Use Of Unmanned Aerial Vehicles Research Articles

Modelling Water Availability in Livestock Ponds by Remote Sensing: Enhancing Management in Iberian Agrosilvopastoral Systems

Extensive livestock farming plays a crucial role in the economy of agrosilvopastoral systems of the southwestern Iberian Peninsula (known as dehesas and montados in Spanish and Portuguese, respectively) as well as providing essential ecosystem services. The existence of livestock in these areas heavily relies on the effective management of natural resources (annual pastures and water stored in ponds built ad hoc). The present work aims to assess the water availability in these ponds by developing equations to estimate the water volume based on the surface area, which can be quantified by means of remote sensing techniques. For this purpose, field surveys were carried out in September 2021, 2022 and 2023 at ponds located in representative farms, using unmanned aerial vehicles (UAVs) equipped with RGB sensors and survey-grade global navigation satellite systems and inertial measurement units (GNSS-IMU). These datasets were used to produce high-resolution 3D models by means of Structure-from-Motion and Multi-View Stereo photogrammetry, facilitating the estimation of the stored water volume within a Geographic Information System (GIS). The Volume–Area–Height relationships were calibrated to allow conversions between these parameters. Regression analyses were performed using the maximum volume and area data to derive mathematical models (power and quadratic functions) that resulted in significant statistical relationships (r2 > 0.90, p < 0.0001). The root mean square error (RMSE) varied from 1.59 to 17.06 m3 and 0.16 to 3.93 m3 for the power and quadratic function, respectively. Both obtained equations (i.e., power and quadratic general functions) were applied to the estimated water storage in similar water bodies using available aerial or satellite imagery for the period from 1984 to 2021.

Drone-Based Visible–Thermal Object Detection with Transformers and Prompt Tuning

The use of unmanned aerial vehicles (UAVs) for visible–thermal object detection has emerged as a powerful technique to improve accuracy and resilience in challenging contexts, including dim lighting and severe weather conditions. However, most existing research relies on Convolutional Neural Network (CNN) frameworks, limiting the application of the Transformer’s attention mechanism to mere fusion modules and neglecting its potential for comprehensive global feature modeling. In response to this limitation, this study introduces an innovative dual-modal object detection framework called Visual Prompt multi-modal Detection (VIP-Det) that harnesses the Transformer architecture as the primary feature extractor and integrates vision prompts for refined feature fusion. Our approach begins with the training of a single-modal baseline model to solidify robust model representations, which is then refined through fine-tuning that incorporates additional modal data and prompts. Tests on the DroneVehicle dataset show that our algorithm achieves remarkable accuracy, outperforming comparable Transformer-based methods. These findings indicate that our proposed methodology marks a significant advancement in the realm of UAV-based object detection, holding significant promise for enhancing autonomous surveillance and monitoring capabilities in varied and challenging environments.

A survey on security of UAV and deep reinforcement learning

Recently, the use of unmanned aerial vehicles (UAV)s for accomplishing various tasks has gained a significant interest from both civilian and military organizations due to their adaptive, autonomous, and flexibility nature in different environments. The characteristics of UAV systems introduce new threats from which cyber attacks may benefit. Adaptive security solutions for UAVs are required to counter the growing threat surface. The security of UAV systems has therefore become one of the fastest growing research topics. Machine learning based security mechanisms have a potential to provide effective countermeasures that complement traditional security mechanisms. The main motivation of this survey is to the lack of a comprehensive literature review about reinforcement learning based security solutions for UAV systems. In this paper, we present a comprehensive review on the security of UAV systems focusing on deep-reinforcement learning-based security solutions. We present a general architecture of an UAV system that includes communication systems to show potential sources of vulnerabilities. Then, the threat surface of UAV systems is explored. We explain attacks on UAV systems according to the threats in a systematic way. In addition, we present countermeasures in the literature for each attack on UAVs. Furthermore, traditional defense mechanisms are explained to highlight requirements for reinforcement based security solutions on UAVs. Next, we present the main reinforcement algorithms. We examine security solutions with reinforcement learning algorithms and their limitations in a holistic approach. We also identify research challenges about reinforcement based security solutions on UAVs. Briefly, this survey provides key guidelines on UAV systems, threats, attacks, reinforcement learning algorithms, the security of UAV systems, and research challenges.

