During the cultivation and growth of green plums, various defects frequently occur, potentially affecting their overall quality and economic value. Accurate classification and identification of these defects have become essential components of the harvesting process, particularly when employing smart agricultural equipment. These defects pose significant challenges to the yield and quality of green plums, making their precise detection crucial for ensuring optimal output and economic efficiency. However, most contemporary research on fruit defect classification and grading using artificial intelligence techniques primarily focuses on accuracy, often neglecting the constraints imposed by limited resources. This study addresses the aforementioned challenges by employing knowledge distillation techniques to optimize the performance of a lightweight model. Specifically, during the knowledge distillation process, the vision transformer model, known for its robust recognition capabilities, was selected as the teacher model. The lightweight MobileNetv3 model, chosen for its ease of deployment, served as the student model and was trained using the Lion optimizer. In addition, the dual guidance learning module was designed to enhance knowledge transfer between the teacher and student models, thereby improving the overall capability of the student model. Experimental validation demonstrated that the proposed method excels in the green plum defect recognition task, with the student model, MobileNetv3, achieving an accuracy of 99.17% and exhibiting high performance in key metrics such as precision, recall, and F1-score. Notably, MobileNetv3 not only delivers exceptional performance but also features a low parameter count and computational complexity, facilitating its efficient deployment in practical applications. This study provides an effective and practical solution for the automatic identification and sorting of green plum defects, significantly advancing the development and application of smart agricultural technologies.
{"title":"An Approach Based on Knowledge Distillation for Lightweight Defect Classification of Green Plums","authors":"Jinhai Wang;Wei Wang;Lan Liao;Lufeng Luo;Xuemin Lin;Xinan Zeng","doi":"10.1109/TAFE.2024.3488196","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3488196","url":null,"abstract":"During the cultivation and growth of green plums, various defects frequently occur, potentially affecting their overall quality and economic value. Accurate classification and identification of these defects have become essential components of the harvesting process, particularly when employing smart agricultural equipment. These defects pose significant challenges to the yield and quality of green plums, making their precise detection crucial for ensuring optimal output and economic efficiency. However, most contemporary research on fruit defect classification and grading using artificial intelligence techniques primarily focuses on accuracy, often neglecting the constraints imposed by limited resources. This study addresses the aforementioned challenges by employing knowledge distillation techniques to optimize the performance of a lightweight model. Specifically, during the knowledge distillation process, the vision transformer model, known for its robust recognition capabilities, was selected as the teacher model. The lightweight MobileNetv3 model, chosen for its ease of deployment, served as the student model and was trained using the Lion optimizer. In addition, the dual guidance learning module was designed to enhance knowledge transfer between the teacher and student models, thereby improving the overall capability of the student model. Experimental validation demonstrated that the proposed method excels in the green plum defect recognition task, with the student model, MobileNetv3, achieving an accuracy of 99.17% and exhibiting high performance in key metrics such as precision, recall, and F1-score. Notably, MobileNetv3 not only delivers exceptional performance but also features a low parameter count and computational complexity, facilitating its efficient deployment in practical applications. This study provides an effective and practical solution for the automatic identification and sorting of green plum defects, significantly advancing the development and application of smart agricultural technologies.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"213-223"},"PeriodicalIF":0.0,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821598","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 : 2024-12-23DOI: 10.1109/TAFE.2024.3513637
Hari Chandana Pichhika;Priyambada Subudhi;Raja Vara Prasad Yerra
Automatic yield estimation is crucial for fruit cultivation, impacting everything from harvesting to marketing. This article introduces an efficient tracking mechanism for accurate yield estimation in mango farming, addressing challenges such as fruit detection inconsistency and over-counting. We utilized this tracking-based solution on a video dataset collected in a $360^circ$ viewpoint of each mango tree in one-acre Banginapalle orchard during daylight. The videos underwent preprocessing, including gamma correction, Gaussian smoothing, and stabilization to minimize the quivering of video frames. We also implemented a cosine similarity technique to remove redundant frames with 90% similarity and segmented the canopy to identify the regions of interest. The mango detection system employs YOLOv8s and an extended Kalman filter that preserves special Euclidean group [SE(3)] equivariance, ensuring accurate mango tracking across frames, which is robust to camera movements through angular estimation. Our method surpasses existing tracking-bas algorithms such as Sort, DeepSort, and Bot-sort in tests with ten video sequences. In addition, the results are also comparable to the harvest count obtained from the farmer and the labeling count performed manually in the video frames, achieving results close to a mean absolute error of 0.341 and 0.089, respectively.
