Pub Date : 2025-10-09DOI: 10.1109/TAFE.2025.3615014
{"title":"IEEE Circuits and Systems Society Information","authors":"","doi":"10.1109/TAFE.2025.3615014","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3615014","url":null,"abstract":"","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"C3-C3"},"PeriodicalIF":0.0,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11197716","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248068","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 : 2025-09-18DOI: 10.1109/TAFE.2025.3604870
Jiayi Xiong;Jianyang Gao;Libin Li;Jiayi Li;Lei Liu
To address the identification requirements for tobacco thinning targets and their growth points in automated tobacco thinning operations, a dual-model collaborative recognition approach integrating target detection and instance segmentation was proposed. First, for thinning target identification, a lightweight you only look once dilated dual-path network (YOLO-DDPNet) segmentation network was developed by integrating a DDPNet module into the YOLOv8 architecture. This network achieved a tobacco seedling segmentation accuracy of 98.7% (3.6% higher than YOLOv8n), enabling thinning target screening by comparing the segmentation mask areas of tobacco seedlings within a seedling hole. Second, for seedling growth point detection, the original C2f module in YOLOv8 was replaced with C3x while incorporating the SE attention mechanism and SPPCSPC multiscale feature fusion module to construct a lightweight YOLO-TGPD detection network. This network attained a growth point detection accuracy of 94.3% (8.2% higher than YOLOv8n). Notably, this study pioneered the synergistic use of segmentation and detection strategies to simultaneously complete thinning target screening and growth point detection. The proposed model outperformed advanced models (e.g., YOLOv9 and YOLOv11) on the tobacco seedling dataset, holding significant potential for advancing tobacco thinning automation technology.
{"title":"Dual-YOLO Network: Recognition of Thinning Targets and Growing Point for Tobacco Seedlings","authors":"Jiayi Xiong;Jianyang Gao;Libin Li;Jiayi Li;Lei Liu","doi":"10.1109/TAFE.2025.3604870","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3604870","url":null,"abstract":"To address the identification requirements for tobacco thinning targets and their growth points in automated tobacco thinning operations, a dual-model collaborative recognition approach integrating target detection and instance segmentation was proposed. First, for thinning target identification, a lightweight you only look once dilated dual-path network (YOLO-DDPNet) segmentation network was developed by integrating a DDPNet module into the YOLOv8 architecture. This network achieved a tobacco seedling segmentation accuracy of 98.7% (3.6% higher than YOLOv8n), enabling thinning target screening by comparing the segmentation mask areas of tobacco seedlings within a seedling hole. Second, for seedling growth point detection, the original C2f module in YOLOv8 was replaced with C3x while incorporating the SE attention mechanism and SPPCSPC multiscale feature fusion module to construct a lightweight YOLO-TGPD detection network. This network attained a growth point detection accuracy of 94.3% (8.2% higher than YOLOv8n). Notably, this study pioneered the synergistic use of segmentation and detection strategies to simultaneously complete thinning target screening and growth point detection. The proposed model outperformed advanced models (e.g., YOLOv9 and YOLOv11) on the tobacco seedling dataset, holding significant potential for advancing tobacco thinning automation technology.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"633-648"},"PeriodicalIF":0.0,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11173177","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248069","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}
Soil salinity is a critical factor influencing agricultural productivity and environmental sustainability, requiring precise monitoring tools. This article focuses on developing a frequency-dependent model to predict soil salinity based on electrical conductivity (EC) and volumetric water content (VWC). A dataset of 40 soil samples with varying levels of salinity and moisture, consisting of two soil types (sandy and clayey), was experimentally measured for EC in the frequency range of 10 to 295 MHz using EC spectroscopy measurement with the dielectric assessment kit–vector network analyzer) system. A new, more comprehensive frequency-dependent model is proposed, surpassing previous models that lacked frequency considerations. This modeling approach was conducted in stages: initially, a frequency-independent model for EC as a function of salinity and VWC was developed. Next, a frequency-dependent model was introduced. Finally, a comparison between pure sandy soil and a sandy–clay mixture led to the final model, which also incorporates effective porosity. The results of the proposed model, comparing measured and predicted values, provide a robust approach to accurately predict soil salinity. Findings demonstrate that the model can enhance salinity prediction accuracy, extending its applicability beyond agriculture to geological and hydrological applications in real-world scenarios.
