Smart agricultural technology最新文献

{"title":"Remote sensing for pasture biomass quantity and quality assessment: Challenges and future prospects","authors":"Nicola Furnitto ,&nbsp;Juan Miguel Ramírez-Cuesta ,&nbsp;Diego S. Intrigliolo ,&nbsp;Giuseppe Todde ,&nbsp;Sabina Failla","doi":"10.1016/j.atech.2025.101057","DOIUrl":"10.1016/j.atech.2025.101057","url":null,"abstract":"<div><div>Optimizing pasture use through careful management is critical to ensure the economic and environmental sustainability of pasture-based agriculture. Maximizing grass utilization and accurately measuring grass quantity and quality by adopting precision agriculture technologies, including estimates from satellite or unmanned aerial vehicles (UAVs), are key aspects to improve production efficiency and reducing environmental impact. With these goals, the review explores the crucial role of biomass quality and assessment estimation in pasture-based agricultural practices, with a focus on the potential offered by remote sensing technologies. This review examined recent advances in biomass and grassland quality assessment, highlighting the most widely used methodologies, remaining challenges and future prospects. The analysis focused particularly on applications of UAV and satellite platforms, discussing the advantages and limitations of the different techniques. Their and applications including machine learning (ML) technologies. A deep analysis of the main indices, electromagnetic regions and ML approaches utilized was also addressed, distinguishing among those intended to biomass quantity and quality assessment. Through the integration of innovative technologies and improved measurement protocols, the full potential of more sustainable and productive pasture-based agriculture can be realised, ensuring improved animal productivity and economic viability for farmers. These advances will pave the way for more effective management practices and contribute significantly to the global effort toward more sustainable agricultural systems.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"12 ","pages":"Article 101057"},"PeriodicalIF":6.3,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144242534","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}

{"title":"Smart and accurate: A new tool to identify stressed soybean seeds based on multispectral images and machine learning models","authors":"Ana Carolina Picinini Petronilio ,&nbsp;Clíssia Barboza Mastrangelo ,&nbsp;Thiago Barbosa Batista ,&nbsp;Gustavo Roberto Fonseca de Oliveira ,&nbsp;Isabela Lopes dos Santos ,&nbsp;Edvaldo Aparecido Amaral da Silva","doi":"10.1016/j.atech.2025.101042","DOIUrl":"10.1016/j.atech.2025.101042","url":null,"abstract":"<div><div>Extreme environmental conditions have been recurrent during the last few years and have impacted crop seed quality worldwide, mainly but not limited to, soybeans (<em>Glycine</em> max (L) Merrill). To overcome this, seed companies often demand innovative tools to address seed quality factors. Machine learning models based on multispectral imaging are a novel seed quality analysis approach. Thus, we hypothesize that segmenting stressed (those produced under conditions that are not favorable to the mother-plant) and non-stressed (produced under conditions favorable to the mother-plant) soybean seeds would be possible with this technology, opening a new opportunity for seed quality management and elucidating quality factors. Soybean seeds (cultivar BR/MG 46-Conquista) were produced under water deficit and heat during maturation (from R5.5 onwards). Multispectral images were acquired from stressed and non-stressed seeds, and the reflectance, autofluorescence, physical properties, and chlorophyll parameters were extracted from the images. In parallel, we determined seed vigor. We designed machine learning models using multispectral imaging data based on three algorithms: neural network, support vector machine, and random forest. Our results demonstrated that the stressed seeds have spectral markers that enable their recognition. Concomitantly, these markers had a direct relationship with seed vigor. The machine learning models developed based on neural network algorithm showed the highest performance in segmenting stressed seeds (≥90 % of accuracy, precision, recall, specificity and F1 score) in contrast to random forest- and support vector machine algorithm (≥88 % of accuracy, precision, recall, specificity and F1 score). Here, we report a new approach for multispectral imaging with the potential to identify soybean seeds of lower vigor as a result of unfavorable environmental conditions during seed maturation.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"12 ","pages":"Article 101042"},"PeriodicalIF":6.3,"publicationDate":"2025-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144291142","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}

