Smart agricultural technology最新文献

{"title":"Exploring bare soil digital mapping: identifying alternative variables to replace ECa via remote sensing, a case study on two Italian fields at different latitude","authors":"Matteo Petito ,&nbsp;Emanuele Barca ,&nbsp;Antonio Berti ,&nbsp;Silvia Cantalamessa ,&nbsp;Giancarlo Pagnani ,&nbsp;Michele Pisante","doi":"10.1016/j.atech.2025.100955","DOIUrl":"10.1016/j.atech.2025.100955","url":null,"abstract":"<div><div>Site-specific management in agriculture, which accounts for variability within a field, is a cornerstone of sustainable agronomic practices. However, despite the availability of numerous methods to measure spatial variability, their limitations hinder large-scale adoption, posing challenges to the broader implementation of precision agriculture. This study aims to identify spectral indices derived from bare-soil analysis as potential substitutes for electrical conductivity (ECa) in mapping spatial variability. The approach aligns with the need for cost-effective, scalable, and less labor-intensive solutions to manage field variability. Using multi-temporal bare-soil imagery spanning five years across two fields under intermittent cultivation in Italy, we applied principal component analysis to evaluate correlations between spectral indices and ECa. Both fields demonstrated strong correlations between ECa and the first principal component (PC1). Key variables identified as highly correlated with ECa included the Brightness Index (0.66), Near-Infrared (0.53), and Red reflectance (0.58). The percentage variance explained by PC1 was 75.4 % for Field 1 and 79.0 % for Field 2. Finally, PC1 is correlated with ECa in the two areas in the measure of 0.73 and 0.53, respectively. This work addresses the problem of substituting ECa with covariates derived from bare-soil analysis from a purely statistical perspective as a first necessary step aiming at identifying the most promising covariates. A subsequent study will address this issue from a pedological standpoint. These findings highlight the potential of remote sensing data and spectral indices from multi-temporal imagery to replace direct ECa measurements, enabling rapid and accurate mapping of spatial variability in annual croplands.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100955"},"PeriodicalIF":6.3,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143843321","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":"Deep learning-based detection of green-ripe pineapples via bract wilting rate analysis","authors":"Guo-Fong Hong ,&nbsp;Sumesh Nair ,&nbsp;Chun-Yu Lin ,&nbsp;Ching-Shan Kuan ,&nbsp;Shean-Jen Chen","doi":"10.1016/j.atech.2025.100949","DOIUrl":"10.1016/j.atech.2025.100949","url":null,"abstract":"<div><div>Green-ripe pineapples are ideal for long-term transportation and storage during summer. However, accurately identifying them during <em>in-situ</em> harvesting remains a challenge for farmers. To address this issue, this study proposes a deep learning-based YOLO<img>NAS-L algorithm to detect green-ripe pineapples by analyzing the wilting rate of floral bracts at the fruit's base. An unmanned tracked vehicle equipped with an Intel D405 depth camera was used to traverse pineapple fields, capturing images from a distance of 300–400 mm. Each image covered approximately 20 floral bracts, with a detection resolution of around 30 × 30 pixels. The camera also provided three-dimensional coordinates of the pineapples to support automated harvesting. To mitigate ambient light variations, a white LED lighting system (24V/5A) was implemented for illumination enhancement. Experimental results indicate that analyzing floral bract wilting improves green-ripe pineapple recognition accuracy by 13.6 %, reaching 95.4 %, compared to solely identifying the pineapple's base. These findings demonstrate that deep learning-based floral bract wilting analysis significantly enhances recognition accuracy and provides robust support for automated harvesting.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100949"},"PeriodicalIF":6.3,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143828410","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":"Early and accurate nutrient deficiency detection in hydroponic crops using ensemble machine learning and hyperspectral imaging","authors":"Nagarajan S․ ,&nbsp;Maria Merin Antony ,&nbsp;Murukeshan Vadakke Matham","doi":"10.1016/j.atech.2025.100952","DOIUrl":"10.1016/j.atech.2025.100952","url":null,"abstract":"<div><div>Vertical indoor hydroponic farms are growing as a technological solution fostering agriculture productivity to address the ever increasing food demands in sustainable cities. These farms provide extensive control over the growing conditions to ensure all-weather cultivation of diverse crops within the available limited space. However, to assure the quality of hydroponic crops, continuous close-range crop monitoring and early detection of deficiencies are essential. Sensitive techniques such as hyperspectral imaging combined with ensemble based machine learning techniques have proven to provide improved reliable results. However, despite their potential, the application of these methods for early-stage nutrient deficiency detection in crops remains relatively underexplored. In this context, this research presents and proposes different machine learning-based approaches that utilizes ensemble techniques such as Random Forest (RF), Bagging or Bootstrap Aggregating, Adaboost or Adaptive Boosting, and eXtreme Gradient Boosting (XGB) classifiers for early detection of nutrient deficiencies in hydroponic crops. In the proposed approach, the features extracted from hyperspectral datacubes are trained to create machine learning models. Among the investigated models, the XGB classifier demonstrated the fastest computational time and test accuracy of 18.07 s and 99.6 %, respectively. This research also proposes a novel computer vision (CV) based approach to improve the tedious manual data labelling process involved in HSI dataset creation. Envisioned as an invaluable tool, the proposed non-invasive imaging system could detect as early as 3 days after stress induction and revolutionize the automated monitoring of indoor hydroponic farms with enhanced accuracy for a sustainable future.