{"title":"Soil Salinity Frequency-Dependent Prediction Model Using Electrical Conductivity Spectroscopy Measurement","authors":"Javad Jafaryahya;Rasool Keshavarz;Taro Kikuchi;Negin Shariati","doi":"10.1109/TAFE.2025.3602029","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3602029","url":null,"abstract":"Soil salinity is a critical factor influencing agricultural productivity and environmental sustainability, requiring precise monitoring tools. This article focuses on developing a frequency-dependent model to predict soil salinity based on electrical conductivity (EC) and volumetric water content (VWC). A dataset of 40 soil samples with varying levels of salinity and moisture, consisting of two soil types (sandy and clayey), was experimentally measured for EC in the frequency range of 10 to 295 MHz using EC spectroscopy measurement with the dielectric assessment kit–vector network analyzer) system. A new, more comprehensive frequency-dependent model is proposed, surpassing previous models that lacked frequency considerations. This modeling approach was conducted in stages: initially, a frequency-independent model for EC as a function of salinity and VWC was developed. Next, a frequency-dependent model was introduced. Finally, a comparison between pure sandy soil and a sandy–clay mixture led to the final model, which also incorporates effective porosity. The results of the proposed model, comparing measured and predicted values, provide a robust approach to accurately predict soil salinity. Findings demonstrate that the model can enhance salinity prediction accuracy, extending its applicability beyond agriculture to geological and hydrological applications in real-world scenarios.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"623-632"},"PeriodicalIF":0.0,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248064","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 : 2025-09-08DOI: 10.1109/TAFE.2025.3601808
Shubhajyoti Das;Pritam Bikram;Arindam Biswas;Vimalkumar C.;Parimal Sinha;Bhargab B. Bhattacharya
Leaf blast disease is a significant constraint in world-wide rice production systems, necessitating effective monitoring for optimized crop-yield management. Satellite-derived land-surface temperature data can be an essential input for detecting such a disease, as it plays a critical role in the pathogen’s development and spread. When combined with other environmental factors, such as humidity and leaf wetness, it serves as a key indicator of potential outbreaks. Vegetation and moisture indexes captured by the European Space Agency satellite Sentinel 2, have been used to analyze rice blast disease on a large scale. However, due to substantial geo-spatial and temporal variability, predicting disease occurrence remains a challenge. To address this gap, we propose a fusion of convolutional neural networks (CNN) and transformer-based models to reveal both local and global syndromes in images associated with the risk of rice blast disease. A novel multichannel attention mechanism within the CNN helps extract essential spectral information, where each RGB channel’s spatial intensity is leveraged to focus on critical details through multihead attention. The transformer network with dynamic tokenization and self-attention captures global information, enabling lightweight transformers to highlight discriminative global features. Dynamic tokenization selects tokens or patches based on attention factors, facilitating the extraction of important sequential information. The aggregated network output enhances the classification accuracy of leaf blast risk prediction while reducing computational complexity. The proposed approach outperforms existing models in spectral image analysis for predicting the spread of leaf blast disease.