{"title":"Path planning for spot spraying with UAVs combining TSP and area coverages","authors":"Mogens Plessen","doi":"10.1016/j.atech.2025.100965","DOIUrl":"10.1016/j.atech.2025.100965","url":null,"abstract":"<div><div>This paper addresses the following task: given a set of patches or areas of varying sizes that are to be serviced within a bounding closed contour, calculate a minimal length path plan for an unmanned aerial vehicle (UAV) such that all patches are serviced, the path additionally avoids any obstacles areas within the bounding contour and the path never leaves the bounding contour. The application in mind is agricultural spot spraying, where the bounding contour represents the field contour and multiple patches represent multiple weed areas meant to be sprayed. Obstacle areas are ponds or tree islands. The proposed method combines a heuristic solution to a traveling salesman problem (TSP) with optimised area coverage path planning. Two TSP-initialisation and 4 TSP-refinement heuristics as well as two area coverage path planning methods are evaluated on three real-world experiments with three obstacle areas and 15, 19 and 197 patches, respectively. The unsuitability of a Boustrophedon-path for area coverage gap avoidance is discussed and inclusion of a headland path for area coverage is motivated. Two main findings are (i) the particular suitability of one TSP-refinement heuristic, and (ii) the unexpected high contribution of patches areas coverage pathlengths on total pathlength, highlighting the importance of optimised area coverage path planning for spot spraying.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100965"},"PeriodicalIF":6.3,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899881","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}

{"title":"Estimation of microbial load in Ganoderma lucidum using a solar-electric hybrid dryer enhanced by machine learning and IoT","authors":"Pinit Nuangpirom ,&nbsp;Siwasit Pitjamit ,&nbsp;Weerin Pheerathamrongrat ,&nbsp;Wasawat Nakkiew ,&nbsp;Parida Jewpanya","doi":"10.1016/j.atech.2025.100977","DOIUrl":"10.1016/j.atech.2025.100977","url":null,"abstract":"<div><div>This study focuses on developing a hybrid-powered dryer that uses both solar and electric energy to dry Ganoderma lucidum mushrooms. Integrated with an Internet of Things (IoT) platform, the system enables real-time monitoring of temperature, time, and humidity. The analysis evaluated reductions in weight, moisture content, water activity, and microbial counts (bacteria, fungus, and yeast) across temperatures ranging from 40 °C to 80 °C over 480 min. The results indicated that higher temperatures, particularly 80 °C, were most effective in reducing microbial counts, achieving near-zero levels after 240 to 480 min. Machine learning (ML) models random forest regression (RFR), decision tree regression (DTR), and multiple linear regression (MLR) were trained to estimate microbial levels based on input variables such as time, temperature, and weight. RFR had the highest accuracy for estimating bacteria, while DTR excelled for fungus and yeast. However, MLR proved most suitable for IoT applications due to its simplicity in real-time implementation on devices. Therefore, the ML models were selected based on accuracy performance (RFR and DTR) and ease of integration into IoT systems (MLR). This study demonstrates the hybrid dryer's efficiency and the potential of ML models to optimize the drying process, contributing to energy efficiency and product quality control. Initially designed for small-scale on-farm use, the system also has the potential for future scaling to industrial processing facilities.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100977"},"PeriodicalIF":6.3,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143891419","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}

{"title":"Digital transformation in Moroccan agriculture: Applications, used technologies, impacts on marketing, limitations, and orientations for future research","authors":"Mohammed Fakhraddine ,&nbsp;Najib Zerrad ,&nbsp;Hicham Berhili ,&nbsp;Meryeme Morchid","doi":"10.1016/j.atech.2025.100978","DOIUrl":"10.1016/j.atech.2025.100978","url":null,"abstract":"<div><div>Integration of digital knowledge in the agricultural field presents a prospect to enhance the ascendancy of agricultural strategy and stimulate economic progress in Morocco through the dissemination of information, decision-making tools, and transmission methodologies. This study aimed to review current data on digital transformation in the agriculture field of Morocco. We presented a literature review on digital technologies and agriculture in Morocco from 1990 to 2025. We used related keywords, and data was recorded from databases, governmental sites, and published documents. The results obtained showed that in Morocco, digital transformation was adopted in the last decade of the 21st century, with applications in different fields, including education, finance, health, administration, etc. Various digital technologies have been implemented in agriculture, including Big Data Analytics (BDA), Blockchain, Unmanned Aerial Vehicles (UAVs), Deep Learning (DL), Unmanned Ground Vehicles (UGVs), Information and Communications Technologies (ICT), Machine Learning (ML), Cloud Computing (CC), Artificial Intelligence (AI), robotics, and the Internet of Things (IoT). These technologies have addressed different fields, including production, farming, logistics, marketing, export, and improved services, visibility, tracking of products, and commercialization. In Morocco, any endeavours by relevant authorities or stakeholders to promote and develop automated agriculture through the integration of new technologies have much ground to make up compared to the top spots in the sector. As a result, we suggested areas that need more improvement in terms of research,  farmer awareness, and activities. This study summarizes for the first time data concerning digital transformation in agriculture and suggests it as a reference for future research and to farmers and cooperatives in Morocco.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100978"},"PeriodicalIF":6.3,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899760","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}