{"title":"Extended Kalman Filter Based Tracking Method for Accurate Fruit Yield Estimation Preserving SE(3) Equivariance","authors":"Hari Chandana Pichhika;Priyambada Subudhi;Raja Vara Prasad Yerra","doi":"10.1109/TAFE.2024.3513637","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3513637","url":null,"abstract":"Automatic yield estimation is crucial for fruit cultivation, impacting everything from harvesting to marketing. This article introduces an efficient tracking mechanism for accurate yield estimation in mango farming, addressing challenges such as fruit detection inconsistency and over-counting. We utilized this tracking-based solution on a video dataset collected in a <inline-formula><tex-math>$360^circ$</tex-math></inline-formula> viewpoint of each mango tree in one-acre Banginapalle orchard during daylight. The videos underwent preprocessing, including gamma correction, Gaussian smoothing, and stabilization to minimize the quivering of video frames. We also implemented a cosine similarity technique to remove redundant frames with 90% similarity and segmented the canopy to identify the regions of interest. The mango detection system employs YOLOv8s and an extended Kalman filter that preserves special Euclidean group [SE(3)] equivariance, ensuring accurate mango tracking across frames, which is robust to camera movements through angular estimation. Our method surpasses existing tracking-bas algorithms such as Sort, DeepSort, and Bot-sort in tests with ten video sequences. In addition, the results are also comparable to the harvest count obtained from the farmer and the labeling count performed manually in the video frames, achieving results close to a mean absolute error of 0.341 and 0.089, respectively.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"200-212"},"PeriodicalIF":0.0,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821873","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 : 2024-12-11DOI: 10.1109/TAFE.2024.3485949
Yu Zhang;Kang Liu;Xulong Wang;Rujing Wang;Po Yang
Precision fertilization is essential in agricultural systems for balancing soil nutrients, conserving fertilizer, decreasing emissions, and increasing crop yields. Access to comprehensive and diverse agricultural data is problematic due to the lack of sophisticated sensor and network technologies on the majority of farms, and available agricultural data are generally unstructured and difficult to mine. The absence of agricultural data is, consequently, a significant impediment to the utilization of machine learning approaches for precision fertilization. In this research, we investigate newly gathered genuine agricultural dataset from nine real winter wheat farms in the United Kingdom, which encompass an extensive variety of agricultural variables, including climate, soil nutrients, and farming data. To deal with the spatio-temporal characteristics of agricultural dataset and to address the problem of scarcity in agricultural data, we propose a novel machine learning approach integrating multi-task learning and one-shot learning, which utilizes a multi-dimensional tensor constructed from original data combined with fertilization temporal patterns extracted by contrasting with environmental information from existing real farms to accurately predict the quantity and timing of base and top dressing fertilization. Specifically, agricultural data are converted into a 3-D tensor and tensor decomposition technique is utilized to derive a set of comprehensible spatio-temporal latent factors from the original data. The latent factors are subsequently utilized to construct the spatio-temporal tensor prediction model as multi-task relationships. The proposed one-shot learning approach utilizes the Mahalanobis distance to evaluate the similarity of environmental information between the target farm and existing real-world farms as a determinant of whether to transfer the fertilization temporal pattern of existing farm to the target farm. Comprehensive experiments are conducted to compare the proposed approach with standard regression models utilizing the real-world agricultural dataset. The experimental results demonstrate that our proposed approach presents superior accuracy and stability for fertilization prediction.
