Smart agricultural technology最新文献_第8页

Fruit-flower-leaf dynamic response of Lycium barbarum L. for vibration harvesting

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-16 DOI: 10.1016/j.atech.2024.100722

Qingyu Chen, Naishuo Wei, Yunlei Fan, Zeyu Wang, Jianguo Zhou, Zening Gao, Yu Chen, Jun Chen

For the characteristic of infinite inflorescence, the aim of vibration harvesting of Lycium barbarum L. (L. barbarum) is to picking ripe fruits and leaving unripe fruits, flowers and leaves. Therefore, it is essential to understand the dynamic response of each component during vibration. In this study, the 3D models of a branch, ripe fruit, unripe fruit, flower and leaf were established respectively, which were assembled into a branch-fruit-flower-leaf system model based on the growth characteristics of L. barbarum. The vibration harvester was designed based on the hedge agronomy and simplified as rods, and a rigid-flexible coupling model was established using the kinetic analysis and the finite element method. The dynamic response of branch-fruit-flower-leaf during vibration harvesting was obtained when the rods continuously excited the branch. Results showed that the maximum acceleration of fruit-flower-leaf during vibration harvesting was in the same order of magnitude. And according to the formula of inertia force, the detachment force and mass were calculated and found that the detachment acceleration of ripe fruit was much smaller than that of unripe fruit, flower, and leaf. This study presents a theoretical basis for achieving the harvesting target. Scripts were written to simulate the vibration detachment process of ripe fruit, and high-speed photography was used for experimental verification. The results indicate that the simulation error was 9.15 %, demonstrating that the rigid-flexible coupling simulation is more effective. The field test showed that the picking rate of ripe fruit of 82.69 %, the picking rate of unripe fruit of 3.13 %, and the damage rate of ripe fruit of 4.06 %. The research provides a new analytical approach and theoretical basis for researching vibration harvesting.

{"title":"Fruit-flower-leaf dynamic response of Lycium barbarum L. for vibration harvesting","authors":"Qingyu Chen, Naishuo Wei, Yunlei Fan, Zeyu Wang, Jianguo Zhou, Zening Gao, Yu Chen, Jun Chen","doi":"10.1016/j.atech.2024.100722","DOIUrl":"10.1016/j.atech.2024.100722","url":null,"abstract":"<div><div>For the characteristic of infinite inflorescence, the aim of vibration harvesting of <em>Lycium barbarum</em> L. (<em>L. barbarum</em>) is to picking ripe fruits and leaving unripe fruits, flowers and leaves. Therefore, it is essential to understand the dynamic response of each component during vibration. In this study, the 3D models of a branch, ripe fruit, unripe fruit, flower and leaf were established respectively, which were assembled into a branch-fruit-flower-leaf system model based on the growth characteristics of <em>L. barbarum</em>. The vibration harvester was designed based on the hedge agronomy and simplified as rods, and a rigid-flexible coupling model was established using the kinetic analysis and the finite element method. The dynamic response of branch-fruit-flower-leaf during vibration harvesting was obtained when the rods continuously excited the branch. Results showed that the maximum acceleration of fruit-flower-leaf during vibration harvesting was in the same order of magnitude. And according to the formula of inertia force, the detachment force and mass were calculated and found that the detachment acceleration of ripe fruit was much smaller than that of unripe fruit, flower, and leaf. This study presents a theoretical basis for achieving the harvesting target. Scripts were written to simulate the vibration detachment process of ripe fruit, and high-speed photography was used for experimental verification. The results indicate that the simulation error was 9.15 %, demonstrating that the rigid-flexible coupling simulation is more effective. The field test showed that the picking rate of ripe fruit of 82.69 %, the picking rate of unripe fruit of 3.13 %, and the damage rate of ripe fruit of 4.06 %. The research provides a new analytical approach and theoretical basis for researching vibration harvesting.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100722"},"PeriodicalIF":6.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183127","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}

引用次数: 0

Galvanic Janus paper: A sustainable electrochemical approach for enhanced crop germination and growth

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-16 DOI: 10.1016/j.atech.2024.100724

Jihoon Kim , Keun Ho Lee , Chi-Do Wee , Soosang Chae

New farming methods are needed to boost crop productivity in an environmentally friendly and sustainable way. One promising solution is the use of electrical stimulation, known as electroculture, to enhance plant growth. However, traditional approaches to electroculture often require additional complex electrical equipment, which can increase costs and implementation challenges. We present a simple yet innovative approach to enhance agricultural productivity using a Galvanic Janus Paper, which exploits the natural electrochemical processes between Zn and Cu layers on a paper substrate. When integrated into soil, this material significantly improves seed germination, cotyledon area, and plant height by generating a galvanic voltage that facilitates ion movement and serves as a growth stimulant. In lettuce (Lactuca sativa), the use of a single Galvanic Janus Paper without fertilizer improved germination potential by 26 % and increased leaf area to 64 mm², compared to 50 mm² in the control. Similarly, in pak choi (Brassica rapa subsp. chinensis), germination potential exceeded 25 % by the fourth day, and leaf area increased to 40 mm² under the most effective treatment. This method offers a sustainable, cost-effective alternative for boosting crop yields without the need for external power sources or complex systems, making it a promising solution for large-scale agricultural applications.

