Smart agricultural technology最新文献

{"title":"3D neural architecture search to optimize segmentation of plant parts","authors":"Farah Saeed ,&nbsp;Chenjiao Tan ,&nbsp;Tianming Liu ,&nbsp;Changying Li","doi":"10.1016/j.atech.2025.100776","DOIUrl":"10.1016/j.atech.2025.100776","url":null,"abstract":"<div><div>Accurately segmenting plant parts from imagery is vital for improving crop phenotypic traits. However, current 3D deep learning models for segmentation in point cloud data require specific network architectures that are usually manually designed, which is both tedious and suboptimal. To overcome this issue, a 3D neural architecture search (NAS) was performed in this study to optimize cotton plant part segmentation. The search space was designed using Point Voxel Convolution (PVConv) as the basic building block of the network. The NAS framework included a supernetwork with weight sharing and an evolutionary search to find optimal candidates, with three surrogate learners to predict mean IoU, latency, and memory footprint. The optimal candidate searched from the proposed method consisted of five PVConv layers with either 32 or 512 output channels, achieving mean IoU and accuracy of over 90 % and 96 %, respectively, and outperforming manually designed architectures. Additionally, the evolutionary search was updated to search for architectures satisfying memory and time constraints, with searched architectures achieving mean IoU and accuracy of &gt;84 % and 94 %, respectively. Furthermore, a differentiable architecture search (DARTS) utilizing PVConv operation was implemented for comparison, and our method demonstrated better segmentation performance with a margin of &gt;2 % and 1 % in mean IoU and accuracy, respectively. Overall, the proposed method can be applied to segment cotton plants with an accuracy over 94 %, while adjusting to available resource constraints.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100776"},"PeriodicalIF":6.3,"publicationDate":"2025-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181948","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}

{"title":"Economic and environmental benefits of digital agricultural technological solutions in livestock farming: A review","authors":"George Papadopoulos ,&nbsp;Maria-Zoi Papantonatou ,&nbsp;Havva Uyar ,&nbsp;Olga Kriezi ,&nbsp;Alexandros Mavrommatis ,&nbsp;Vasilis Psiroukis ,&nbsp;Aikaterini Kasimati ,&nbsp;Eleni Tsiplakou ,&nbsp;Spyros Fountas","doi":"10.1016/j.atech.2025.100783","DOIUrl":"10.1016/j.atech.2025.100783","url":null,"abstract":"<div><div>This review paper delved into the economic and environmental benefits of Digital Agricultural Technological Solutions (DATSs) in livestock farming systems. Synthesising data from 52 peer-reviewed papers it presents the outcomes of a systematic literature review on livestock farming DATSs, conducted with the use of the PRISMA methodology. The analysis highlighted the contribution of DATSs across three main livestock farming DATSs categories: Automated Milking Systems (AMS), Feed and Live Weight Measurement technologies, and Health Monitoring Systems. The results showed that AMS has the potential to boost cow productivity by up to 15 % while also reducing energy consumption by 35 %. Feed and Live Weight Measurement technologies contribute notably to sustainability and cost savings, with feed waste reductions of 75 % and feeding savings of 33 %. Health Monitoring Systems are especially effective in improving herd health and productivity through early detection of clinical issues, which directly enhances animal welfare and farm efficiency. Environmentally, AMS and health monitoring tools play a vital role in reducing greenhouse gas emissions, with AMS lowering global warming potential by up to 5.83 %. Overall, the findings of this review highlight the potentials of livestock DATSs towards economic viability and environmental sustainability, suggesting that the wider adoption could offer substantial benefits for the livestock farming sector. Up to now, DATSs have shown great potential in dairy cattle by improving milk yield, quality, and animal health, with advancements such as AMS increasing productivity and health monitoring systems enhancing early disease detection. In contrast, their application in sheep, goats, and pigs is still in its early stages, mainly limited to basic health monitoring and feeding technologies, despite the economic importance of these species, especially in the Mediterranean area, where most of the studies are conducted.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100783"},"PeriodicalIF":6.3,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143182111","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}