{"title":"Precision Fertilization via Spatio-temporal Tensor Multi-task Learning and One-Shot Learning","authors":"Yu Zhang;Kang Liu;Xulong Wang;Rujing Wang;Po Yang","doi":"10.1109/TAFE.2024.3485949","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3485949","url":null,"abstract":"Precision fertilization is essential in agricultural systems for balancing soil nutrients, conserving fertilizer, decreasing emissions, and increasing crop yields. Access to comprehensive and diverse agricultural data is problematic due to the lack of sophisticated sensor and network technologies on the majority of farms, and available agricultural data are generally unstructured and difficult to mine. The absence of agricultural data is, consequently, a significant impediment to the utilization of machine learning approaches for precision fertilization. In this research, we investigate newly gathered genuine agricultural dataset from nine real winter wheat farms in the United Kingdom, which encompass an extensive variety of agricultural variables, including climate, soil nutrients, and farming data. To deal with the spatio-temporal characteristics of agricultural dataset and to address the problem of scarcity in agricultural data, we propose a novel machine learning approach integrating multi-task learning and one-shot learning, which utilizes a multi-dimensional tensor constructed from original data combined with fertilization temporal patterns extracted by contrasting with environmental information from existing real farms to accurately predict the quantity and timing of base and top dressing fertilization. Specifically, agricultural data are converted into a 3-D tensor and tensor decomposition technique is utilized to derive a set of comprehensible spatio-temporal latent factors from the original data. The latent factors are subsequently utilized to construct the spatio-temporal tensor prediction model as multi-task relationships. The proposed one-shot learning approach utilizes the Mahalanobis distance to evaluate the similarity of environmental information between the target farm and existing real-world farms as a determinant of whether to transfer the fertilization temporal pattern of existing farm to the target farm. Comprehensive experiments are conducted to compare the proposed approach with standard regression models utilizing the real-world agricultural dataset. The experimental results demonstrate that our proposed approach presents superior accuracy and stability for fertilization prediction.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"190-199"},"PeriodicalIF":0.0,"publicationDate":"2024-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143820329","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 : 2024-12-05DOI: 10.1109/TAFE.2024.3504956
Muhib Ullah;Abdul Bais;Tyler Wist
Crop health monitoring is crucial for implementing timely and effective interventions that ensure sustainability and maximize crop yield. Flea beetles (FB), Crucifer (Phyllotreta cruciferae) and Striped (Phyllotreta striolata), pose a significant threat to canola crop health and cause substantial damage if not addressed promptly. Accurate and timely damage quantification is crucial for implementing targeted pest management strategies if insecticidal seed treatments are overcome by FB feeding to minimize yield losses if the action threshold is exceeded. Traditional manual field monitoring for FB damage is time-consuming and error-prone due to reliance on human visual estimates of FB damage. This article proposes TranSEF, a novel self-ensemble semantic segmentation algorithm that utilizes a hybrid convolutional neural network-vision transformer (ViT) encoder–decoder framework. The encoder employs a modified cross-stage partial DenseNet (CSPDenseNet), MCSPDNet, which enhances attention to tiny regions by aggregating spatially aware features from shallow layers with deeper, more abstract features. ViTs effectively capture the global context in the decoder by modeling long-range dependencies and relationships across the image. Each decoder independently processes inputs from different stages of the MCSPDNet, acting as a weak learner within an ensemble-like approach. Unlike traditional ensemble learning approaches that train weak learners separately, TranSEF is trained end-to-end, making it a self-ensembling framework. TranSEF uses hybrid supervision with a composite loss function, where decoders generate independent predictions and simultaneously supervise each other. TranSEF achieves IoU scores of 0.831 for canola leaves and 0.807 for FB damage, and the overall mIoU improved by 2.29% and 1.56% over DeepLabv3+ and SegFormer, respectively, while utilizing only 35.42 M trainable parameters-significantly fewer than DeepLabv3+ (63 M) and SegFormer (61 M).