{"title":"Obstacle detection and avoidance system based on layered costmaps for robot tractors","authors":"Ricardo Ospina ,&nbsp;Kota Itakura","doi":"10.1016/j.atech.2025.100973","DOIUrl":"10.1016/j.atech.2025.100973","url":null,"abstract":"<div><div>In the context of automated navigation for agricultural vehicles, efficient obstacle avoidance remains a significant challenge, particularly on farm roads where road conditions vary. This paper presents a novel obstacle detection and avoidance system based on layered costmaps, designed to enhance the safety and efficiency of robot tractors navigating farm roads. The system integrates a cost-effective 2D LiDAR sensor for obstacle detection, combined with real-time avoidance maneuver calculation to ensure continuous and safe operation. A static layer map was created using a simple image processing technique, so it can be easily integrated with the layered costmaps. The system’s performance was validated through three experimental setups. For single obstacle avoidance, the system achieved an RMSE of 0.15 m in lateral avoidance distance. For two parallel obstacles, the RMSE was 0.19 m, and for two consecutively aligned obstacles, the RMSE was below 0.28 m. These results demonstrate the effectiveness of the proposed system in ensuring stable obstacle detection and avoidance, highlighting its potential for practical use in agricultural machinery for field operations. The method provides a cost-efficient solution, bypassing the need for complex sensor fusion and synchronization, making it highly suitable for real-world deployment.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100973"},"PeriodicalIF":6.3,"publicationDate":"2025-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143887242","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}

{"title":"Enhancing lettuce classification: Optimizing spectral wavelength selection via CCARS and PLS-DA","authors":"Nicola Dilillo ,&nbsp;Andrea Sanna ,&nbsp;Elena Belcore ,&nbsp;Kyra Smith ,&nbsp;Marco Piras ,&nbsp;Bartolomeo Montrucchio ,&nbsp;Renato Ferrero","doi":"10.1016/j.atech.2025.100962","DOIUrl":"10.1016/j.atech.2025.100962","url":null,"abstract":"<div><div>Spectroscopy is a valuable tool for analyzing the inside of plants. In this field, plant health is evaluated through light analysis, specifically by examining wavelengths beyond the visible spectrum, making it essential to select the most representative wavelength. The Competitive Adaptive Reweighted Sampling (CARS) algorithm has been applied efficiently in the literature to select the best variables in several applications, including agricultural monitoring, nutrient analysis, and chemometrics. This study presents the Calibrated CARS (CCARS) algorithm, an extension of CARS, alongside the Partial Least Square Discriminant Analysis (PLS-DA) model. The algorithm is developed to identify critical informative wavelengths of a spectral dataset of lettuce to facilitate the creation of streamlined and efficient models for lettuce health classification. While effective with spectral data, the PLS-DA models tend to overfit, and to address this problem a rigorous systematic evaluation approach is employed. Permutation tests are conducted to verify the model's robustness, while learning curve analyses ensure the model's capacity to generalize data. With this comprehensive evaluation method, confidence in the robustness of the PLS-DA models is instilled, ensuring model stability, which is achieved thanks to the CCARS algorithm instead of the original version. The results demonstrate that using CCARS with 3 or 4 PLS components and only 30 or 19 selected wavelengths reduces the number of variables by 97%, without sacrificing accuracy, and with a statistically significant robust model.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100962"},"PeriodicalIF":6.3,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879022","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}

{"title":"Predicting the greenhouse crop morphological parameters based on RGB-D Computer Vision","authors":"Ziqiu Kang ,&nbsp;Bo Zhou ,&nbsp;Shulang Fei ,&nbsp;Nan Wang","doi":"10.1016/j.atech.2025.100968","DOIUrl":"10.1016/j.atech.2025.100968","url":null,"abstract":"<div><div>Accurate data acquisition of crop morphological parameters is crucial for effective greenhouse management decision-making and remote sensing technologies are increasingly being applied to automate the data collection process. This research utilised an RGB-D based computer vision method to investigate the correlation between the computer vision features and the lettuce morphological parameters, including leaf area, plant height, diameter, and fresh weight. A dataset of lettuce containing over 300 RGB images and depth images of the 3rd Autonomous Greenhouse Challenge was used, and Random Forest, XGBoost and linear regression models were applied in the prediction. The best NRMSE values for diameter, dry matter content, dry weight, fresh weight, height, and leaf area are 0.08, 0.08, 0.07, 0.07, 0.08, and 0.07, which showed a promising accuracy compared to similar studies. This research demonstrates a novel approach to non-destructively estimate greenhouse leafy vegetable morphological parameters.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100968"},"PeriodicalIF":6.3,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143881716","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}