{"title":"Non-invasive prediction of sweet cherry soluble solids content using dielectric spectroscopy and down-sampling techniques","authors":"Kamil Sacilik ,&nbsp;Necati Cetin ,&nbsp;Burak Ozbey ,&nbsp;Fernando Auat Cheein","doi":"10.1016/j.atech.2025.100782","DOIUrl":"10.1016/j.atech.2025.100782","url":null,"abstract":"<div><div>The soluble solid content (SSC) in fruits significantly influences consumers' taste, aroma, and flavor preferences. It also plays a crucial role for farmers and wholesalers in determining the optimal harvest period for marketing. Dielectric spectroscopy, an innovative and non-invasive technique, has shown promise for various applications in the food and agriculture sectors. This study introduces an open-ended coaxial line probe measurement system to non-invasively determine the SSC of sweet cherries at different radio and microwave frequencies. Key parameters such as the dielectric constant (ε′), loss factor (ε′′), loss tangent (tan δ), and SSC of sweet cherries were measured across different harvest periods. The dielectric property frequency ranges were down-sampled from 300 MHz to 15 MHz. Using dielectric spectroscopy, we implemented predictive models: support vector regression (SVR) and multilayer perceptron (MLP), that demonstrated extremely low MAE and RMSE, with correlation coefficients (R) exceeding 0.97 for SVR and 0.96 for MLP. The down-sampled frequency ranges for dielectric properties yielded consistently high performance across all subsets, demonstrating comparable results. These findings suggest that a dielectric measurement system designed for SSC estimation using fewer frequencies could effectively reduce costs while maintaining accuracy.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100782"},"PeriodicalIF":6.3,"publicationDate":"2025-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181510","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}

{"title":"AI-powered cow detection in complex farm environments","authors":"Voncarlos M. Araújo ,&nbsp;Ines Rili ,&nbsp;Thomas Gisiger ,&nbsp;Sébastien Gambs ,&nbsp;Elsa Vasseur ,&nbsp;Marjorie Cellier ,&nbsp;Abdoulaye Baniré Diallo","doi":"10.1016/j.atech.2025.100770","DOIUrl":"10.1016/j.atech.2025.100770","url":null,"abstract":"<div><div>Animal welfare has become a critical issue in contemporary society, emphasizing our ethical responsibilities toward animals, particularly within livestock farming. In addition, the advent of Artificial Intelligence (AI) technologies, specifically computer vision, offers a innovative approach to monitoring and enhancing animal welfare. Cows, as essential contributors to sustainable agriculture and climate management, being a central part of it. However, existing cow detection algorithms face significant challenges in real-world farming environments, such as complex lighting, occlusions, pose variations and background interference, which hinder accurate and reliable detection. Additionally, the model generalization power is highly desirable as it enables the model to adapt and perform well across different contexts and conditions, beyond its training environment or dataset. This study addresses these challenges in diverse cow dataset composed of six different environments, including indoor and outdoor scenarios. More precisely, we propose a novel detection model that combines YOLOv8 with the CBAM (Convolutional Block Attention Module) and assess its performance against baseline models, including Mask R-CNN, YOLOv5 and YOLOv8. Our findings indicate that while baseline models show promise, their performance degrades in complex real-world conditions, which our approach improves using the CBAM attention module. Overall, YOLOv8-CBAM outperformed YOLOv8 by 2.3% in mAP across all camera types, achieving a precision of 95.2% and an [email protected]:0.95 of 82.6%, demonstrating superior generalization and enhanced detection accuracy in complex backgrounds. Thus, the primary contributions of this research are: (1) providing an in-depth analysis of current limitations in cow detection under challenging indoor and outdoor environments, (2) proposing a robust general model that effectively detects cows in complex real-world conditions and (3) evaluating and benchmarking state-of-the-art detection algorithms. Potential application scenarios of the model include automated health monitoring, behavioral analysis and tracking within smart farm management systems, enabling precise detection of individual cows, even in challenging environments. By addressing these critical challenges, this study paves the way for future innovations in AI-driven livestock monitoring, aiming to improve the welfare and management of farm animals while advancing smart agriculture.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100770"},"PeriodicalIF":6.3,"publicationDate":"2025-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181509","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}

{"title":"Multi-task neural networks for multi-step soil moisture forecasting in vineyards using Internet-of-Things sensors","authors":"Ada Baldi ,&nbsp;Laura Carnevali ,&nbsp;Giovanni Collodi ,&nbsp;Marco Lippi ,&nbsp;Antonio Manes","doi":"10.1016/j.atech.2025.100769","DOIUrl":"10.1016/j.atech.2025.100769","url":null,"abstract":"<div><div>Promoting an efficient management of water resources is one of the most crucial challenges in smart farming for the coming years. In this context, developing accurate soil moisture forecasting methods is fundamental in order to optimize irrigation and avoid waste. In this paper, we present a deep learning approach based on the multi-task paradigm, which is exploited to jointly forecast soil moisture at multiple time steps in the future, using a multivariate time-series as input features. Experiments are conducted on a real data set collected via data fusion techniques from Internet-of-Things (IoT) sensors located in a vineyard in Montalcino (Tuscany), showing the advantages of joint multi-step forecasting for prediction horizons that range from 24 to 48 hours ahead.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100769"},"PeriodicalIF":6.3,"publicationDate":"2025-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181954","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}