{"title":"Galvanic Janus paper: A sustainable electrochemical approach for enhanced crop germination and growth","authors":"Jihoon Kim , Keun Ho Lee , Chi-Do Wee , Soosang Chae","doi":"10.1016/j.atech.2024.100724","DOIUrl":"10.1016/j.atech.2024.100724","url":null,"abstract":"<div><div>New farming methods are needed to boost crop productivity in an environmentally friendly and sustainable way. One promising solution is the use of electrical stimulation, known as electroculture, to enhance plant growth. However, traditional approaches to electroculture often require additional complex electrical equipment, which can increase costs and implementation challenges. We present a simple yet innovative approach to enhance agricultural productivity using a Galvanic Janus Paper, which exploits the natural electrochemical processes between Zn and Cu layers on a paper substrate. When integrated into soil, this material significantly improves seed germination, cotyledon area, and plant height by generating a galvanic voltage that facilitates ion movement and serves as a growth stimulant. In lettuce (<em>Lactuca sativa</em>), the use of a single Galvanic Janus Paper without fertilizer improved germination potential by 26 % and increased leaf area to 64 mm², compared to 50 mm² in the control. Similarly, in pak choi (<em>Brassica rapa subsp. chinensis</em>), germination potential exceeded 25 % by the fourth day, and leaf area increased to 40 mm² under the most effective treatment. This method offers a sustainable, cost-effective alternative for boosting crop yields without the need for external power sources or complex systems, making it a promising solution for large-scale agricultural applications.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100724"},"PeriodicalIF":6.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183197","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}

引用次数: 0

Optimizing dataset diversity for a robust deep-learning model in rice blast disease identification to enhance crop health assessment across diverse conditions

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-16 DOI: 10.1016/j.atech.2024.100726

Reuben Alfred , Judith Leo , Shubi Felix Kaijage

Magnaporthe oryzae, the pathogen that causes rice blast disease, poses a significant global threat to rice production. This disease may lead to yield losses exceeding 30 % in susceptible rice varieties. There is an urgent need for more effective detection solutions, as traditional methods—primarily based on visual inspection—are time-consuming and prone to errors. Deep-learning models presented effective solutions for disease identification due to their ability to analyze large datasets. However, the diversity of the training dataset is significant for optimal performance and generalizability of the model. This study evaluated the impact of dataset diversity on model performance and generalizability by developing two models, referred to in this study as the High-Diverse Model and the Low-Diverse Model. The High-Diverse Model was trained on a diverse dataset comprising images from different geographical regions, rice species, environmental conditions, plant growth stages, and disease severity levels. In contrast, the Low-Diverse Model was trained on a less diverse dataset with significantly limited variability. The results showed that the High-Diverse Model significantly outperformed the Low-Diverse Model, achieving a training accuracy of 95.26 % and a validation accuracy of 94.43 %, indicating effective generalization. The Low-Diverse Model achieved an accuracy of 98.37 % on the training data but only 35.38 % on the validation data, indicating a severe overfitting issue associated with limited dataset diversity. This highlights the importance of dataset diversity in developing effective and scalable deep-learning models for crop health assessment.

引起稻瘟病的病原体 Magnaporthe oryzae 对全球水稻生产构成重大威胁。这种病害可能导致易感水稻品种的产量损失超过 30%。由于传统方法（主要基于目测）耗时且容易出错，因此迫切需要更有效的检测解决方案。深度学习模型因其分析大型数据集的能力，为病害识别提供了有效的解决方案。然而，训练数据集的多样性对模型的最佳性能和通用性至关重要。本研究通过开发两个模型，评估了数据集多样性对模型性能和普适性的影响，这两个模型在本研究中被称为高多样性模型和低多样性模型。高多样性模型是在由不同地理区域、水稻品种、环境条件、植物生长阶段和病害严重程度的图像组成的多样性数据集上进行训练的。相比之下，低多样性模型是在多样性较低的数据集上进行训练的，其可变性非常有限。结果表明，高分辨率模型的表现明显优于低分辨率模型，训练准确率达到 95.26%，验证准确率达到 94.43%，这表明高分辨率模型具有有效的泛化能力。低多样性模型在训练数据上的准确率达到了 98.37%，但在验证数据上的准确率仅为 35.38%，这表明数据集多样性有限导致了严重的过拟合问题。这凸显了数据集多样性在开发有效、可扩展的作物健康评估深度学习模型中的重要性。