Features of display in account of compensation of losses from written-off stocks during the period of martial state

The paper is devoted to the issues of reflecting in the accounts compensation for losses from written-off stocks during the period of martial law. The relevance of this direction of research is due to the fact that, unfortunately, at enterprises in places where assets are stored, processes of theft, improper storage and deterioration of stocks sometimes occur, as a result of which business entities suffer losses. Therefore, the use of compensation mechanisms by the guilty parties (employees of the enterprise, the aggressor country) and their reflection in accounting lead to a decrease in costs during the write-off of stocks from the enterprise's balance sheet. Every enterprise in Ukraine, during its activity, necessarily records stocks that can be used in production, trade, administration or in other areas of work. Special attention is paid to the places of their storage during the receipt of these assets. We believe that the data of warehouse accounting of stocks should coincide with the data of accounting. But, unfortunately, there are situations in which, after an inventory or military aggression, it is necessary to write off undetected (stolen), damaged or destroyed stocks. In today's conditions, Ukrainian enterprises suffer significant losses as a result of martial law. The use of unmanned aerial vehicles and missiles by Russia, the occupation of a large part of Ukrainian territories, the location of enterprises in war zones or close to them leads to the loss of enterprise assets. We have researched that these losses can occur as a result of direct ammunition hits, fires, and breakdown of equipment used to preserve stocks. Unfortunately, the loss and deterioration of stocks leads to the mismatch of their recognition as assets and the subsequent receipt of economic benefits.

The Use of Unmanned Aerial Vehicles in the Documentation of Recreation Areas Within The Scope Of Cultural Heritage: The Example of The Ancient City Of Kilistra

Historical buildings have been damaged by natural events and human hands for centuries. For this reason, the importance of studies on the preservation, documentation and restoration of historical buildings is increasing all over the world. In addition, the use of technological opportunities is also becoming important in order to carry out more comprehensive studies on the sustainability of historical buildings. Studies on modeling many cultural heritage (CH) works around the world continue. In order to evaluate these models in reconstruction studies, precise data about the structure must be obtained. In parallel with technological developments, the measurement sensitivity of the devices used in this field has also increased. In addition, different technological tools that will assist in measurement activities are entering our lives day by day. Unmanned Aerial Vehicle (UAV) systems are one of them. 3D data obtained from UAVs and photogrammetric systems are sensitive data obtained quickly and at low cost. This data provides the opportunity to be used in interdisciplinary studies. This study investigates the use of UAVs in the digital documentation study of the Ancient City of Kilistra in Konya/Gökyurt. The results of the study showed that UAVs are much more practical and effective than traditional measurement techniques. It has been revealed that, if appropriate conditions are provided, UAVs can be a suitable alternative in terms of time and cost in documenting recreational CH sites. Finally, it has been determined that the high-resolution digital maps obtained not only speed up the process of documenting large areas, but also provide advantages in terms of accuracy and precision.

Autonomous UAV navigation using deep learning-based computer vision frameworks: A systematic literature review

The increasing use of unmanned aerial vehicles (UAVs) in both military and civilian applications, such as infrastructure inspection, package delivery, and recreational activities, underscores the importance of enhancing their autonomous functionalities. Artificial intelligence (AI), particularly deep learning-based computer vision (DL-based CV), plays a crucial role in this enhancement. This paper aims to provide a systematic literature review (SLR) of Scopus-indexed research studies published from 2019 to 2024, focusing on DL-based CV approaches for autonomous UAV applications. By analyzing 173 studies, we categorize the research into four domains: sensing and inspection, landing, surveillance and tracking, and search and rescue. Our review reveals a significant increase in research utilizing computer vision for UAV applications, with over 39.5 % of studies employing the You Only Look Once (YOLO) framework. We discuss the key findings, including the dominant trends, challenges, and opportunities in the field, and highlight emerging technologies such as in-sensor computing. This review provides valuable insights into the current state and future directions of DL-based CV for autonomous UAVs, emphasizing its growing significance as legislative frameworks evolve to support these technologies.