{"title":"TranSEF: Transformer Enhanced Self-Ensemble Framework for Damage Assessment in Canola Crops","authors":"Muhib Ullah;Abdul Bais;Tyler Wist","doi":"10.1109/TAFE.2024.3504956","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3504956","url":null,"abstract":"Crop health monitoring is crucial for implementing timely and effective interventions that ensure sustainability and maximize crop yield. Flea beetles (FB), Crucifer (Phyllotreta cruciferae) and Striped (Phyllotreta striolata), pose a significant threat to canola crop health and cause substantial damage if not addressed promptly. Accurate and timely damage quantification is crucial for implementing targeted pest management strategies if insecticidal seed treatments are overcome by FB feeding to minimize yield losses if the action threshold is exceeded. Traditional manual field monitoring for FB damage is time-consuming and error-prone due to reliance on human visual estimates of FB damage. This article proposes TranSEF, a novel self-ensemble semantic segmentation algorithm that utilizes a hybrid convolutional neural network-vision transformer (ViT) encoder–decoder framework. The encoder employs a modified cross-stage partial DenseNet (CSPDenseNet), MCSPDNet, which enhances attention to tiny regions by aggregating spatially aware features from shallow layers with deeper, more abstract features. ViTs effectively capture the global context in the decoder by modeling long-range dependencies and relationships across the image. Each decoder independently processes inputs from different stages of the MCSPDNet, acting as a weak learner within an ensemble-like approach. Unlike traditional ensemble learning approaches that train weak learners separately, TranSEF is trained end-to-end, making it a self-ensembling framework. TranSEF uses hybrid supervision with a composite loss function, where decoders generate independent predictions and simultaneously supervise each other. TranSEF achieves IoU scores of 0.831 for canola leaves and 0.807 for FB damage, and the overall mIoU improved by 2.29% and 1.56% over DeepLabv3+ and SegFormer, respectively, while utilizing only 35.42 M trainable parameters-significantly fewer than DeepLabv3+ (63 M) and SegFormer (61 M).","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"179-189"},"PeriodicalIF":0.0,"publicationDate":"2024-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821526","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}
The nutritional value of agricultural products is an important indicator for evaluating their quality, which directly affects people's dietary choices and overall well-being. Nutritional assessment studies provide a scientific basis for the production, processing, and marketing of food by analyzing the nutrients they contain. Traditional methods often struggle with suboptimal accuracy and can be time consuming, as well as a shortage of professionals. The progress in artificial intelligence has revolutionized dietary health by offering more accessible methods for food nutritional assessment using vision-based approaches. However, existing vision-based methods using RGB images often face challenges due to varying lighting conditions, impacting the accuracy of nutritional assessment. An alternative is the RGB-D fusion method, which combines RGB images and depth maps. Yet, these methods typically rely on simple fusion techniques that do not ensure precise assessment. Additionally, current vision-based methods struggle to detect small components like oils and sugars on food surfaces, crucial for determining ingredient information and ensuring accurate nutritional assessment. In this pursuit, we propose a novel ingredient-guided RGB-D fusion network that integrates RGB images with depth maps and enables more reliable nutritional assessment guided by ingredient information. Specifically, the multifrequency bimodality fusion module is designed to leverage the correlation between the RGB image and the depth map within the frequency domain. Furthermore, the progressive-fusion module and ingredient-guided module leverage ingredient information to explore the potential correlation between ingredients and nutrients, thereby enhancing the guidance for nutritional assessment learning. We evaluate our approach on a variety of ablation settings on Nutrition5k, where it consistently outperforms state-of-the-art methods.