{"title":"Data generation using Pix2Pix to improve YOLO v8 performance in UAV-based Yuzu detection","authors":"Zhen Zhang ,&nbsp;Yuu Tanimoto ,&nbsp;Makoto Iwata ,&nbsp;Shinichi Yoshida","doi":"10.1016/j.atech.2025.100777","DOIUrl":"10.1016/j.atech.2025.100777","url":null,"abstract":"<div><div>Unmanned aerial vehicle (UAV) detection using deep learning techniques plays a crucial role in the pre-harvest estimation of yuzu (Citrus Junos) yield. However, the detection performance of deep learning models heavily depends on the quantity and quality of training data. One of the current challenges is that the work of labeling data is difficult and expensive, because of the high density of fruits, the similarity in color between fruits and leaves, and the varying lighting conditions in the captured images of fruit trees. To address these challenges, we propose to use generative adversarial networks (GANs) for data generation, and then utilize the generated data to improve the yuzu detection performance of YOLO (You Only Look Once) v8 models.</div><div>In this study, the experimental images were photographed using UAVs from two orchards of Kochi agricultural research center between 2020 and 2022. In our approach, we first trained a conditional GAN called Pix2Pix using pairs of images, where the training inputs are the images of fruit trees with all fruits removed, and the training targets are the original images. Subsequently, we created new regions of interest on the images of fruit trees and used the trained Pix2Pix network to generate yuzu fruits within these regions, thereby generating new labeled images. In the experiments, we merged real and generated images to train YOLO v8-series models and explored to reduce the dependency on real training images through the proposed data augmentation approach.</div><div>The results showed that the combined training of these generated and real images can significantly improve the detection performance of YOLO v8-series models, with the maximum improvements of 5.4% in F1-scores, 5.6% in mAP50, and 7.1% in mAP50–90, respectively. Moreover, the proposed data augmentation approach allowed for up to a 50% reduction in the amount of real training images while still achieving improved detection results.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100777"},"PeriodicalIF":6.3,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181508","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}

{"title":"Design and experiment of active obstacle avoidance control system for grapevine interplant weeding based on GNSS","authors":"Hao Zhang,&nbsp;Zejiang Meng,&nbsp;Shiwei Wen,&nbsp;Guangyao Liu,&nbsp;Guangrui Hu,&nbsp;Jun Chen,&nbsp;Shuo Zhang","doi":"10.1016/j.atech.2025.100781","DOIUrl":"10.1016/j.atech.2025.100781","url":null,"abstract":"<div><div>Traditional passive obstacle avoidance mechanical weeding strategies heavily relied on touch rods, which led to a high crop damage rate and low weeding efficiency during operations. This study proposed an obstacle avoidance information collection scheme that integrates precise detection of obstacle positions and coordinate conversion of weeding tool positions. An active obstacle avoidance control system based on obstacle positions and real-time tool status was designed. This system consisted of the autonomous navigation equipment, obstacle avoidance information collection units, the control system module, hydraulic execution components, and the real-time monitoring sensor. Based on the requirements for active obstacle avoidance, the study established the relationship between the obstacle avoidance information collection units, hydraulic execution components, and the real-time monitoring sensor, and determined a precise active obstacle avoidance control scheme. Field tests were conducted using machine forward speed as the test factor, with inter-row weeding coverage rate and plant damage rate as evaluation indicators. The test results indicated that when the machine forward speed was 460 mm/s, the combined effect of inter-row weeding coverage and operational efficiency was optimal, with an average inter-row weeding coverage rate of 94.62 % and a plant damage rate of 1.94 %. The active obstacle avoidance weeding scheme proposed in this study provided a technical reference for improving inter-row weeding effectiveness in orchards.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100781"},"PeriodicalIF":6.3,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181951","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}

{"title":"IoT-enabled solar-powered smart irrigation for precision agriculture","authors":"Md. Rasel Al Mamun,&nbsp;Abu Kawsar Ahmed,&nbsp;Sidratul Muntaha Upoma,&nbsp;Md．Mashurul Haque,&nbsp;Muhammad Ashik-E-Rabbani","doi":"10.1016/j.atech.2025.100773","DOIUrl":"10.1016/j.atech.2025.100773","url":null,"abstract":"<div><div>The Internet of Things (IoT) can enable the fourth industrial revolution, significantly boosting production and efficiency in the agricultural sector by optimizing farming practices. This research aims to develop a solar-powered IoT irrigating system. The system comprised a 20W solar panel for powering the base station, a Raspberry Pi 4 for pump control, and a 12V 7.5Ah battery for energy storage. Multiple data-collection substations were established to gather field data. The ESP8266 microcontroller was integrated with a Capacitive Soil Moisture Sensor (V1.2) and a DHT 22 sensor to relay soil moisture, air temperature, and humidity data to the base station via the Message Queuing Telemetry Transport (MQTT) protocol. The battery can power the motor for at least two hours at night, considering a maximum discharge of 75 %, enough to operate the system at the data collection substation. The threshold for pump activation was set at soil moisture below 45 %, with deactivation occurring at or above 80 % to maintain optimal moisture levels. A website was created utilizing the Python Django framework, and an SQLite3 database was implemented, enabling real-time monitoring and remote control of the irrigation pump. Multiple criteria for irrigation were established to enhance the pump's performance so that the developed irrigation system could operate efficiently. This method enables farmers to remotely monitor field conditions and manage irrigation via a website, thereby decreasing reliance on traditional energy sources and reducing water loss during irrigation.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100773"},"PeriodicalIF":6.3,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181132","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}