The Use of Unmanned Aerial Vehicles to Assist Lifeguards Identifying, Preventing, and Rescuing: A Systematic Review.

When dealing with drowning scenarios, time is of capital importance. In this sense, the search for innovative strategies that provide more effectiveness and safety for lifesaving professionals, as well as for bathers, is essential. The aim of this study was to systematically review the feasibility and effectiveness of unmanned aerial vehicles as a tool to assist lifeguards in performing their duties. A systematic literature search was conducted in October 2023 across the databases PubMed, Web of Science, Academic Search Complete, and Scopus, using the search terms: ((drone) OR (Unmanned Aerial Vehicles)) AND ((drowning) OR (water rescue)). A total of 5 studies were included in this review, whose quality ranged from moderate to strong, according to the EPHPP Quality Assessment Tool for Quantitative Studies. The results suggest that unmanned aerial vehicles may assist lifeguards in the prevention, detection, and rescue of drowning victims and people in distress in the water. Although preliminary, evidence suggests that lifeguards would benefit from the use of unmanned aerial vehicles regarding lifeguards' and others' safety.

Factors controlling peat soil thickness and carbon storage in temperate peatlands based on UAV high-resolution remote sensing

Peatlands store a large amount of carbon. However, peatlands are complex ecosystems, and acquiring reliable estimates of how much carbon is stored underneath the Earth’s surface is inherently challenging, even at small scales. Here, we aim to establish links between the above- and below-ground factors that control soil carbon status, identify the key environmental variables associated with carbon storage, as well as to explore the potential for using Unmanned Aerial Vehicle (UAV) remote sensing for spatial mapping of peatlands. We combine UAVs equipped with Red-Green-Blue (RGB), multispectral, thermal infrared, and light detection and ranging (LiDAR) sensors with ground-penetrating radar (GPR) technology and traditional field surveys to provide a comprehensive, 3-dimensional mapping of a peatland hillslope-floodplain landscape in the Belgian Hautes Fagnes. Our results indicate that both peat thickness and soil organic carbon (SOC) stock (top 1 m) are spatially heterogeneous and that the contributions from the surface topography to peat thickness and SOC stock varied from micro- to macro-scales. Peat thickness was more strongly controlled by macro-topography (R2 = 0.46) than SOC stock, which was more influenced by micro-topography (R2 = 0.21). Current vegetation had little predictive power for explaining their spatial variability. Additionally, the UAV data provided accurate estimates of both peat thickness and SOC stock, with RMSE and R2 values of 0.16 m and 0.85 for the peat thickness, and 59.25 t/ha and 0.85 for the SOC stock. However, similar performance can already be achieved by using only topographical data from the LiDAR sensor (for peat thickness) and a combination of peat thickness and topography (for SOC stock) as predictor variables. Our study bridges the gap between surface observations and the hidden carbon reservoir below. This not only allows us to improve our ability to assess the spatial distribution of SOC stocks, but also contributes to our understanding of the environmental factors associated with SOC storage in these highly heterogeneous landscapes, providing insights for environmental science and climate projections.

Assessing the spatial-temporal impacts of underlying surfaces on 3D thermal environment: A field study based on UAV vertical measurements

The composition of underlying surfaces has a significant impact on the thermal environment within urban canopies. Current research predominantly focuses on surface temperature (Ts) and near-surface air temperature (Ta) at the pedestrian scale. However, studies on the Spatial-temporal distribution characteristics of the three-dimensional (3D) thermal environment within urban canopies remain scarce. This study selected 7 typical underlying surfaces and investigated the Spatial-temporal distribution characteristics of Ts and vertical Ta from 1.5 to 200 m by using unmanned aerial vehicle (UAV) and fixed measurement techniques. The results indicated that the largest differences in 3D thermal environment characteristics across different underlying surfaces occurred at 14:00, with the smallest at 05:30. Each underlying surface exhibited distinct day-night variations in their impacts on the 3D thermal environment, with the water body showing significant day-night difference. The dense forest positively regulated the urban thermal environment, but at night, it slightly adversely effected the thermal environment at the pedestrian scale. The bare soil had the second most adverse effect on the 3D thermal environment during the day, following the impervious pavement; thus, exposure of the land should be minimized in urban environments. There was a significant positive correlation between Ts and Ta at different heights (P < 0.05). At 05:30, 14:00, and 21:00, the heights of vertical influence of Ts on Ta were 2 m, 50 m, and 32 m, respectively. These results provide important references for urban environmental managers to mitigate the 3D urban heat island (3D UHI).