{"title":"Transformer-Embedded Attentive CNN for Spectral Image Analysis of Rice Blast Syndromes","authors":"Shubhajyoti Das;Pritam Bikram;Arindam Biswas;Vimalkumar C.;Parimal Sinha;Bhargab B. Bhattacharya","doi":"10.1109/TAFE.2025.3601808","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3601808","url":null,"abstract":"Leaf blast disease is a significant constraint in world-wide rice production systems, necessitating effective monitoring for optimized crop-yield management. Satellite-derived land-surface temperature data can be an essential input for detecting such a disease, as it plays a critical role in the pathogen’s development and spread. When combined with other environmental factors, such as humidity and leaf wetness, it serves as a key indicator of potential outbreaks. Vegetation and moisture indexes captured by the European Space Agency satellite Sentinel 2, have been used to analyze rice blast disease on a large scale. However, due to substantial geo-spatial and temporal variability, predicting disease occurrence remains a challenge. To address this gap, we propose a fusion of convolutional neural networks (CNN) and transformer-based models to reveal both local and global syndromes in images associated with the risk of rice blast disease. A novel multichannel attention mechanism within the CNN helps extract essential spectral information, where each RGB channel’s spatial intensity is leveraged to focus on critical details through multihead attention. The transformer network with dynamic tokenization and self-attention captures global information, enabling lightweight transformers to highlight discriminative global features. Dynamic tokenization selects tokens or patches based on attention factors, facilitating the extraction of important sequential information. The aggregated network output enhances the classification accuracy of leaf blast risk prediction while reducing computational complexity. The proposed approach outperforms existing models in spectral image analysis for predicting the spread of leaf blast disease.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"615-622"},"PeriodicalIF":0.0,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248034","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 : 2025-09-04DOI: 10.1109/TAFE.2025.3600008
Yuzhu Zheng;Jun Qi;Yun Yang;Po Yang;Zhipeng Yuan
Aphids are among the most destructive pests that threaten global crop yields, harming crops through feeding and virus transmission. Accurate detection of aphids in fields is a crucial step in implementing sustainable agricultural pest management. However, the tiny size of aphids and the complex image background present significant challenges for accurate identification and classification for in-field detection. In response to the challenges, this study proposes a lightweight real-time object detection model, Aphid-YOLO (A-YOLO), for in-field aphid identification and counting. Specifically, a tiny path aggregation network with C2f-CG modules is proposed to enhance the detection ability of tiny objects while maintaining a low computational cost through efficiently fusing multilayer features. For model training, a normalized Wasserstein distance loss function is adopted to address the optimization challenges caused by the tiny size of aphids. In addition, an optimized data augmentation method, Mosaic9, is introduced to enrich training samples and positive supervised signals for addressing the classification challenge of tiny aphids. To validate the effectiveness of A-YOLO, this study conducts comprehensive experiments on an aphid detection dataset with images collected by hand-held devices from a complex field environment. Experimental results demonstrate that A-YOLO achieves outstanding detection efficiency, with an mAP@0.5 of 83.4%, an mAP@0.5:0.95 of 33.7%, an inference speed of 72 FPS, and a model size of 30.6 MB. Compared to the YOLOv8m model employing traditional Mosaic data augmentation, the proposed method improves mAP@0.5 by 5.8%, mAP@0.5:0.95 by 2.7%, increases inference speed by 5 FPS, and reduces model size by 38.4% .
{"title":"Aphid-YOLO: A Lightweight Detection Model for Real-Time Identification and Counting of Aphids in Complex Field Environments","authors":"Yuzhu Zheng;Jun Qi;Yun Yang;Po Yang;Zhipeng Yuan","doi":"10.1109/TAFE.2025.3600008","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3600008","url":null,"abstract":"Aphids are among the most destructive pests that threaten global crop yields, harming crops through feeding and virus transmission. Accurate detection of aphids in fields is a crucial step in implementing sustainable agricultural pest management. However, the tiny size of aphids and the complex image background present significant challenges for accurate identification and classification for in-field detection. In response to the challenges, this study proposes a lightweight real-time object detection model, Aphid-YOLO (A-YOLO), for in-field aphid identification and counting. Specifically, a tiny path aggregation network with C2f-CG modules is proposed to enhance the detection ability of tiny objects while maintaining a low computational cost through efficiently fusing multilayer features. For model training, a normalized Wasserstein distance loss function is adopted to address the optimization challenges caused by the tiny size of aphids. In