{"title":"Optimizing dataset diversity for a robust deep-learning model in rice blast disease identification to enhance crop health assessment across diverse conditions","authors":"Reuben Alfred , Judith Leo , Shubi Felix Kaijage","doi":"10.1016/j.atech.2024.100726","DOIUrl":"10.1016/j.atech.2024.100726","url":null,"abstract":"<div><div><em>Magnaporthe oryzae,</em> the pathogen that causes rice blast disease, poses a significant global threat to rice production. This disease may lead to yield losses exceeding 30 % in susceptible rice varieties. There is an urgent need for more effective detection solutions, as traditional methods—primarily based on visual inspection—are time-consuming and prone to errors. Deep-learning models presented effective solutions for disease identification due to their ability to analyze large datasets. However, the diversity of the training dataset is significant for optimal performance and generalizability of the model. This study evaluated the impact of dataset diversity on model performance and generalizability by developing two models, referred to in this study as the <em>High-Diverse Model</em> and the <em>Low-Diverse Model</em>. The <em>High-Diverse Model</em> was trained on a diverse dataset comprising images from different geographical regions, rice species, environmental conditions, plant growth stages, and disease severity levels. In contrast, the Low-Diverse Model was trained on a less diverse dataset with significantly limited variability. The results showed that the High-Diverse Model significantly outperformed the Low-Diverse Model, achieving a training accuracy of 95.26 % and a validation accuracy of 94.43 %, indicating effective generalization. The Low-Diverse Model achieved an accuracy of 98.37 % on the training data but only 35.38 % on the validation data, indicating a severe overfitting issue associated with limited dataset diversity<em>.</em> This highlights the importance of dataset diversity in developing effective and scalable deep-learning models for crop health assessment.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100726"},"PeriodicalIF":6.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183198","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}

引用次数: 0

Performance analysis of modified DeepLabv3+ architecture for fruit detection and localization in apple orchards 用于苹果园水果检测和定位的改进型 DeepLabv3+ 架构的性能分析

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-16 DOI: 10.1016/j.atech.2024.100729

Prabhakar Maheswari , Purushothaman Raja , Manoj Karkee , Mugundhan Raja , Rahmath Ulla Baig , Kiet Tran Trung , Vinh Truong Hoang

Deep learning plays an important role in automating various operations in fruit crop production including irrigation, nutrition management, yield estimation and harvesting. Yield estimation is essential in fruit crop production as it helps farmers optimize cultivation, harvesting, logistics and marketing operations. Furthermore, fruit detection and localization is a very important step in the development of an automated fruit harvesting system. Hence, an intelligent system was proposed in this study for apple fruit detection and localization using modified DeepLabv3+, semantic segmentation based architecture. The finetuned customizations (such as modifying the activation function, optimization technique and loss function) were performed in the original architecture of DeepLabv3+ and its performance was analyzed. The modified model was trained with the training dataset of 2600 apple tree images. Images were split into 80 % of training and 20 % of validation. The modified architecture was also compared with the other variants of DeepLabv3+ architectures. After training, the model was tested with the unobserved test dataset of 101 images. The test results demonstrated the Mean Accuracy (

M_{A c c}

) of 98.58 % and the Mean Intersection over Union (

M_{I o U}

) of 96.66 % without compromising the inference time (i.e., 15 ms). The proposed model revealed the improved results than the original model which attained a

M_{A c c}

of 92.12 % and

M_{I o U}

of 88.94 % for the same dataset with the inference time of 40 ms. To ascertain further, the modified model was compared with other single stage detectors, including Fully Convolutional Network (FCN) and U-Net. FCN attained a

M_{A c c}

and

M_{I o U}

of 77.5 % and 77.27 %, respectively whereas U-Net resulted a

M_{A c c}

and

M_{I o U}

of 83.95 % and 81.09 %, respectively. Results demonstrated that the modified DeepLabv3+ with ResNet18 is capable of detecting the apple fruits by mitigating the effects of class imbalance which is the major drawback in single stage detectors. Further, better detection and localization of apple fruits can lead to the precise picking by the robotic system.

深度学习在水果作物生产的各种自动化操作中发挥着重要作用，包括灌溉、营养管理、产量估计和收获。产量估算在水果作物生产中至关重要，因为它能帮助农民优化种植、收获、物流和营销操作。此外，水果检测和定位是开发水果自动收获系统的重要步骤。因此，本研究提出了一种智能系统，利用改进的 DeepLabv3+、基于语义分割的架构进行苹果果实检测和定位。对 DeepLabv3+ 的原始架构进行了微调定制（如修改激活函数、优化技术和损失函数），并对其性能进行了分析。修改后的模型使用 2600 张苹果树图像的训练数据集进行训练。图像分为 80% 的训练图像和 20% 的验证图像。修改后的架构还与 DeepLabv3+ 架构的其他变体进行了比较。训练结束后，该模型在包含 101 张图像的未观察测试数据集上进行了测试。测试结果表明，在不影响推理时间（即 15 毫秒）的情况下，平均准确率（MAcc）为 98.58%，平均联合交叉率（MIoU）为 96.66%。在推理时间为 40 毫秒的相同数据集上，原模型的 MAcc 为 92.12%，MIoU 为 88.94%。为了进一步确定结果，我们将修改后的模型与其他单级检测器进行了比较，包括全卷积网络（FCN）和 U-Net。FCN 的 MAcc 和 MIoU 分别为 77.5 % 和 77.27 %，而 U-Net 的 MAcc 和 MIoU 分别为 83.95 % 和 81.09 %。结果表明，改进后的 DeepLabv3+ 与 ResNet18 能够减轻单级检测器的主要缺点--类不平衡的影响，从而检测出苹果水果。此外，对苹果果实进行更好的检测和定位可帮助机器人系统进行精确采摘。