{"title":"Ingredient-Guided RGB-D Fusion Network for Nutritional Assessment","authors":"Zhihui Feng;Hao Xiong;Weiqing Min;Sujuan Hou;Huichuan Duan;Zhonghua Liu;Shuqiang Jiang","doi":"10.1109/TAFE.2024.3493332","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3493332","url":null,"abstract":"The nutritional value of agricultural products is an important indicator for evaluating their quality, which directly affects people's dietary choices and overall well-being. Nutritional assessment studies provide a scientific basis for the production, processing, and marketing of food by analyzing the nutrients they contain. Traditional methods often struggle with suboptimal accuracy and can be time consuming, as well as a shortage of professionals. The progress in artificial intelligence has revolutionized dietary health by offering more accessible methods for food nutritional assessment using vision-based approaches. However, existing vision-based methods using RGB images often face challenges due to varying lighting conditions, impacting the accuracy of nutritional assessment. An alternative is the RGB-D fusion method, which combines RGB images and depth maps. Yet, these methods typically rely on simple fusion techniques that do not ensure precise assessment. Additionally, current vision-based methods struggle to detect small components like oils and sugars on food surfaces, crucial for determining ingredient information and ensuring accurate nutritional assessment. In this pursuit, we propose a novel ingredient-guided RGB-D fusion network that integrates RGB images with depth maps and enables more reliable nutritional assessment guided by ingredient information. Specifically, the multifrequency bimodality fusion module is designed to leverage the correlation between the RGB image and the depth map within the frequency domain. Furthermore, the progressive-fusion module and ingredient-guided module leverage ingredient information to explore the potential correlation between ingredients and nutrients, thereby enhancing the guidance for nutritional assessment learning. We evaluate our approach on a variety of ablation settings on Nutrition5k, where it consistently outperforms state-of-the-art methods.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"156-166"},"PeriodicalIF":0.0,"publicationDate":"2024-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821604","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}
Plant pests are a major threat to sustainable food supply, causing damage to food production and agriculture industries around the world. Despite these negative impacts, on several occasions, plant pests have also been used to improve the quality of agricultural products. Although deep learning-based automated plant pest identification techniques have shown tremendous success in the recent past, they are often limited by increased computational cost, large training data requirements, and impaired performance when they present in complex backgrounds. Therefore, to alleviate these challenges, a lightweight attention-based convolutional neural network architecture, called LiRAN, based on a novel simplified attention mask module and an extended MobileNetV2 architecture, is proposed in this study. The experimental results reveal that the proposed architecture can attain 96.25%, 98.9%, and 91% accuracies on three variants of publicly available datasets with 5869, 545, and 500 sample images, respectively, showcasing high performance consistently in large and small data conditions. More importantly, this model can be deployed on smartphones or other resource-constrained embedded devices for in-field realization, only requiring $approx$ 9.3 MB of storage space with around 2.37 M parameters and 0.34 giga multiply-and-accumulate FLOPs with an input image size of 224 × 224.
{"title":"LiRAN: A Lightweight Residual Attention Network for In-Field Plant Pest Recognition","authors":"Sivasubramaniam Janarthan;Selvarajah Thuseethan;Sutharshan Rajasegarar;Qiang Lyu;Yongqiang Zheng;John Yearwood","doi":"10.1109/TAFE.2024.3496798","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3496798","url":null,"abstract":"Plant pests are a major threat to sustainable food supply, causing damage to food production and agriculture industries around the world. Despite these negative impacts, on several occasions, plant pests have also been used to improve the quality of agricultural products. Although deep learning-based automated plant pest identification techniques have shown tremendous success in the recent past, they are often limited by increased computational cost, large training data requirements, and impaired performance when they present in complex backgrounds. Therefore, to alleviate these challenges, a lightweight attention-based convolutional neural network architecture, called LiRAN, based on a novel simplified attention mask module and an extended MobileNetV2 architecture, is proposed in this study. The experimental results reveal that the proposed architecture can attain 96.25%, 98.9%, and 91% accuracies on three variants of publicly available datasets with 5869, 545, and 500 sample images, respectively, showcasing high performance consistently in large and small data conditions. More importantly, this model can be deployed on smartphones or other resource-constrained embedded devices for in-field realization, only requiring <inline-formula><tex-math>$approx$</tex-math></inline-formula> 9.3 MB of storage space with around 2.37 M parameters and 0.34 giga multiply-and-accumulate FLOPs with an input image size of 224 × 224.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"167-178"},"PeriodicalIF":0.0,"publicationDate":"2024-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821872","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 : 2024-11-21DOI: 10.1109/TAFE.2024.3492323
Olakunle Elijah;Abiodun Emmanuel Abioye;Tawanda E. Maguvu
Ginger (Zingiber officinale), a globally cultivated spice crop, is vital to numerous economies. However, its production faces significant challenges due to pests and diseases, which can lead to substantial yield losses. Traditional methods for detecting these threats often rely on visual inspection by human experts, a process that is time-consuming, labor-intensive, and prone to errors. This article examines the potential of artificial intelligence (AI) to address these limitations and transform ginger cultivation. It provides a comprehensive analysis of conventional pest and disease management strategies, identifying their short comings and exploring the potential of emerging AI technologies, including the AI of things’ applications, for accurate, efficient, and timely detection and control. By pinpointing the challenges and outlining promising avenues for future research, this study aims to equip agriculturists and researchers with the knowledge necessary to optimize ginger production, enhance food security, and foster sustainable farming practices.