Adaptive Clustering and Scheduling for UAV-Enabled Data Aggregation

Using unmanned aerial vehicles (UAVs) is an effective way to gather data from Internet of Things (IoT) devices. To reduce data gathering time and redundancy, thereby enabling the timely response of state-of-the-art systems, one can partition a network into clusters and perform aggregation within each cluster. Existing works solved the UAV trajectory planning problem, in which the energy consumption and/or flight time of the UAV is the minimization objective. The aggregation scheduling within each cluster was neglected, and they assumed that data must be ready when the UAV arrives at the cluster heads (CHs). This paper addresses the minimum time aggregation scheduling problem in duty-cycled networks with a single UAV. We propose an adaptive clustering method that takes into account the trajectory and speed of the UAV. The transmission schedule of IoT devices and the UAV departure times are jointly computed so that (1) the UAV flies continuously throughout the shortest path among the CHs to minimize the hovering time and energy consumption, and (2) data are aggregated at each CH right before the UAV arrival, to maximize the data freshness. Intensive simulation shows that the proposed scheme reduces up to 35% of the aggregation delay compared to other benchmarking methods.

Prediction accuracy and repeatability of UAV based biomass estimation in wheat variety trials as affected by variable type, modelling strategy and sampling location

BackgroundThis study explores the use of Unmanned Aerial Vehicles (UAVs) for estimating wheat biomass, focusing on the impact of phenotyping and analytical protocols in the context of late-stage variety selection programs. It emphasizes the importance of variable selection, model specificity, and sampling location within the experimental plot in predicting biomass, aiming to refine UAV-based estimation techniques for enhanced selection accuracy and throughput in variety testing programs.ResultsThe research uncovered that integrating geometric and spectral traits led to an increase in prediction accuracy, whilst a recursive feature elimination (RFE) based variable selection workflowled to slight reductions in accuracy with the benefit of increased interpretability. Models, tailored to specific experiments were more accurate than those modelling all experiments together, while models trained for broad-growth stages did not significantly increase accuracy. The comparison between a permanent and a precise region of interest (ROI) within the plot showed negligible differences in biomass prediction accuracy, indicating the robustness of the approach across different sampling locations within the plot. Significant differences in the within-season repeatability (w2) of biomass predictions across different experiments highlighted the need for further investigation into the optimal timing of measurement for prediction.ConclusionsThe study highlights the promising potential of UAV technology in biomass prediction for wheat at a small plot scale. It suggests that the accuracy of biomass predictions can be significantly improved through optimizing analytical and modelling protocols (i.e., variable selection, algorithm selection, stage-specific model development). Future work should focus on exploring the applicability of these findings under a wider variety of conditions and from a more diverse set of genotypes.

Yield Prediction Models for Rice Varieties Using UAV Multispectral Imagery in the Amazon Lowlands of Peru