{"title":"Advancing soybean biomass estimation through multi-source UAV data fusion and machine learning algorithms","authors":"Haitao Da ,&nbsp;Yaxin Li ,&nbsp;Le Xu ,&nbsp;Shuai Wang ,&nbsp;Limin Hu ,&nbsp;Zhengbang Hu ,&nbsp;Qiaorong Wei ,&nbsp;Rongsheng Zhu ,&nbsp;Qingshan Chen ,&nbsp;Dawei Xin ,&nbsp;Zhenqing Zhao","doi":"10.1016/j.atech.2025.100778","DOIUrl":"10.1016/j.atech.2025.100778","url":null,"abstract":"<div><div>Technological advances in unmanned aerial vehicle (UAV) systems offer significant potential for the rapid and efficient monitoring of soybean aboveground biomass (AGB) in precision agriculture, providing an alternative to traditional AGB measurement techniques. However, recent studies have indicated that relying solely on vegetation indices (VIs) can lead to inaccurate AGB estimations due to variability in crop cultivars, growth stages, and environmental conditions. This study evaluated the performance of UAV-derived features (including canopy spectral, textural, and structural features) in estimating AGB across fifty soybean cultivars and multiple growth stages in a two-year field experiment, utilizing various machine learning algorithms (decision tree, DT; random forest, RF; neural network, NN; extreme gradient boosting, XGBoost; and ensemble learning, EL). The findings revealed that: (1) The integration of UAV digital imagery with the canopy height model (CHM) facilitated the estimation of soybean plant height, with the coefficient of determination (R²) and root mean square error (RMSE) values for ground-measured and UAV-derived plant height across different growth stages ranging from 0.72 to 0.88 and 3.35 to 6.13 cm, respectively. (2) Textural and structural features demonstrated good sensitivity to AGB variability across cultivars and growth stages, despite each feature type having its limitations. The fusion of UAV-derived spectral, textural, and structural features yielded the highest accuracy (R² = 0.85), significantly improving model performance compared to using dual (R² ranging from 0.79 to 0.81) feature types. (3) Model accuracy significantly varied across different growth stages. For machine learning algorithms, the EL model outperformed DT, RF, NN, and XGBoost in AGB prediction, consistently providing accurate estimations multiple soybean growth stages. These findings highlight the potential of integrating multi-source UAV features to enhance soybean AGB estimation, facilitating farmers decision-making in precision crop management and assisting breeders to select high- and sustainable-yielding cultivars in large-scale breeding program.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100778"},"PeriodicalIF":6.3,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181512","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}

{"title":"Using internet technology for business entrepreneurial choice: Evidence from Chinese farming households","authors":"Baoling Zou ,&nbsp;Feiyun Yang ,&nbsp;Ashok K. Mishra","doi":"10.1016/j.atech.2025.100775","DOIUrl":"10.1016/j.atech.2025.100775","url":null,"abstract":"<div><div>Entrepreneurship has the potential to stimulate employment, diversify and increase income, and foster shared prosperity. Rural farm families have harnessed the advantages of digital technologies like the internet by using them for input acquisition, new technologies, production methods, advertising and marketing, income diversification (farm and off-farm work), and entrepreneurial choices. Farm families work off-farm by engaging in self-employment and creating their off-farm businesses (entrepreneurship). This study analyzes farmers’ Internet usage in terms of entrepreneurial decisions using the 2018 China Family Panel Studies survey. The results show that Internet usage significantly promotes farmers’ entrepreneurial choice, and the estimated effects are robust. The mechanism analysis reveals that Internet usage mainly affects farmers’ entrepreneurial choices through financial capital, social networks, and risk preference attitudes. Finally, heterogeneity analysis indicates that Internet usage encourages entrepreneurial choices among farmers with higher education or family income.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"10 ","pages":"Article 100775"},"PeriodicalIF":6.3,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183259","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}