{"title":"Pest and Disease Management in Ginger Plants: Artificial Intelligence of Things (AIoT)","authors":"Olakunle Elijah;Abiodun Emmanuel Abioye;Tawanda E. Maguvu","doi":"10.1109/TAFE.2024.3492323","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3492323","url":null,"abstract":"Ginger (<italic>Zingiber officinale</i>), a globally cultivated spice crop, is vital to numerous economies. However, its production faces significant challenges due to pests and diseases, which can lead to substantial yield losses. Traditional methods for detecting these threats often rely on visual inspection by human experts, a process that is time-consuming, labor-intensive, and prone to errors. This article examines the potential of artificial intelligence (AI) to address these limitations and transform ginger cultivation. It provides a comprehensive analysis of conventional pest and disease management strategies, identifying their short comings and exploring the potential of emerging AI technologies, including the AI of things’ applications, for accurate, efficient, and timely detection and control. By pinpointing the challenges and outlining promising avenues for future research, this study aims to equip agriculturists and researchers with the knowledge necessary to optimize ginger production, enhance food security, and foster sustainable farming practices.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"86-97"},"PeriodicalIF":0.0,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821874","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 : 2024-11-14DOI: 10.1109/TAFE.2024.3488747
Baofeng Ji;Jingming Zhao;Fazhan Tao;Ji Zhang;Gaoyuan Zhang;Nan Wang;Ping Zhang;Huitao Fan
Fruit yield assessment is an important aspect of orchard management. In this context, target detection of fruit is of paramount importance. However, due to complex factors in real orchard environments, such as fruit occlusion, insufficient lighting, and overlapping fruits, traditional detection and counting methods often suffer from low detection accuracy and inadequate classification precision, failing to meet the requirements of practical applications. To address this issue, we focus on nectarine fruit and propose an improved YOLOv8n-based object detection algorithm model, YOLOv8n-global feature extraction enhancement (GFE). We integrate the effective squeeze-and-excitation attention mechanism into the YOLOv8n model. This integration allows our approach to adaptively adjust the weight of each channel, which enhances both detection efficiency and target recognition accuracy. Then, we introduce focal distance-intersection over union loss to address the misjudgment of hard samples. This further contributes to improving detection accuracy. In addition, we incorporate the gather-and-distribute mechanism from GOLD-YOLO, replacing the traditional feature pyramid network structure. This enhancement improves the information fusion capability in the neck of the model, leading to a higher mean average precision (mAP@0.5). In addition, the output of the improved model can be used as an input to DEEPSORT to classify and count nectarine fruit. This functionality can be used for estimating fruit maturity and yield in orchards. Experimental results demonstrate that the YOLOv8n-GFE model achieves a mAP@0.5 of 92.5%, which is an improvement of 3.2% over the original YOLOv8n model, meeting the required accuracy for recognizing nectarine fruit maturity in practical applications.