addition, an optimized data augmentation method, Mosaic9, is introduced to enrich training samples and positive supervised signals for addressing the classification challenge of tiny aphids. To validate the effectiveness of A-YOLO, this study conducts comprehensive experiments on an aphid detection dataset with images collected by hand-held devices from a complex field environment. Experimental results demonstrate that A-YOLO achieves outstanding detection efficiency, with an mAP@0.5 of 83.4%, an mAP@0.5:0.95 of 33.7%, an inference speed of 72 FPS, and a model size of 30.6 MB. Compared to the YOLOv8m model employing traditional Mosaic data augmentation, the proposed method improves mAP@0.5 by 5.8%, mAP@0.5:0.95 by 2.7%, increases inference speed by 5 FPS, and reduces model size by 38.4% .","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"605-614"},"PeriodicalIF":0.0,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248082","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 : 2025-08-19DOI: 10.1109/TAFE.2025.3554512
Fazhan Tao;Dong Yang;Dayong Xu;Zhumu Fu
The tobacco industry attaches great importance to the development of slim cigarettes, and the content of stem sticks in slim cigarettes is extremely important to the quality of cigarettes. Therefore, in order to solve the problem of difficult detection of stem sticks in cut tobacco, a stem sticks detection algorithm in cut tobacco based on hyperspectral image technology combined with improved YOLOv8n is proposed. First, a principal component analysis method was used to process the hyperspectral image data to improve the differentiation between cut tobacco and stem sticks, and to construct the dataset. Second, the YOLOv8n algorithm was optimized to obtain the GMCM-YOLOv8n algorithm. Multiscale convolutional attention was introduced in the backbone network to capture detail information. Then, ghost convolution (GhostConv) was introduced to replace the regular convolution to simplify the network. M-BiFPN modules are proposed in neck networks as a way to improve the detection of small-sized stem sticks. The C2f module is also improved to obtain P-C2f with a view to reducing the model parameters and computational volume. Finally, the effectiveness of the GMCM-YOLOv8n algorithm is experimentally verified on self-constructed dataset. The results of the experiment showed that: the algorithm achieved a mean average precision of 93.9%, with parameters and floating point operations of 2.2 M and 6.2 G, respectively, and frames per second maintained at 73.5 fps. Compared with YOLOv8n, the proposed improved algorithm exhibited better comprehensive performance, which provided a valuable reference for realizing the task of quickly and accurately detecting the content of stem sticks in cut tobacco in practical production.
{"title":"Hyperspectral Images-Based Stem Sticks Signature Detection of Cut Tobacco Using Improved YOLOv8n Algorithm","authors":"Fazhan Tao;Dong Yang;Dayong Xu;Zhumu Fu","doi":"10.1109/TAFE.2025.3554512","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3554512","url":null,"abstract":"The tobacco industry attaches great importance to the development of slim cigarettes, and the content of stem sticks in slim cigarettes is extremely important to the quality of cigarettes. Therefore, in order to solve the problem of difficult detection of stem sticks in cut tobacco, a stem sticks detection algorithm in cut tobacco based on hyperspectral image technology combined with improved YOLOv8n is proposed. First, a principal component analysis method was used to process the hyperspectral image data to improve the differentiation between cut tobacco and stem sticks, and to construct the dataset. Second, the YOLOv8n algorithm was optimized to obtain the GMCM-YOLOv8n algorithm. Multiscale convolutional attention was introduced in the backbone network to capture detail information. Then, ghost convolution (GhostConv) was introduced to replace the regular convolution to simplify the network. M-BiFPN modules are proposed in neck networks as a way to improve the detection of small-sized stem sticks. The C2f module is also improved to obtain P-C2f with a view to reducing the model parameters and computational volume. Finally, the effectiveness of the GMCM-YOLOv8n algorithm is experimentally verified on self-constructed dataset. The results of the experiment showed that: the algorithm achieved a mean average precision of 93.9%, with parameters and floating point operations of 2.2 M and 6.2 G, respectively, and frames per second maintained at 73.5 fps. Compared with YOLOv8n, the proposed improved algorithm exhibited better comprehensive performance, which provided a valuable reference for realizing the task of quickly and accurately detecting the content of stem sticks in cut tobacco in practical production.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"591-604"},"PeriodicalIF":0.0,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248036","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 : 2025-08-07DOI: 10.1109/TAFE.2025.3585065
Pooja Garg;Anusha Mishra;Rameez Raja;Ahlad Kumar;Manjunath V. Joshi;Vinay S Palaparthy