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100952"},"PeriodicalIF":6.3,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143839454","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":"Effects of viewing angle and field of view on detection, tracking, and counting of pine seedlings towards automated forest nursery inventory","authors":"Ashish Reddy Mulaka ,&nbsp;Rafael Bidese ,&nbsp;Yin Bao","doi":"10.1016/j.atech.2025.100951","DOIUrl":"10.1016/j.atech.2025.100951","url":null,"abstract":"<div><div>The current inventory practice in bareroot forest nurseries relies on manually counting tree seedlings in randomly sampled plots to estimate the stock for each seed lot. This method is labor-intensive, time-consuming, and susceptible to human error. Recent advances in deep learning-based object detection and efficient tracking algorithms have enabled automated object counting in video data across various domains, including crop seedling counting in agriculture. This study investigates the effects of viewing angle (VA) and field of view (FoV) on detection, tracking, and counting early-stage pine seedlings in nadir-view videos using a detect-and-track approach. We evaluated the performance of YOLOv8–10 models in conjunction with three multi-object tracking (MOT) algorithms (SORT, ByteTrack, and BoT-SORT) on a custom MOT dataset comprising an average of 153 seedlings per frame and totaling 166,440 seedlings. Detection results and statistical tests showed that increasing horizontal VA reduces the intersection over union (IoU) of seedling detections, primarily due to the perspective differences introduced by oblique viewing angles. MOT evaluations further demonstrated that BoT-SORT consistently delivered high counting accuracy when the vertical FoV encompassed at least the entire seedling. In contrast, ByteTrack and SORT exhibited significantly lower performance, producing reasonable counting accuracy only when the vertical FoV was sufficiently large. The superior performance of BoT-SORT is attributed to its camera motion compensation, which effectively reduces identity switches and tracking failures in scenes involving stationary yet overlapping seedlings. Notably, BoT-SORT achieved 100 % counting accuracy under a 20° horizontal VA across YOLO model sizes. Furthermore, larger YOLO models showed greater robustness to increases in horizontal VA. These findings provide valuable guidance for optimizing camera configurations and model selection towards the development of a real-time inventory systems for precision forest nursery management.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100951"},"PeriodicalIF":6.3,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143839453","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":"Deoxynivalenol prediction and spatial mapping in wheat based on online hyperspectral imagery scanning","authors":"Muhammad Baraa Almoujahed ,&nbsp;Orly Enrique Apolo-Apolo ,&nbsp;Mohammad Alhussein ,&nbsp;Marius Kazlauskas ,&nbsp;Zita Kriaučiūnienė ,&nbsp;Egidijus Šarauskis ,&nbsp;Abdul Mounem Mouazen","doi":"10.1016/j.atech.2025.100947","DOIUrl":"10.1016/j.atech.2025.100947","url":null,"abstract":"<div><div>Deoxynivalenol (DON), a harmful mycotoxin produced by several <em>Fusarium</em> species, poses critical challenges to wheat production and food safety. However, reducing risks of human toxicity requires pre-harvest detection of DON concentration across different zones of a field. This study investigates the potential of integrating hyperspectral imaging (HSI) in the 400–1000 nm range with machine learning (ML) models for online field detection and mapping of DON contamination in wheat (<em>Triticum aestivum)</em>. Using a tractor-mounted push-broom hyperspectral camera, spectral data were collected across four commercial fields in Lithuania and Belgium. A total of 76 wheat samples collected during crop scanning were analyzed for DON levels using liquid chromatography-mass spectrometry (LC-MS). Initial analysis of spectral data alone revealed relatively low classification accuracy, with light gradient boosting machine (LGBM) achieving 55.92 % and decision tree classifier (DTC) achieving 51.97 %. However, the inclusion of fusarium head blight (FHB) severity as an additional feature significantly improved performance, boosting accuracy to 90.79 % for LGBM (a 62.4 % increase) and 86.18 % for DTC (a 65.8 % increase). Moreover, the use of mutual information (MI) for feature selection enhanced model accuracy, achieving 93.42 % for LGBM and 90.13 % for DTC. Spatial mapping of DON contamination demonstrated fair to substantial agreement with ground truth maps, providing valuable tools for farmers to understand DON distribution and implement targeted harvesting strategy. This study highlights the potential of integrating online HSI, ML, and feature selection techniques, for pre-harvest DON detection and mapping, providing valuable information for reducing risks of human toxicity and improving the economic value of wheat grain.