Rice is cataloged as one of the most widely cultivated crops globally, providing food for a large proportion of the global population. Integrating Geographic Information Systems (GISs), such as unmanned aerial vehicles (UAVs), into agricultural practices offers numerous benefits. UAVs, equipped with imaging sensors and geolocation technology, enable precise crop monitoring and management, enhancing yield and efficiency. However, Peru lacks sufficient experience with the application of these technologies, making them somewhat unfamiliar in the context of modern agriculture. In this study, we conducted experiments involving four distinct rice varieties (n = 24) at various stages of growth to predict yield using vegetation indices (VIs). A total of nine VIs (NDVI, GNDVI, ReCL, CIgreen, MCARI, SAVI, CVI, LCI, and EVI) were assessed across four dates: 88, 103, 116, and 130 days after sowing (DAS). Pearson correlation analysis, principal component analysis (PCA), and multiple linear regression were used to build prediction models. The results showed a general prediction model (including all the varieties) with the best performance at 130 days after sowing (DAS) using NDVI, EVI, and SAVI, with a coefficient of determination (adjusted-R2 = 0.43). The prediction models by variety showed the best performance for Esperanza at 88 DAS (adjusted-R2 = 0.94) using EVI as the vegetation index. The other varieties showed their best performance using different indices at different times: Capirona (LCI and CIgreen, 130 DAS, adjusted-R2 = 0.62); Conquista Certificada (MCARI, 116 DAS, R2 = 0.52); and Conquista Registrada (CVI and LCI, 116 DAS, adjusted-R2 = 0.79). These results provide critical information for optimizing rice crop management and support the use of unmanned aerial vehicles (UAVs) to inform timely decision making and mitigate yield losses in Peruvian agriculture.

Advancing primate surveillance with image recognition techniques from unmanned aerial vehicles.

Using unmanned aerial vehicles (UAVs) for surveys on thermostatic animals has gained prominence due to their ability to provide practical and precise dynamic censuses, contributing to developing and refining conservation strategies. However, the practical application of UAVs for animal monitoring necessitates the automation of image interpretation to enhance their effectiveness. Based on our past experiences, we present the Sichuan snub-nosed monkey (Rhinopithecus roxellana) as a case study to illustrate the effective use of thermal cameras mounted on UAVs for monitoring monkey populations in Qinling, a region characterized by magnificent biodiversity. We used the local contrast method for a small infrared target detection algorithm to collect the total population size. Through the experimental group, we determined the average optimal grayscale threshold, while the validation group confirmed that this threshold enables automatic detection and counting of target animals in similar datasets. The precision rate obtained from the experiments ranged from 85.14% to 97.60%. Our findings reveal a negative correlation between the minimum average distance between thermal spots and the count of detected individuals, indicating higher interference in images with closer thermal spots. We propose a formula for adjusting primate population estimates based on detection rates obtained from UAV surveys. Our results demonstrate the practical application of UAV-based thermal imagery and automated detection algorithms for primate monitoring, albeit with consideration of environmental factors and the need for data preprocessing. This study contributes to advancing the application of UAV technology in wildlife monitoring, with implications for conservation management and research.

Defect detection with ego-noise reduction based on multimodal information in UAV hammering inspection

In this paper, we introduce a novel approach for defect detection in hammering inspections using Unmanned Aerial Vehicles (UAVs). Despite the promising application of UAVs for inspecting hard-to-reach structures, like bridges, their efficiency is often compromised by the significant ego noise produced by their motor-propeller systems. This noise complicates the discrimination between healthy and defective hammering sounds. In previous research, methods to improve robustness through supervised learning have been proposed; however, these methods require the labeling of hammering sounds by skilled inspectors to train the discrimination model. To overcome this problem, we propose an ego-noise reduction method based on propeller vibrations. By reducing ego noise and thereby making the characteristics of hammering sound more dramatically clear, we enable unsupervised defect detection amidst ego noise. Our experiments with concrete specimens demonstrate that our technique achieves defect detection with an accuracy on par with the supervised method. The proposed method proves especially beneficial for hammering inspections, in which the domain gap–the variability in acoustic signatures of hammering sounds caused by differences in concrete mix ratios and curing conditions from one site to another–presents a significant challenge. Our approach effectively adapts to these variations, ensuring reliable defect detection across diverse construction environments.

Optimized Radio Frequency Footprint Identification Based on UAV Telemetry Radios.