In agricultural applications, traditional image and sensor-based methods for plant disease prediction face notable limitations. Image-based approaches often struggle with early-stage detection, while sensor-based methods prone to reliability issues due to potential system failures. This study addresses these challenges by integrating complementary data of the Jamun (Syzygium cumini) plant from Internet of Things (IoT)-enabled sensors and mobile-captured images to develop a hybrid machine learning (ML) model for early and accurate plant disease detection. The proposed model combines a multilayer perceptron (MLP) for processing numerical sensor inputs—ambient temperature, soil temperature, relative humidity, soil moisture, and leaf wetness duration—and a convolutional neural network (CNN) for analyzing leaf images labeled as leaf spot, anthracnose, or healthy. Outputs from the MLP and CNN concatenated and processed through an additional MLP to classify plant health effectively. Optimized with hidden layer configurations of 8-16-32-8 for the sensor-data MLP, 16--32-64-128_32-8-4 for the image-data CNN, and 4-3 layers for the final MLP, the model achieves a loss of 1% and an accuracy of 95%, outperforming state-of-the-art methods, such as DenseNet201-support vector machines (SVM) (87.23%) and gray level co-occurrence matrix-SVM (90%). Performance metrics demonstrate high precision (leaf spot: 0.93, anthracnose: 0.93, and healthy: 0.98), recall (leaf spot: 0.92, anthracnose: 0.95, and healthy: 0.96), and F1-scores (leaf spot: 0.92, anthracnose: 0.94, and healthy: 0.97). The model’s deployment on an Amazon Web Services cloud server enables real-time disease detection and classification, making it accessible for practical agricultural use. This sensor and image data integration offers a novel and robust solution to address the limitations of single-modality approaches.
{"title":"Multimodal Data Fusion by Integrating IoT-Enabled Sensors and Images for Jamun Crop Disease Detection With Machine Learning","authors":"Pooja Garg;Anusha Mishra;Rameez Raja;Ahlad Kumar;Manjunath V. Joshi;Vinay S Palaparthy","doi":"10.1109/TAFE.2025.3585065","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3585065","url":null,"abstract":"In agricultural applications, traditional image and sensor-based methods for plant disease prediction face notable limitations. Image-based approaches often struggle with early-stage detection, while sensor-based methods prone to reliability issues due to potential system failures. This study addresses these challenges by integrating complementary data of the Jamun (Syzygium cumini) plant from Internet of Things (IoT)-enabled sensors and mobile-captured images to develop a hybrid machine learning (ML) model for early and accurate plant disease detection. The proposed model combines a multilayer perceptron (MLP) for processing numerical sensor inputs—ambient temperature, soil temperature, relative humidity, soil moisture, and leaf wetness duration—and a convolutional neural network (CNN) for analyzing leaf images labeled as leaf spot, anthracnose, or healthy. Outputs from the MLP and CNN concatenated and processed through an additional MLP to classify plant health effectively. Optimized with hidden layer configurations of 8-16-32-8 for the sensor-data MLP, 16--32-64-128_32-8-4 for the image-data CNN, and 4-3 layers for the final MLP, the model achieves a loss of 1% and an accuracy of 95%, outperforming state-of-the-art methods, such as DenseNet201-support vector machines (SVM) (87.23%) and gray level co-occurrence matrix-SVM (90%). Performance metrics demonstrate high precision (leaf spot: 0.93, anthracnose: 0.93, and healthy: 0.98), recall (leaf spot: 0.92, anthracnose: 0.95, and healthy: 0.96), and F1-scores (leaf spot: 0.92, anthracnose: 0.94, and healthy: 0.97). The model’s deployment on an Amazon Web Services cloud server enables real-time disease detection and classification, making it accessible for practical agricultural use. This sensor and image data integration offers a novel and robust solution to address the limitations of single-modality approaches.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"582-590"},"PeriodicalIF":0.0,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248039","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 : 2025-08-05DOI: 10.1109/TAFE.2025.3591910
Xueqian Fu;Hai Long
To unlock the regulation potential of agricultural loads and enhance the ecofriendly and cost- efficient operation of microgrids, this article proposes a virtual power plant (VPP) scheduling model in alignment with the VPP policy implemented in Shandong Province, China. The proposed model coordinates the control of distributed energy resources (DERs) and greenhouse loads while considering the operational constraints of agricultural production and electricity consumption on the demand side, as well as the generation limitations of DERs. By fully exploiting the flexibility of greenhouse supplemental lighting systems, the proposed model enables an integrated optimization of load demand and energy supply, thereby achieving cost-effective peak-shaving strategies. Compared with the conventional agricultural microgrid operation strategies, the proposed VPP scheduling strategy exhibits superior economic performance. Simulation results demonstrate that the proposed model achieves a peak-shaving capacity of 4.28 MW over a 3-h period and yields an additional daily revenue of CNY 818 for the agricultural microgrid, highlighting its potential to facilitate the expansion of VPP applications from urban to rural settings.