{"title":"Performance analysis of modified DeepLabv3+ architecture for fruit detection and localization in apple orchards","authors":"Prabhakar Maheswari , Purushothaman Raja , Manoj Karkee , Mugundhan Raja , Rahmath Ulla Baig , Kiet Tran Trung , Vinh Truong Hoang","doi":"10.1016/j.atech.2024.100729","DOIUrl":"10.1016/j.atech.2024.100729","url":null,"abstract":"<div><div>Deep learning plays an important role in automating various operations in fruit crop production including irrigation, nutrition management, yield estimation and harvesting. Yield estimation is essential in fruit crop production as it helps farmers optimize cultivation, harvesting, logistics and marketing operations. Furthermore, fruit detection and localization is a very important step in the development of an automated fruit harvesting system. Hence, an intelligent system was proposed in this study for apple fruit detection and localization using modified DeepLabv3+, semantic segmentation based architecture. The finetuned customizations (such as modifying the activation function, optimization technique and loss function) were performed in the original architecture of DeepLabv3+ and its performance was analyzed. The modified model was trained with the training dataset of 2600 apple tree images. Images were split into 80 % of training and 20 % of validation. The modified architecture was also compared with the other variants of DeepLabv3+ architectures. After training, the model was tested with the unobserved test dataset of 101 images. The test results demonstrated the Mean Accuracy (<span><math><msub><mi>M</mi><mrow><mi>A</mi><mi>c</mi><mi>c</mi></mrow></msub></math></span>) of 98.58 % and the Mean Intersection over Union (<span><math><msub><mi>M</mi><mrow><mi>I</mi><mi>o</mi><mi>U</mi></mrow></msub></math></span>) of 96.66 % without compromising the inference time (i.e., 15 ms). The proposed model revealed the improved results than the original model which attained a <span><math><msub><mi>M</mi><mrow><mi>A</mi><mi>c</mi><mi>c</mi></mrow></msub></math></span> of 92.12 % and <span><math><msub><mi>M</mi><mrow><mi>I</mi><mi>o</mi><mi>U</mi></mrow></msub></math></span> of 88.94 % for the same dataset with the inference time of 40 ms. To ascertain further, the modified model was compared with other single stage detectors, including Fully Convolutional Network (FCN) and U-Net. FCN attained a <span><math><msub><mi>M</mi><mrow><mi>A</mi><mi>c</mi><mi>c</mi></mrow></msub></math></span>and<span><math><msub><mi>M</mi><mrow><mi>I</mi><mi>o</mi><mi>U</mi></mrow></msub></math></span> of 77.5 % and 77.27 %, respectively whereas U-Net resulted a <span><math><msub><mi>M</mi><mrow><mi>A</mi><mi>c</mi><mi>c</mi></mrow></msub></math></span> and <span><math><msub><mi>M</mi><mrow><mi>I</mi><mi>o</mi><mi>U</mi></mrow></msub></math></span> of 83.95 % and 81.09 %, respectively. Results demonstrated that the modified DeepLabv3+ with ResNet18 is capable of detecting the apple fruits by mitigating the effects of class imbalance which is the major drawback in single stage detectors. Further, better detection and localization of apple fruits can lead to the precise picking by the robotic system.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100729"},"PeriodicalIF":6.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143182108","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}

引用次数: 0

Crop yield prediction using machine learning: An extensive and systematic literature review

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-16 DOI: 10.1016/j.atech.2024.100718

Sarowar Morshed Shawon , Falguny Barua Ema , Asura Khanom Mahi , Fahima Lokman Niha , H.T. Zubair

In recent years, agriculture has gained much attention regarding forecasting and prediction with the advancement of artificial intelligence techniques. Advancements in Machine Learning (ML) have significantly improved agricultural activities. In order to ensure food security and optimize resource allocation, precise crop yield prediction has become essential due to the growing global population and the effects of climate change on agricultural production. In this research, we conducted a Systematic Literature Review (SLR) to identify and synthesize techniques and attributes utilized in crop yield prediction research between the years of 2017 and 2024. This extensive search yielded 184 eligible papers from eight electronic sources. A total of 97 papers have been chosen for further analysis based on inclusion and exclusion criteria. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) technique has been employed to search, screen, and select relevant research papers, resulting in a comprehensive and unbiased review. According to this analysis, the most used features are temperature, soil type, and vegetation. Also, the most applied machine learning algorithms are Linear Regression (LR), Random Forest (RF), and Gradient Boosting Trees (GBT) whereas the most applied deep learning algorithms are Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM). An additional search has been performed in order to identify some hybrid ML-based models. As per the evaluation metrics, RMSE, R-square, and MAE are found to be the mostly favored. Eventually, this review offers valuable insights into the state-of-the-art algorithms in crop yield prediction and suggests future directions for researchers aiming to address existing difficulties and limitations.