{"title":"A Novel Nectarine Fruit Maturity Detection and Classification Counting Model Based on YOLOv8n","authors":"Baofeng Ji;Jingming Zhao;Fazhan Tao;Ji Zhang;Gaoyuan Zhang;Nan Wang;Ping Zhang;Huitao Fan","doi":"10.1109/TAFE.2024.3488747","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3488747","url":null,"abstract":"Fruit yield assessment is an important aspect of orchard management. In this context, target detection of fruit is of paramount importance. However, due to complex factors in real orchard environments, such as fruit occlusion, insufficient lighting, and overlapping fruits, traditional detection and counting methods often suffer from low detection accuracy and inadequate classification precision, failing to meet the requirements of practical applications. To address this issue, we focus on nectarine fruit and propose an improved YOLOv8n-based object detection algorithm model, YOLOv8n-global feature extraction enhancement (GFE). We integrate the effective squeeze-and-excitation attention mechanism into the YOLOv8n model. This integration allows our approach to adaptively adjust the weight of each channel, which enhances both detection efficiency and target recognition accuracy. Then, we introduce focal distance-intersection over union loss to address the misjudgment of hard samples. This further contributes to improving detection accuracy. In addition, we incorporate the gather-and-distribute mechanism from GOLD-YOLO, replacing the traditional feature pyramid network structure. This enhancement improves the information fusion capability in the neck of the model, leading to a higher mean average precision (mAP@0.5). In addition, the output of the improved model can be used as an input to DEEPSORT to classify and count nectarine fruit. This functionality can be used for estimating fruit maturity and yield in orchards. Experimental results demonstrate that the YOLOv8n-GFE model achieves a mAP@0.5 of 92.5%, which is an improvement of 3.2% over the original YOLOv8n model, meeting the required accuracy for recognizing nectarine fruit maturity in practical applications.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"144-155"},"PeriodicalIF":0.0,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821550","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 : 2024-11-11DOI: 10.1109/TAFE.2024.3472493
Marcel Balle;Wenxiu Xu;Kevin FA Darras;Thomas Cherico Wanger
Recent advances in Internet of Things and artificial intelligence technologies have shifted automated monitoring in smart agriculture toward low power sensors and embedded vision on powerful processing units. Vision-based monitoring devices need an effective power management and control system with system-adapted power input and output capabilities to achieve power-efficient and self-sustainable operation. Here, we present a universal power management solution for automated monitoring devices in agricultural systems, compatible with commonly used off-the-shelf edge processing units (EPUs). The proposed design is specifically adapted for battery-powered EPU systems by incorporating power-matched energy harvesting, a power switch with low-power sleep mode, and simple system integration in an microcontroller unit-less architecture with automated operation. We use a four-month case study to monitor the effects of plastic pollution in agricultural soils on plant growth under 4-mg microplastic exposure, demonstrating that the setup achieved continuous and sustainable operation. In this agricultural application, our power management module is deployed in an embedded vision camera equipped with a 5-W solar panel and five various environmental sensors, effectively monitoring environmental stress and plant growth state. This work highlights the application of the power management board in embedded agricultural monitoring devices for precision farming.
{"title":"A Power Management and Control System for Environmental Monitoring Devices","authors":"Marcel Balle;Wenxiu Xu;Kevin FA Darras;Thomas Cherico Wanger","doi":"10.1109/TAFE.2024.3472493","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3472493","url":null,"abstract":"Recent advances in Internet of Things and artificial intelligence technologies have shifted automated monitoring in smart agriculture toward low power sensors and embedded vision on powerful processing units. Vision-based monitoring devices need an effective power management and control system with system-adapted power input and output capabilities to achieve power-efficient and self-sustainable operation. Here, we present a universal power management solution for automated monitoring devices in agricultural systems, compatible with commonly used off-the-shelf edge processing units (EPUs). The proposed design is specifically adapted for battery-powered EPU systems by incorporating power-matched energy harvesting, a power switch with low-power sleep mode, and simple system integration in an microcontroller unit-less architecture with automated operation. We use a four-month case study to monitor the effects of plastic pollution in agricultural soils on plant growth under 4-mg microplastic exposure, demonstrating that the setup achieved continuous and sustainable operation. In this agricultural application, our power management module is deployed in an embedded vision camera equipped with a 5-W solar panel and five various environmental sensors, effectively monitoring environmental stress and plant growth state. This work highlights the application of the power management board in embedded agricultural monitoring devices for precision farming.","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"3 1","pages":"134-143"},"PeriodicalIF":0.0,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821602","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 : 2024-10-24DOI: 10.1109/TAFE.2024.3483630
{"title":"2024 Index IEEE Transactions on AgriFood Electronics Vol. 2","authors":"","doi":"10.1109/TAFE.2024.3483630","DOIUrl":"https://doi.org/10.1109/TAFE.2024.3483630","url":null,"abstract":"","PeriodicalId":100637,"journal":{"name":"IEEE Transactions on AgriFood Electronics","volume":"2 2","pages":"638-652"},"PeriodicalIF":0.0,"publicationDate":"2024-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10734674","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142540444","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}
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：