{"title":"Metabolic mapping for precision grape maturation: Application of a tomography-like method for site-specific management","authors":"Renan Tosin ,&nbsp;Leandro Rodrigues ,&nbsp;Maria Santos-Campos ,&nbsp;Igor Gonçalves ,&nbsp;Catarina Barbosa ,&nbsp;Filipe Santos ,&nbsp;Rui Martins ,&nbsp;Mario Cunha","doi":"10.1016/j.atech.2025.100967","DOIUrl":"10.1016/j.atech.2025.100967","url":null,"abstract":"<div><div>This study demonstrates the application of a tomography-like (TL) method to monitor grape maturation dynamics over two growing seasons (2021–2022) in the Douro Wine Region. Using a Vis-NIR point-of-measurement sensor, which employs visible and near-infrared light to penetrate grape tissues non-destructively and provide spectral data to predict internal composition, this approach captures non-destructive measurements of key physicochemical properties, including soluble solids content (SSC), weight-to-volume ratio, chlorophyll and anthocyanin levels across internal grape tissues - skin, pulp, and seeds - over six post-veraison stages. The collected data were used to generate detailed metabolic maps of maturation, integrating topographical factors such as altitude and NDVI-based (normalised difference vegetation index) vigour assessments, which revealed significant (<em>p</em> &lt; 0.05) variations in SSC, chlorophyll, and anthocyanin levels across vineyard zones. The metabolic maps generated from the TL method enable high-throughput data to reveal the impact of environmental variability on grape maturation across distinct vineyard areas. Predictive models using random forest (RF) and self-learning artificial intelligence (SL-AI) algorithms showed RF’s robustness, achieving stable predictions with R² ≥ 0.86 and MAPE ≤ 33.83 %. To illustrate the TL method’s practical value, three hypothetical decision models were developed for targeted winemaking objectives based on SSC, chlorophyll in the pulp, and anthocyanin in the skin and seeds. These models underscore the TL method’s ability to support site-specific management (SSM) by providing actionable agricultural practices (e.g. harvest) into vineyard management, guiding winemakers to implement tailored interventions based on metabolic profiles rather than only cultivar characteristics. This precision viticulture (PV) approach enhances wine quality and production efficiency by aligning vineyard practices with specific wine quality goals.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100967"},"PeriodicalIF":6.3,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873695","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}

{"title":"Weed instance segmentation from UAV Orthomosaic Images based on Deep Learning","authors":"Chenghao Lu ,&nbsp;Klaus Gehring ,&nbsp;Stefan Kopfinger ,&nbsp;Heinz Bernhardt ,&nbsp;Michael Beck ,&nbsp;Simon Walther ,&nbsp;Thomas Ebertseder ,&nbsp;Mirjana Minceva ,&nbsp;Yuncai Hu ,&nbsp;Kang Yu","doi":"10.1016/j.atech.2025.100966","DOIUrl":"10.1016/j.atech.2025.100966","url":null,"abstract":"<div><div>Weeds significantly impact agricultural production, and traditional weed control methods often harm soil health and environment. This study aimed to develop deep learning-based segmentation models in identifying weeds in potato fields captured by Unmanned Aerial Vehicle (UAV<em>)</em> orthophotos and to explore the effects of weeds on potato yield. Previous studies predominantly employed U-Net for weed segmentation, but its performance often declines under complex field environments and low-image resolution conditions. Some studies attempted to overcome this limitation by reducing flight altitude or using high-cost cameras, but these approaches are not always practical. To address these challenges, this study uniquely integrated Real-ESRGAN Super-Resolution (SR) for UAV image enhancement and the Segment Anything Model (SAM) for semi-automatic annotation. Subsequently, we trained the YOLOv8 and Mask R-CNN models for segmentation. Results showed that the detection accuracy mAP50 scores were 0.902 and 0.920 for YOLOv8 and Mask R-CNN, respectively. Real-ESRGAN reconstruction slightly improved accuracy. When multiple weed types were present, accuracy generally decreased. The YOLOv8 model characterized plant and weed coverage areas could explained 41.2 % of potato yield variations (R² = 0.412, p-value = 0.01), underscoring the practical utility of UAV-based segmentation for yield estimation. Both YOLOv8 and Mask R-CNN achieved high accuracy, with YOLOv8 converging faster. While different nitrogen fertilizer treatments had no significant effect on yield, weed control treatments significantly impacted yield, highlighting the importance of precise weed mapping for spot-specific weed management. This study provides insights into weed segmentation using Deep Leaning and contributes to environmentally friendly precision weed control.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100966"},"PeriodicalIF":6.3,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143868969","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}