{"title":"Crop yield prediction using machine learning: An extensive and systematic literature review","authors":"Sarowar Morshed Shawon , Falguny Barua Ema , Asura Khanom Mahi , Fahima Lokman Niha , H.T. Zubair","doi":"10.1016/j.atech.2024.100718","DOIUrl":"10.1016/j.atech.2024.100718","url":null,"abstract":"<div><div>In recent years, agriculture has gained much attention regarding forecasting and prediction with the advancement of artificial intelligence techniques. Advancements in Machine Learning (ML) have significantly improved agricultural activities. In order to ensure food security and optimize resource allocation, precise crop yield prediction has become essential due to the growing global population and the effects of climate change on agricultural production. In this research, we conducted a Systematic Literature Review (SLR) to identify and synthesize techniques and attributes utilized in crop yield prediction research between the years of 2017 and 2024. This extensive search yielded 184 eligible papers from eight electronic sources. A total of 97 papers have been chosen for further analysis based on inclusion and exclusion criteria. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) technique has been employed to search, screen, and select relevant research papers, resulting in a comprehensive and unbiased review. According to this analysis, the most used features are temperature, soil type, and vegetation. Also, the most applied machine learning algorithms are Linear Regression (LR), Random Forest (RF), and Gradient Boosting Trees (GBT) whereas the most applied deep learning algorithms are Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM). An additional search has been performed in order to identify some hybrid ML-based models. As per the evaluation metrics, RMSE, R-square, and MAE are found to be the mostly favored. Eventually, this review offers valuable insights into the state-of-the-art algorithms in crop yield prediction and suggests future directions for researchers aiming to address existing difficulties and limitations.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100718"},"PeriodicalIF":6.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183129","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}

引用次数: 0

A motion planning method for winter jujube harvesting robotic arm based on optimized Informed-RRT* algorithm

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-16 DOI: 10.1016/j.atech.2024.100732

Anxiang Huang , Chenhao Yu , Junzhe Feng , Xing Tong , Ayanori Yorozu , Akihisa Ohya , Yaohua Hu

Winter jujube is a fruit that is rich in nutritional value and has a delicious taste. Since the ripe winter jujubes are tender and easy to be damaged, during the picking process the mechanical arm needs to maintain smooth and stable movement to avoid the impact of vibration or oscillation on the picking task. However, existing motion planning algorithms may not guarantee the smoothness and stability of the mechanical arm's movement. The research discovered a motion planning method based on the optimized Informed-RRT* algorithm. By adding target bias, adaptive step size, and pruning strategies, the optimization of search paths was achieved, reducing unnecessary movement of the mechanical arm during operation. This can ensure high picking success rate and low damage rate. Through comparative experiments, the optimized Informed-RRT* algorithm has good performance in the two-dimensional and three-dimensional spaces. Within the specified time, the planned paths by the optimized Informed-RRT* algorithm are shorter and smoother, significantly improving the efficiency of the mechanical arm. Additionally, this research deploys the optimized Informed-RRT* algorithm to the Robot Operating System (ROS) and conducts three-dimensional modeling of the picking environment through Moveit! to obtain real-time environmental information and obstacle detection. This allows for effective avoidance of obstacles while ensuring the optimal path. To ensure the safety of the mechanical arm's movement, this research monitors the position changes of each joint in real-time. The results indicated that during the movement process, the angular velocity and angular acceleration of the mechanical arm exhibit smooth and continuous trends, demonstrating good dynamic stability and control performance during movement and further proving the effectiveness of the optimized Informed-RRT* algorithm.

{"title":"A motion planning method for winter jujube harvesting robotic arm based on optimized Informed-RRT* algorithm","authors":"Anxiang Huang , Chenhao Yu , Junzhe Feng , Xing Tong , Ayanori Yorozu , Akihisa Ohya , Yaohua Hu","doi":"10.1016/j.atech.2024.100732","DOIUrl":"10.1016/j.atech.2024.100732","url":null,"abstract":"<div><div>Winter jujube is a fruit that is rich in nutritional value and has a delicious taste. Since the ripe winter jujubes are tender and easy to be damaged, during the picking process the mechanical arm needs to maintain smooth and stable movement to avoid the impact of vibration or oscillation on the picking task. However, existing motion planning algorithms may not guarantee the smoothness and stability of the mechanical arm's movement. The research discovered a motion planning method based on the optimized Informed-RRT* algorithm. By adding target bias, adaptive step size, and pruning strategies, the optimization of search paths was achieved, reducing unnecessary movement of the mechanical arm during operation. This can ensure high picking success rate and low damage rate. Through comparative experiments, the optimized Informed-RRT* algorithm has good performance in the two-dimensional and three-dimensional spaces. Within the specified time, the planned paths by the optimized Informed-RRT* algorithm are shorter and smoother, significantly improving the efficiency of the mechanical arm. Additionally, this research deploys the optimized Informed-RRT* algorithm to the Robot Operating System (ROS) and conducts three-dimensional modeling of the picking environment through Moveit! to obtain real-time environmental information and obstacle detection. This allows for effective avoidance of obstacles while ensuring the optimal path. To ensure the safety of the mechanical arm's movement, this research monitors the position changes of each joint in real-time. The results indicated that during the movement process, the angular velocity and angular acceleration of the mechanical arm exhibit smooth and continuous trends, demonstrating good dynamic stability and control performance during movement and further proving the effectiveness of the optimized Informed-RRT* algorithm.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100732"},"PeriodicalIF":6.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183124","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}