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100947"},"PeriodicalIF":6.3,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143828411","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":"Deep learning framework for fruit counting and yield mapping in tart cherry using YOLOv8 and YOLO11","authors":"Anderson L.S. Safre ,&nbsp;Alfonso Torres-Rua ,&nbsp;Brent L. Black ,&nbsp;Sierra Young","doi":"10.1016/j.atech.2025.100948","DOIUrl":"10.1016/j.atech.2025.100948","url":null,"abstract":"<div><div>Object detection for fruit counting has significant potential for orchard yield estimation. Tart cherries are mechanically harvested, creating opportunities for developing new yield mapping technologies. However, there is a lack of dedicated technologies for this purpose, motivating the evaluation of computer vision-based approaches in tart cherries. In this study, we compared the nano (n) and extra-large (x) configurations of YOLOv8 and YOLO11 for tart cherry detection and fruit counting on the harvester. The models demonstrated robust performance, even in high object density conditions, with YOLOv11x achieving a mAP50 of 0.92. While YOLOv8n and YOLO11n produced similar detection results, YOLOv8n had a faster inference time, making it more suitable for real-time applications. Fruit counting was performed using a combination of YOLO models and the BoT-SORT tracking algorithm. The resulting number of fruits was compared to the actual weights of harvested fruit from individual trees. The results indicated a linear relationship, with YOLO11x achieving an R² of 0.62 and an RMSE of 10 kg. To the best of our knowledge, this is the first study to evaluate object detection and fruit counting performance in tart cherries during harvest. Additionally, we introduce a new dataset with annotated cherries on the conveyor belt of the harvester which can support further research and development. This approach addresses the existing technology gap in yield monitoring for tart cherry orchards, facilitating the application of precision agriculture and site-specific management strategies in the industry.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100948"},"PeriodicalIF":6.3,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143834385","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":"Analysis and experiment on seed-filling performance of double seed-taking precision dibbler for cotton","authors":"Zibin Mao ,&nbsp;Luochuan Xu ,&nbsp;Mengyu Guo ,&nbsp;Junwei Li ,&nbsp;Xin Luo ,&nbsp;Bin Hu ,&nbsp;Xiyang Li","doi":"10.1016/j.atech.2025.100946","DOIUrl":"10.1016/j.atech.2025.100946","url":null,"abstract":"<div><div>Aiming at the problem that cottonseeds cannot be separated and migrated from the population in time due to disordered seed-filling, short filling time, and fast seed-filling speed in the filling process of the existing mechanical precision dibbler, this paper innovated a mechanical precision dibbler for double seed-taking, in which the seed-taking disc is provided with a finite migration space that can effectively disrupt the population and assist seed-filling. It can ensure the seed-filling performance of the cotton precision dibbler at high speed (&gt; 4 km·h^-1). A numerical simulation of the seed-filling process was conducted using EDEM software, and the effects of kinetic energy and velocity of the population, moving track, and velocity of a single cottonseed on filling performance were examined. A three-factor, five-level quadratic rotational orthogonal test with rotational speed, population height, and thickness as test factors, filling index, and missing index as test indices. When the rotational speed of the seed-taking disc was 42.3 r·min^-1, that is, the working speed of the planter was 4.06 km·h^-1, the population height was 0.165 kg, and the thickness of the seed-taking disc was 5.5 mm, the filling index and missing index were 96.65% and 3.35%, respectively. This study not only provides a reference for the high-speed seed-filling theory of the type hole in precision dibblers but also contributes to the local seed cluster formation of ellipsoidal materials in the relative rotation space and the accelerated migration of a single target material to fill type holes.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100946"},"PeriodicalIF":6.3,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143828409","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":"Design of tabletop hemispherical light transmittance characterization system for small scale samples","authors":"Chun-Ting Cho,&nbsp;Johan de Haas,&nbsp;Erik van der Kolk","doi":"10.1016/j.atech.2025.100936","DOIUrl":"10.1016/j.atech.2025.100936","url":null,"abstract":"<div><div>Greenhouse owners highly value the hemispherical light transmittance (T_HEM) of roofing materials because sunlight rarely projects at a perpendicular angle, especially in high-latitude regions. With growing interest in research of advanced multi-functional greenhouse roofing, a compact and efficient T_HEM characterization system for lab-scale samples is needed to promote research in the horticulture field. In this study, we developed a tabletop system capable of characterizing T_HEM of lab-scale samples with a size one-third of that required by the current characterization system. Key designing parameters, such as the beam cross-section area, port area, and port edge thickness were systematically varied to evaluate their impact on T_HEM characterization. The results indicated that the total port area should be limited to under 1% of the sphere surface area with minimized edge thickness since reflection from the edge area can not be corrected by a double-beam measurement. Furthermore, the collimated beam cross-section area should exceed the port area by a factor of 1.5 to ensure that T_HEM remains unaffected by the sphere rotation center. The system provides a consistent and reliable method for T_HEM measurement and offers essential guidelines for future users to construct a similar setup.