choices impact their physical health and have implications for climate change and global warming. Efficient food image recognition can assist individuals in making more environmentally friendly and healthier dietary choices using end devices, such as smartphones. Simultaneously, it can enhance the efficiency of server-side training, thereby reducing carbon emissions. We propose a lightweight deep neural network named Global Shuffle Net (GSNet) that can efficiently recognize food images. In GSNet, we develop a novel convolution method called global shuffle convolution, which captures the dependence between long-range pixels. Merging global shuffle convolution with classic local convolution yields a framework that works as the backbone of GSNet. Through GSNet's ability to capture the dependence between long-range pixels at the start of the network, by restricting the number of layers in the middle and rear, the parameters and floating operation operations (FLOPs) can be minimized without compromising the performance, thus permitting a lightweight goal to be achieved. Experimental results on four popular food recognition datasets demonstrate that our approach achieves state-of-the-art performance with higher accuracy and fewer FLOPs and parameters. For example, in comparison to the current state-of-the-art model of MobileViTv2, GSNet achieved 87.9% accuracy of the top-1 level on the Eidgenössische Technische Hochschule Zürich (ETHZ) Food-101 dataset with 28% reduction in the parameters, 37% reduction in the FLOPs, but a 0.7% more accuracy.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 2","pages":"392-402"},"PeriodicalIF":0.0,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142408757","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}
This article presents a wireless biosensor based on an organic electrochemical transistor, a low-power electronic system, with narrow band (NB)-Internet of Things (IoT)/Cat-M1 radio interface, and server with web interface. The biosensor, implanted in the plant stem, allows the in vivo evaluation of the concentration of nutrients dissolved as cations in the sap. The electronic circuit enables the real-time monitoring of the plant in the crop. The NB-IoT or Cat-M1 link, both available in the system-in-package device selected for the proposed system, ensures almost ubiquitous availability of the network link, without severe limitations on the data payload. The server stores in cloud the data obtained from the field, and a web interface enables the remote monitoring of the plant physiological mechanisms, with consumer devices, such as laptops or smartphones. The low power consumption of the biosensor allows more than three months of battery lifetime, adequate for most seasonal crops. With duty-cycle approach, more than one year of lifetime can be obtained for perennial crops, such as vineyards and orchards. Measurements on KCl solutions showed adequate sensor linearity up to 10-mM K $^+$ concentration, while those performed on a sap of kiwi vines are in agreement with data available in the literature. In vivo measurements carried out on cabbage show how the parameters of the sensor are affected by the circadian cycle. In day time, a reduction of cation concentration, due to water absorption for the photosynthesis and stomatal transpiration, is detected by the wireless-bioristor.
{"title":"A Wireless Biosensor Node for In Vivo and Real-Time Plant Monitoring in Precision Agriculture","authors":"Michele Caselli;Edoardo Graiani;Valentina Bianchi;Filippo Vurro;Manuele Bettelli;Giuseppe Tarabella;Ilaria De Munari;Michela Janni;Andrea Boni","doi":"10.1109/TAFE.2024.3386938","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3386938","url":null,"abstract":"This article presents a wireless biosensor based on an organic electrochemical transistor, a low-power electronic system, with narrow band (NB)-Internet of Things (IoT)/Cat-M1 radio interface, and server with web interface. The biosensor, implanted in the plant stem, allows the in vivo evaluation of the concentration of nutrients dissolved as cations in the sap. The electronic circuit enables the real-time monitoring of the plant in the crop. The NB-IoT or Cat-M1 link, both available in the system-in-package device selected for the proposed system, ensures almost ubiquitous availability of the network link, without severe limitations on the data payload. The server stores in cloud the data obtained from the field, and a web interface enables the remote monitoring of the plant physiological mechanisms, with consumer devices, such as laptops or smartphones. The low power consumption of the biosensor allows more than three months of battery lifetime, adequate for most seasonal crops. With duty-cycle approach, more than one year of lifetime can be obtained for perennial crops, such as vineyards and orchards. Measurements on KCl solutions showed adequate sensor linearity up to 10-mM K\u0000<inline-formula><tex-math>$^+$</tex-math></inline-formula>\u0000 concentration, while those performed on a sap of kiwi vines are in agreement with data available in the literature. In vivo measurements carried out on cabbage show how the parameters of the sensor are affected by the circadian cycle. In day time, a reduction of cation concentration, due to water absorption for the photosynthesis and stomatal transpiration, is detected by the wireless-bioristor.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 2","pages":"268-275"},"PeriodicalIF":0.0,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517755","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142430874","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}
Pub Date : 2024-02-28DOI: 10.1109/TAFE.2024.3352480
Sean J. Harkin;Tomás Crotty;John Warren;Conor Shanahan;Edward Jones;Martin Glavin;Dallan Byrne