引用次数: 0

Prediction of copper contamination in soil across EU using spectroscopy and machine learning: Handling class imbalance problem

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-16 DOI: 10.1016/j.atech.2024.100728

Chongchong Qi , Nana Zhou , Tao Hu , Mengting Wu , Qiusong Chen , Han Wang , Kejing Zhang , Zhang Lin

Soil copper (Cu) pollution is a significant global environmental challenge, necessitating accurate assessment methods for effective control. However, existing classification approaches for Cu content in soil spectral datasets often face imbalances in data distribution, resulting in unreliable identification of Cu-contaminated samples. To address this limitation, we conducted a comprehensive evaluation of three basic machine learning (ML) algorithms and four imbalanced ML algorithms. These methods were used to develop seven continental-scale models for imbalanced classification of soil Cu contamination using visible and near-infrared reflectance spectroscopy. A dataset comprising 18,675 topsoil samples was utilized for training and validation. Hyperparameter optimization was applied to enhance model performance, and multiple statistical metrics were employed for evaluation. Furthermore, feature importance analysis identified key spectral bands influencing Cu classification. Among the tested models, the BalancedRandomForest algorithm demonstrated superior classification performance and generalization ability, achieving an area under the curve of 0.870, recall of 0.816, and balanced accuracy of 0.793. Spectral analysis highlighted the 2310–2320 nm as the most critical spectral region for Cu classification. This study underscores the utility of the optimized model for managing soil Cu pollution and provides a valuable reference for addressing imbalanced learning challenges in soil pollution research.

{"title":"Prediction of copper contamination in soil across EU using spectroscopy and machine learning: Handling class imbalance problem","authors":"Chongchong Qi , Nana Zhou , Tao Hu , Mengting Wu , Qiusong Chen , Han Wang , Kejing Zhang , Zhang Lin","doi":"10.1016/j.atech.2024.100728","DOIUrl":"10.1016/j.atech.2024.100728","url":null,"abstract":"<div><div>Soil copper (Cu) pollution is a significant global environmental challenge, necessitating accurate assessment methods for effective control. However, existing classification approaches for Cu content in soil spectral datasets often face imbalances in data distribution, resulting in unreliable identification of Cu-contaminated samples. To address this limitation, we conducted a comprehensive evaluation of three basic machine learning (ML) algorithms and four imbalanced ML algorithms. These methods were used to develop seven continental-scale models for imbalanced classification of soil Cu contamination using visible and near-infrared reflectance spectroscopy. A dataset comprising 18,675 topsoil samples was utilized for training and validation. Hyperparameter optimization was applied to enhance model performance, and multiple statistical metrics were employed for evaluation. Furthermore, feature importance analysis identified key spectral bands influencing Cu classification. Among the tested models, the BalancedRandomForest algorithm demonstrated superior classification performance and generalization ability, achieving an area under the curve of 0.870, recall of 0.816, and balanced accuracy of 0.793. Spectral analysis highlighted the 2310–2320 nm as the most critical spectral region for Cu classification. This study underscores the utility of the optimized model for managing soil Cu pollution and provides a valuable reference for addressing imbalanced learning challenges in soil pollution research.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100728"},"PeriodicalIF":6.3,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183260","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}

引用次数: 0

Research on detection of wheat tillers in natural environment based on YOLOv8-MRF 基于 YOLOv8-MRF 的自然环境中小麦分蘖检测研究

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-15 DOI: 10.1016/j.atech.2024.100720

Min Liang , Yuchen Zhang , Jian Zhou , Fengcheng Shi , Zhiqiang Wang , Yu Lin , Liang Zhang , Yaxi Liu

To bolster agricultural efficiency and precision, this study introduces the YOLOv8-MRF model (multi-path coordinate attention, receptive field attention convolution, and Focaler-CIoU-optimized YOLOv8), a groundbreaking advancement in automated detection of wheat tillers. This model transcends traditional manual methods prone to subjectivity and inefficiency. This approach integrates an enhanced multi-path coordinate attention (MPCA) mechanism within the backbone network, capturing multi-scale features and significantly elevating tillers recognition. The innovative replacement of the CSPDarknet53 to 2-Stage FPN (C2F) module with receptive field attention convolution (RFCAConv) addresses parameter-sharing limitations, accentuating feature significance, and amplifying network performance. Coupled with the Focaler-CIoU loss for superior detection accuracy, YOLOv8-MRF outperforms RTDETR, YOLOv5, YOLOv7, and YOLOv8 by impressive margins in mAP50, while operating with merely 11 % of the parameters of YOLOv7, achieving a detection precision of 91.7 %, and with enhancements of 2.5 % in precision, 5.5 % in recall, and 4.1 % in mAP50 over the original model. The experimental results demonstrate that this method can realize tillering detection under complex backgrounds, contributing to advancing intelligent farming practices for wheat. Importantly, the YOLOv8-MRF model not only achieves significant technological advancements but also shows strong potential in practical applications, providing an effective tool for agricultural automation and intelligence, which could become pivotal in the development of future precision agriculture technologies.