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100936"},"PeriodicalIF":6.3,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143817320","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":"Transformer-based similarity learning for re-identification of chickens","authors":"Christian Lamping,&nbsp;Gert Kootstra,&nbsp;Marjolein Derks","doi":"10.1016/j.atech.2025.100945","DOIUrl":"10.1016/j.atech.2025.100945","url":null,"abstract":"<div><div>Continuous animal monitoring relies heavily on the ability to re-identify individuals over time, essential for both short-term tracking, such as video analysis, and long-term monitoring of animal conditions. Traditionally, livestock re-identification is approached using tags or sensors, which require additional handling effort and potentially impact animal welfare. In response to these limitations, non-invasive vision-based approaches have emerged recently, with existing research primarily focusing on the re-identification of pigs and cows. Re-identification of chickens, which exhibit high uniformity and are housed in larger groups, remains challenging and has received less research attention. This study addresses this gap by exploring the feasibility of re-identifying individual laying hens within uncontrolled farm environments using images of their heads. It proposes the first similarity-learning approach based on a VisionTransformer architecture to re-identify chickens without requiring training images for each individual bird. In our experiments, we compared the transformer-based approach to traditional CNN architectures while assessing the impact of different model sizes and triplet mining strategies during training. Moreover, we evaluated practical applicability by analyzing the effects of the number of images per chicken and overall population size on re-identification accuracy. Finally, we examined which visual features of the chicken head were most relevant for re-identification. Results show Top-1 accuracies exceeding 80 % for small groups and maintaining over 40 % accuracy for a population of 100 chickens. Moreover, it was shown that the transformer-based architecture outperformed CNN models, with the use of semi-hard negative samples during training yielding the best results. Furthermore, it was revealed that the evaluated models learned to prioritize features such as the comb, wattles, and ear lobes, often aligning with human perception. These results demonstrate promising potential for re-identifying chickens even when recorded in an uncontrolled farm environment, providing a foundation for future applications in animal tracking and monitoring.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100945"},"PeriodicalIF":6.3,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143839466","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":"Investigating Pea (Pisum sativum L.) Flowering with High Throughput Field Phenotyping and Object Detection","authors":"Corina Oppliger ,&nbsp;Radek Zenkl ,&nbsp;Achim Walter ,&nbsp;Beat Keller","doi":"10.1016/j.atech.2025.100942","DOIUrl":"10.1016/j.atech.2025.100942","url":null,"abstract":"<div><div>Flowering is one of the most important and sensitive processes throughout a plant's life and marks the start of the reproductive phase. Flowering traits largely define yield potential and are therefore crucial for crop breeding. To observe flowering dynamics under field conditions, visual ratings have been a standard method for decades. Today, high-throughput field phenotyping (HTFP) methods provide opportunities for objective and efficient data collection. We developed an object detection approach (based on YOLOv8) that allows to collect detailed data about flower and pod density. RGB-images from 12 pea breeding lines were automatically acquired by the field phenotyping platform (FIP) of ETH Zürich in two years. The trained model reached high accuracy for open flower detection, which allowed to monitor flowering dynamics and flower density over time. Maximal flower density (Max.Fl.Dens) was highly correlated (R²= 0.967) to ground truth data taken in the field. Clear differences in timing of flowering and flower density were detected between breeding lines and years. Furthermore, a high correlation was observed between the maximal flower density and yield components. This automated, data-driven method of flower and pod detection proved itself as a reliable tool. Therefore, the results are promising for the use of RGB imaging methods to objectively assess not only flowering dynamics but also flower density and fruiting efficiency. Maximal flower density allows to predict seed amount and therefore has potential as selection trait in breeding programs. Fruiting efficiency could be used to identify stress-tolerant breeding lines.</div></div>","PeriodicalId":74813,"journal":{"name":"Smart agricultural technology","volume":"11 ","pages":"Article 100942"},"PeriodicalIF":6.3,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143843318","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}