Establishing and maintaining farmland geometric boundaries is crucial to increasing agricultural productivity. Accurate field boundaries enable farm machinery contractors and other farm stakeholders to calculate charges, costs and to examine machinery performance. Field segmentation is the process by which agricultural field plots are geofenced into their individual field geometric boundaries. This paper presents a novel coordinate-based method to perform trajectory segmentation and field boundary detection from a tractor towing an implement. The main contribution of this research is a practical, robust algorithm which can solve for challenging field-to-field segmentation cases where the operator engages the towed implement continuously across several fields. The algorithm first isolates raw machinery trajectory data into unique job sites by using a coarse filter on geolocation data and implement power-take off activation. Next, the coordinate data is plotted and image processing techniques are applied to erode any pathway(s) that may present in job sites with adjacent working fields. Georeferenced time series tractor and implement data were aggregated from a five-month-long measurement campaign of a silage baling season in Galway, Ireland. The algorithm was validated against two unique machinery implement datasets, which combined, contain a mixture of 296 road-to-field and 31 field-to-field cases. The results demonstrate that the algorithm achieves an accuracy of 100% on a baler implement dataset and 98.84% on a mower implement dataset. The proposed algorithm is deterministic and does not require any additional labor, land traversal or aerial surveillance to produce results with accuracy metrics registering above 98%.
{"title":"Field-to-Field Coordinate-Based Segmentation Algorithm on Agricultural Harvest Implements","authors":"Sean J. Harkin;Tomás Crotty;John Warren;Conor Shanahan;Edward Jones;Martin Glavin;Dallan Byrne","doi":"10.1109/TAFE.2024.3352480","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3352480","url":null,"abstract":"Establishing and maintaining farmland geometric boundaries is crucial to increasing agricultural productivity. Accurate field boundaries enable farm machinery contractors and other farm stakeholders to calculate charges, costs and to examine machinery performance. Field segmentation is the process by which agricultural field plots are geofenced into their individual field geometric boundaries. This paper presents a novel coordinate-based method to perform trajectory segmentation and field boundary detection from a tractor towing an implement. The main contribution of this research is a practical, robust algorithm which can solve for challenging field-to-field segmentation cases where the operator engages the towed implement continuously across several fields. The algorithm first isolates raw machinery trajectory data into unique job sites by using a coarse filter on geolocation data and implement power-take off activation. Next, the coordinate data is plotted and image processing techniques are applied to erode any pathway(s) that may present in job sites with adjacent working fields. Georeferenced time series tractor and implement data were aggregated from a five-month-long measurement campaign of a silage baling season in Galway, Ireland. The algorithm was validated against two unique machinery implement datasets, which combined, contain a mixture of 296 road-to-field and 31 field-to-field cases. The results demonstrate that the algorithm achieves an accuracy of 100% on a baler implement dataset and 98.84% on a mower implement dataset. The proposed algorithm is deterministic and does not require any additional labor, land traversal or aerial surveillance to produce results with accuracy metrics registering above 98%.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 1","pages":"91-104"},"PeriodicalIF":0.0,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10452413","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140544319","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}
Pub Date : 2024-02-26DOI: 10.1109/TAFE.2024.3359660
Divya Mishra;Sharon Chemweno;Ofer Hadar;Ofer Ben-Tovim;Naftali Lazarovitch;Jhonathan E. Ephrath
Root hairs are essential for nutrient uptake and plant-microbe interactions, playing a vital role in plant health and agricultural productivity. They extend from the surface of root cells, significantly increasing the root surface area, and constitute roughly 70% of the total root area. Given these advantages, detecting root hairs in scenes with low resolution is challenging. Therefore, we have proposed a study that utilizes unsupervised image super-resolution methods to reconstruct finer details for root hairs using the dataset captured from our novel scanning camera known as RootCam. RootCam is a fully automated tool designed for monitoring and capturing plant root images for different vision tasks for a more accurate representation of their morphology and root trait measurements. Root hair super-resolution proves to be a powerful tool for root biology and its applications in precision agriculture. To the best of the authors' knowledge, this research study is the first that mainly focuses on root hairs and their trait measurement improvement using super-resolution. By visualizing the rhizosphere in high-resolution detail, we are able to notice a significant improvement in bell-pepper plant root hair count from 7 to 12, total root length from 0.32 to 1 mm, and root hair density (number of root hairs/mm) from 2.7 to 4.63, as upscaling factors rise from 2 to 8, respectively, when compared with bicubic and contrastive learning semi-supervised remote sensing image super (CLSR) for super-resolution. Researchers and farmers can make informed decisions about nutrient placement, irrigation management, and crop selection, optimizing resource use efficiency and crop yields.