{"title":"Research on detection of wheat tillers in natural environment based on YOLOv8-MRF","authors":"Min Liang , Yuchen Zhang , Jian Zhou , Fengcheng Shi , Zhiqiang Wang , Yu Lin , Liang Zhang , Yaxi Liu","doi":"10.1016/j.atech.2024.100720","DOIUrl":"10.1016/j.atech.2024.100720","url":null,"abstract":"<div><div>To bolster agricultural efficiency and precision, this study introduces the YOLOv8-MRF model (multi-path coordinate attention, receptive field attention convolution, and Focaler-CIoU-optimized YOLOv8), a groundbreaking advancement in automated detection of wheat tillers. This model transcends traditional manual methods prone to subjectivity and inefficiency. This approach integrates an enhanced multi-path coordinate attention (MPCA) mechanism within the backbone network, capturing multi-scale features and significantly elevating tillers recognition. The innovative replacement of the CSPDarknet53 to 2-Stage FPN (C2F) module with receptive field attention convolution (RFCAConv) addresses parameter-sharing limitations, accentuating feature significance, and amplifying network performance. Coupled with the Focaler-CIoU loss for superior detection accuracy, YOLOv8-MRF outperforms RTDETR, YOLOv5, YOLOv7, and YOLOv8 by impressive margins in mAP50, while operating with merely 11 % of the parameters of YOLOv7, achieving a detection precision of 91.7 %, and with enhancements of 2.5 % in precision, 5.5 % in recall, and 4.1 % in mAP50 over the original model. The experimental results demonstrate that this method can realize tillering detection under complex backgrounds, contributing to advancing intelligent farming practices for wheat. Importantly, the YOLOv8-MRF model not only achieves significant technological advancements but also shows strong potential in practical applications, providing an effective tool for agricultural automation and intelligence, which could become pivotal in the development of future precision agriculture technologies.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100720"},"PeriodicalIF":6.3,"publicationDate":"2024-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183220","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}

引用次数: 0

Towards rigorous dataset quality standards for deep learning tasks in precision agriculture: A case study exploration 为精准农业中的深度学习任务制定严格的数据集质量标准：案例研究探索

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-15 DOI: 10.1016/j.atech.2024.100721

A. Carraro , G. Saurio , F. Marinello

Deep Learning (DL) through Convolutional Neural Networks (CNNs) has emerged as a critical player in classifying plant diseases from images. This prominence has intensified the demand for a substantial volume of annotated training data. However, acquiring such data is costly and intricate, fraught with subtle challenges. In the domain of plants, where data collection can be even more complex, this study scrutinises how one dataset was gathered. Specifically, it delves into the nuances of collecting images of grapevine leaves in an open field for a binary classification task, discerning the presence or absence of Esca disease.

Adherence to rigorous dataset quality standards during image collection is paramount in precision agriculture. Errors made in this phase can have devastating repercussions on all subsequent work. For instance, collections of photos may exhibit a consistent disparity in background characteristics between images belonging to different classes. This persistent difference can lead a deep-learning algorithm to learn undesired correlations, even though the algorithm's performances are excellent because the train and test sets possess the same kind of disparity.

通过卷积神经网络（CNN）进行的深度学习（DL）已成为从图像中对植物病害进行分类的重要手段。这一显著地位加剧了对大量注释训练数据的需求。然而，获取这些数据既昂贵又复杂，充满了微妙的挑战。在植物领域，数据收集工作可能更加复杂，本研究仔细研究了一个数据集的收集过程。具体来说，本研究深入探讨了在露天田野中收集葡萄叶片图像的细微差别，以完成二元分类任务，辨别是否存在埃斯卡病。在这一阶段出现的错误会对所有后续工作产生破坏性影响。例如，收集的照片可能会显示出属于不同类别的图像在背景特征上的持续差异。这种持续存在的差异会导致深度学习算法学习到不想要的相关性，即使该算法的性能非常出色，因为训练集和测试集具有相同的差异。

{"title":"Towards rigorous dataset quality standards for deep learning tasks in precision agriculture: A case study exploration","authors":"A. Carraro , G. Saurio , F. Marinello","doi":"10.1016/j.atech.2024.100721","DOIUrl":"10.1016/j.atech.2024.100721","url":null,"abstract":"<div><div>Deep Learning (DL) through Convolutional Neural Networks (CNNs) has emerged as a critical player in classifying plant diseases from images. This prominence has intensified the demand for a substantial volume of annotated training data. However, acquiring such data is costly and intricate, fraught with subtle challenges. In the domain of plants, where data collection can be even more complex, this study scrutinises how one dataset was gathered. Specifically, it delves into the nuances of collecting images of grapevine leaves in an open field for a binary classification task, discerning the presence or absence of Esca disease.</div><div>Adherence to rigorous dataset quality standards during image collection is paramount in precision agriculture. Errors made in this phase can have devastating repercussions on all subsequent work. For instance, collections of photos may exhibit a consistent disparity in background characteristics between images belonging to different classes. This persistent difference can lead a deep-learning algorithm to learn undesired correlations, even though the algorithm's performances are excellent because the train and test sets possess the same kind of disparity.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100721"},"PeriodicalIF":6.3,"publicationDate":"2024-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183196","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}