{"title":"Virtual Power Plant Scheduling in Agricultural Microgrids Through the Control of Distributed Energy Resources and Greenhouse Loads","authors":"Xueqian Fu;Hai Long","doi":"10.1109/TAFE.2025.3591910","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3591910","url":null,"abstract":"To unlock the regulation potential of agricultural loads and enhance the ecofriendly and cost- efficient operation of microgrids, this article proposes a virtual power plant (VPP) scheduling model in alignment with the VPP policy implemented in Shandong Province, China. The proposed model coordinates the control of distributed energy resources (DERs) and greenhouse loads while considering the operational constraints of agricultural production and electricity consumption on the demand side, as well as the generation limitations of DERs. By fully exploiting the flexibility of greenhouse supplemental lighting systems, the proposed model enables an integrated optimization of load demand and energy supply, thereby achieving cost-effective peak-shaving strategies. Compared with the conventional agricultural microgrid operation strategies, the proposed VPP scheduling strategy exhibits superior economic performance. Simulation results demonstrate that the proposed model achieves a peak-shaving capacity of 4.28 MW over a 3-h period and yields an additional daily revenue of CNY 818 for the agricultural microgrid, highlighting its potential to facilitate the expansion of VPP applications from urban to rural settings.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"569-581"},"PeriodicalIF":0.0,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248045","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 : 2025-08-04DOI: 10.1109/TAFE.2025.3590771
Daniel Fiallo;N. Robert Bennett;Michael G. Farrier;Adam Wang;Weixin Cheng;Shiva Abbaszadeh
Computed tomography (CT) serves as a noninvasive technique for pinpointing specific areas within objects, facilitating the examination of soil distributions and localized flow processes within soil pore networks. CT scanning yields cross-sectional sequences that unveil insights into the internal structure of pore networks, which is crucial for understanding root–soil interactions. In this investigation, we explore the potential of employing a high-resolution amorphous selenium (a-Se) direct conversion detector coupled with complementary metal–oxide–semiconductor (CMOS) readouts for micro-CT scanning of soil matrices. This approach aims to visualize the aggregation status and pore network connectivity within intact soil. In addition, we compare the capabilities of the a-Se/CMOS detector with other commercially available detectors evaluating performance in terms of spatial resolution, noise levels, and overall imaging quality. The integration of a-Se’s intrinsic high spatial resolution with small-pixel CMOS readouts enables detailed visualization of soil aggregates in plant samples. By varying X-ray energy and soil thickness, we achieved a spatial resolution of $leq$ 25 $mu$m and a noise-limited performance of eight photons/pixel at 20 keV. Although thick soil presents challenges due to high X-ray attenuation, finer details are discernible in thinner samples, underscoring the importance of careful selection of soil thickness and container material.