{"title":"Unsupervised Image Super-Resolution for Root Hair Enhancement and Improved Root Traits Measurements","authors":"Divya Mishra;Sharon Chemweno;Ofer Hadar;Ofer Ben-Tovim;Naftali Lazarovitch;Jhonathan E. Ephrath","doi":"10.1109/TAFE.2024.3359660","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3359660","url":null,"abstract":"Root hairs are essential for nutrient uptake and plant-microbe interactions, playing a vital role in plant health and agricultural productivity. They extend from the surface of root cells, significantly increasing the root surface area, and constitute roughly 70% of the total root area. Given these advantages, detecting root hairs in scenes with low resolution is challenging. Therefore, we have proposed a study that utilizes unsupervised image super-resolution methods to reconstruct finer details for root hairs using the dataset captured from our novel scanning camera known as RootCam. RootCam is a fully automated tool designed for monitoring and capturing plant root images for different vision tasks for a more accurate representation of their morphology and root trait measurements. Root hair super-resolution proves to be a powerful tool for root biology and its applications in precision agriculture. To the best of the authors' knowledge, this research study is the first that mainly focuses on root hairs and their trait measurement improvement using super-resolution. By visualizing the rhizosphere in high-resolution detail, we are able to notice a significant improvement in bell-pepper plant root hair count from 7 to 12, total root length from 0.32 to 1 mm, and root hair density (number of root hairs/mm) from 2.7 to 4.63, as upscaling factors rise from 2 to 8, respectively, when compared with bicubic and contrastive learning semi-supervised remote sensing image super (CLSR) for super-resolution. Researchers and farmers can make informed decisions about nutrient placement, irrigation management, and crop selection, optimizing resource use efficiency and crop yields.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 1","pages":"81-90"},"PeriodicalIF":0.0,"publicationDate":"2024-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140544168","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}
Pub Date : 2024-02-13DOI: 10.1109/TAFE.2024.3359463
Boon Siong Wee;Cheng Siong Chin;Anurag Sharma
Mushroom harvesting, specifically mushroom picking, is a labor-intensive and time-consuming activity. This article presents a literature survey of the design and evaluation of mushroom harvesting robots and technologies to address the labor-intensive and time-consuming task of manual mushroom harvesting. We classify and look at different classes of harvesting robots and technologies from the 1970s to 2022. We present the robot's overall system and capabilities, including sensors and control systems where available. We also summarized the advantages and disadvantages of each system and the performance metrics where data are available.
{"title":"Survey of Mushroom Harvesting Agricultural Robots and Systems Design","authors":"Boon Siong Wee;Cheng Siong Chin;Anurag Sharma","doi":"10.1109/TAFE.2024.3359463","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3359463","url":null,"abstract":"Mushroom harvesting, specifically mushroom picking, is a labor-intensive and time-consuming activity. This article presents a literature survey of the design and evaluation of mushroom harvesting robots and technologies to address the labor-intensive and time-consuming task of manual mushroom harvesting. We classify and look at different classes of harvesting robots and technologies from the 1970s to 2022. We present the robot's overall system and capabilities, including sensors and control systems where available. We also summarized the advantages and disadvantages of each system and the performance metrics where data are available.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 1","pages":"59-80"},"PeriodicalIF":0.0,"publicationDate":"2024-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140544192","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：