引用次数: 0

A review of model predictive control in precision agriculture 精准农业中的模型预测控制综述

IF 6.3 Q1 AGRICULTURAL ENGINEERING

Smart agricultural technology

Pub Date : 2024-12-15 DOI: 10.1016/j.atech.2024.100716

Erion Bwambale , Joshua Wanyama , Thomas Apusiga Adongo , Etienne Umukiza , Romain Ntole , Sylvester R. Chikavumbwa , Davis Sibale , Zechariah Jeremaih

Precision agriculture, driven by advanced technologies and data-driven decision-making, has emerged as a transformative approach to address global food demand, resource constraints, and sustainability challenges. In this context, Model Predictive Control (MPC) has garnered significant attention as a powerful control strategy capable of optimizing farming processes through predictive and anticipatory control actions. This review comprehensively explores the fundamentals and applications of MPC in precision agriculture. The review begins with an overview of MPC's principles, formulation, and optimization techniques, emphasizing its predictive and adaptable nature. Subsequently, it delves into the diverse applications of MPC in precision agriculture, including crop growth and yield optimization, pest and disease management, and autonomous machinery and robotics. The integration of MPC with precision agriculture machinery and its role in autonomous farming systems are also explored. Success stories and case studies highlight real-world applications of MPC, showcasing its positive impact on crop yields, resource utilization, and economic viability. Additionally, demonstrated benefits such as water conservation, reduced chemical usage, and improved produce quality attest to the significance of MPC in sustainable farming practices. While MPC offers numerous advantages, the review also discusses challenges, such as computational complexity, model uncertainty, and sensor reliability. The review concludes by underscoring MPC's potential in driving precision agriculture towards a more sustainable, efficient, and technologically advanced future.

在先进技术和数据驱动决策的推动下，精准农业已成为应对全球粮食需求、资源限制和可持续发展挑战的变革性方法。在此背景下，模型预测控制（MPC）作为一种强大的控制策略，能够通过预测和预期控制行动优化农业生产流程，因而备受关注。本综述全面探讨了 MPC 在精准农业中的基本原理和应用。综述首先概述了 MPC 的原理、配方和优化技术，强调了其预测性和适应性。随后，文章深入探讨了 MPC 在精准农业中的各种应用，包括作物生长和产量优化、病虫害管理以及自主机械和机器人技术。此外，还探讨了 MPC 与精准农业机械的集成及其在自主耕作系统中的作用。成功案例和案例研究强调了 MPC 在现实世界中的应用，展示了其对作物产量、资源利用和经济可行性的积极影响。此外，节水、减少化学品用量和提高农产品质量等效益也证明了多用途植保在可持续农业实践中的重要意义。虽然 MPC 具有众多优势，但综述也讨论了其面临的挑战，如计算复杂性、模型不确定性和传感器可靠性。综述最后强调了 MPC 在推动精准农业迈向更可持续、更高效、技术更先进的未来方面的潜力。

{"title":"A review of model predictive control in precision agriculture","authors":"Erion Bwambale , Joshua Wanyama , Thomas Apusiga Adongo , Etienne Umukiza , Romain Ntole , Sylvester R. Chikavumbwa , Davis Sibale , Zechariah Jeremaih","doi":"10.1016/j.atech.2024.100716","DOIUrl":"10.1016/j.atech.2024.100716","url":null,"abstract":"<div><div>Precision agriculture, driven by advanced technologies and data-driven decision-making, has emerged as a transformative approach to address global food demand, resource constraints, and sustainability challenges. In this context, Model Predictive Control (MPC) has garnered significant attention as a powerful control strategy capable of optimizing farming processes through predictive and anticipatory control actions. This review comprehensively explores the fundamentals and applications of MPC in precision agriculture. The review begins with an overview of MPC's principles, formulation, and optimization techniques, emphasizing its predictive and adaptable nature. Subsequently, it delves into the diverse applications of MPC in precision agriculture, including crop growth and yield optimization, pest and disease management, and autonomous machinery and robotics. The integration of MPC with precision agriculture machinery and its role in autonomous farming systems are also explored. Success stories and case studies highlight real-world applications of MPC, showcasing its positive impact on crop yields, resource utilization, and economic viability. Additionally, demonstrated benefits such as water conservation, reduced chemical usage, and improved produce quality attest to the significance of MPC in sustainable farming practices. While MPC offers numerous advantages, the review also discusses challenges, such as computational complexity, model uncertainty, and sensor reliability. The review concludes by underscoring MPC's potential in driving precision agriculture towards a more sustainable, efficient, and technologically advanced future.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100716"},"PeriodicalIF":6.3,"publicationDate":"2024-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183194","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}

引用次数: 0