{"title":"Comparison Study of a-Se/CMOS Detector and Commercial Alternatives for High-Resolution X-Ray Imaging of Soil Structure","authors":"Daniel Fiallo;N. Robert Bennett;Michael G. Farrier;Adam Wang;Weixin Cheng;Shiva Abbaszadeh","doi":"10.1109/TAFE.2025.3590771","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3590771","url":null,"abstract":"Computed tomography (CT) serves as a noninvasive technique for pinpointing specific areas within objects, facilitating the examination of soil distributions and localized flow processes within soil pore networks. CT scanning yields cross-sectional sequences that unveil insights into the internal structure of pore networks, which is crucial for understanding root–soil interactions. In this investigation, we explore the potential of employing a high-resolution amorphous selenium (a-Se) direct conversion detector coupled with complementary metal–oxide–semiconductor (CMOS) readouts for micro-CT scanning of soil matrices. This approach aims to visualize the aggregation status and pore network connectivity within intact soil. In addition, we compare the capabilities of the a-Se/CMOS detector with other commercially available detectors evaluating performance in terms of spatial resolution, noise levels, and overall imaging quality. The integration of a-Se’s intrinsic high spatial resolution with small-pixel CMOS readouts enables detailed visualization of soil aggregates in plant samples. By varying X-ray energy and soil thickness, we achieved a spatial resolution of <inline-formula><tex-math>$leq$</tex-math></inline-formula> 25 <inline-formula><tex-math>$mu$</tex-math></inline-formula>m and a noise-limited performance of eight photons/pixel at 20 keV. Although thick soil presents challenges due to high X-ray attenuation, finer details are discernible in thinner samples, underscoring the importance of careful selection of soil thickness and container material.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"561-568"},"PeriodicalIF":0.0,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248032","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}
Many real-world agricultural applications, such as automatic pest recognition, benefit from lightweight deep learning (DL) architectures due to their reduced computational complexity, enabling deployment on resource-constrained devices. However, this paradigm shift comes at the cost of model performance, significantly limiting its extensive use. Traditional data-centric approaches for improving model performance, such as using large training datasets, are often unsuitable for the agricultural domain due to limited labeled data and high data collection costs. On the other hand, architectural improvements, such as attention mechanisms, have demonstrated the potential to enhance the performance of lightweight DL architectures. However, improper integration can lead to increased complexity and diminished performance. To address this challenge, this study proposes a novel mechanism to systematically determine the optimal integration configuration of popular attention techniques with the MobileNet lightweight DL architecture. The proposed method is evaluated on four variants of two benchmark plant pest datasets (D15,869 and D1500, D21599, and D2545) and the best integration configurations are reported along with their results. The Bottleneck Attention Module (BAM) attention mechanism, integrated into 12 different layers of MobileNetV2 (BAM12), demonstrated superior performance on D15869 and D1500, and D21599 and D2545, while integrating BAM into eight layers yielded higher accuracy on D21599. As a result, a comparison with the MobileNet baseline demonstrates that the careful integration of attention mechanisms significantly improves performance.
{"title":"Efficient Attention-Lightweight Deep Learning Architecture Integration for Plant Pest Recognition","authors":"Sivasubramaniam Janarthan;Selvarajah Thuseethan;Charles Joseph;Vigneshwaran Palanisamy;Sutharshan Rajasegarar;John Yearwood","doi":"10.1109/TAFE.2025.3583334","DOIUrl":"https://doi.org/10.1109/TAFE.2025.3583334","url":null,"abstract":"Many real-world agricultural applications, such as automatic pest recognition, benefit from lightweight deep learning (DL) architectures due to their reduced computational complexity, enabling deployment on resource-constrained devices. However, this paradigm shift comes at the cost of model performance, significantly limiting its extensive use. Traditional data-centric approaches for improving model performance, such as using large training datasets, are often unsuitable for the agricultural domain due to limited labeled data and high data collection costs. On the other hand, architectural improvements, such as attention mechanisms, have demonstrated the potential to enhance the performance of lightweight DL architectures. However, improper integration can lead to increased complexity and diminished performance. To address this challenge, this study proposes a novel mechanism to systematically determine the optimal integration configuration of popular attention techniques with the MobileNet lightweight DL architecture. The proposed method is evaluated on four variants of two benchmark plant pest datasets (D1<sub>5,869</sub> and D1<sub>500</sub>, D2<sub>1599</sub>, and D2<sub>545</sub>) and the best integration configurations are reported along with their results. The Bottleneck Attention Module (BAM) attention mechanism, integrated into 12 different layers of MobileNetV2 (BAM12), demonstrated superior performance on D1<sub>5869</sub> and D1<sub>500</sub>, and D2<sub>1599</sub> and D2<sub>545</sub>, while integrating BAM into eight layers yielded higher accuracy on D2<sub>1599</sub>. As a result, a comparison with the MobileNet baseline demonstrates that the careful integration of attention mechanisms significantly improves performance.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 2","pages":"548-560"},"PeriodicalIF":0.0,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145248037","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：