With the widespread use of unmanned aerial vehicles (UAVs), the detection and identification of UAVs is a vital security issue for the safety of airspace and ground facilities in the no-fly zone. Telemetry radios are important wireless communication devices for UAVs, especially in UAVs beyond the visual line of sight (BVLOS) operating mode. This work focuses on the UAV identification approach using transient signals from UAV telemetry radios instead of the signals from UAV controllers that the former research work depended on. In our novel UAV Radio Frequency (RF) identification system framework based on telemetry radio signals, the EC-α algorithm is optimized to detect the starting point of the UAV transient signal and the detection accuracy at different signal-to-noise ratios (SNR) is evaluated. In the training stage, the Convolutional Neural Network (CNN) model is trained to extract features from raw I/Q data of the transient signals with different waveforms. Its architecture and hyperparameters are analyzed and optimized. In the identification stage, the extracted transient signals are clustered through the Self-Organizing Map (SOM) algorithm and the Clustering Signals Joint Identification (CSJI) algorithm is proposed to improve the accuracy of RF fingerprint identification. To evaluate the performance of our proposed approach, we design a testbed, including two UAVs as the flight platform, a Universal Software Radio Peripheral (USRP) as the receiver, and 20 telemetry radios with the same model as targets for identification. Indoor test results show that the optimized identification approach achieves an average accuracy of 92.3% at 30 dB. In comparison, the identification accuracy of SVM and KNN is 69.7% and 74.5%, respectively, at the same SNR condition. Extensive experiments are conducted outdoors to demonstrate the feasibility of this approach.

Training of future policeman for the use of unmanned aerial vehicle in professional activities

Training of future policeman for the use of unmanned aerial vehicle in professional activities

The Importance of the Use of Unmanned Aerial Vehicles (UAVs) in the Oil and Gas Industry

The oil and gas industry, renowned for its extensive and intricate operations, confronts numerous challenges in maintaining and inspecting its infrastructure. Traditional inspection methodologies, often reliant on manual labor, are not only time-consuming and costly but also expose workers to hazardous conditions. These conventional methods necessitate the shutdown of operations, leading to considerable productivity losses and elevated costs. Over the past decade, the advent of unmanned aerial vehicles (UAVs) has introduced a transformative technology in this sector, providing a safer, more efficient, and cost-effective alternative to traditional inspection techniques. UAVs, equipped with advanced sensors and high-resolution cameras, facilitate detailed visual and thermal inspections of critical assets such as pipelines, flare stacks, and offshore platforms. These capabilities enable companies to perform routine inspections without halting operations, thereby minimizing downtime and operational disruptions. The implementation of UAV technology has notably enhanced safety by reducing the need for human intervention in perilous environments. For instance, the traditional inspection of flare stacks necessitates workers to ascend these structures, posing significant risks. UAVs obviate this requirement by delivering real-time visual data from secure distances. Moreover, the precision and accuracy inherent in UAV inspections contribute to the early detection of defects and potential issues, allowing for prompt maintenance and repairs, which further augment safety and operational efficiency. The environmental benefits of UAV technology are also noteworthy. By diminishing the reliance on heavy machinery and extensive transportation typically associated with inspections and maintenance, UAVs aid in reducing carbon emissions, thus aligning with the industry&amp;apos;s sustainability objectives. In summary, the integration of UAVs into the oil and gas industry&amp;apos;s inspection protocols represents a significant technological advancement. This integration aligns with the sector&amp;apos;s goals of enhancing safety, efficiency, and environmental sustainability, marking a pivotal step forward in the industry&amp;apos;s evolution.

Practical Performance Analysis of MDI-QKD with Orbital Angular Momentum on UAV Relay Platform.

The integration of terrestrial- and satellite-based quantum key distribution (QKD) experiments has markedly advanced global-scale quantum networks, showcasing the growing maturity of quantum technologies. Notably, the use of unmanned aerial vehicles (UAVs) as relay nodes has emerged as a promising method to overcome the inherent limitations of fiber-based and low-Earth orbit (LEO) satellite connections. This paper introduces a protocol for measurement-device-independent QKD (MDI-QKD) using photon orbital angular momentum (OAM) encoding, with UAVs as relay platforms. Leveraging UAV mobility, the protocol establishes a secure and efficient link, mitigating threats from untrusted UAVs. Photon OAM encoding addresses reference frame alignment issues exacerbated by UAV jitter. A comprehensive analysis of atmospheric turbulence, state-dependent diffraction (SDD), weather visibility, and pointing errors on free-space OAM-state transmission systems was conducted. This analysis elucidates the relationship between the key generation rate and propagation distance for the proposed protocol. Results indicate that considering SDD significantly decreases the key rate, halving previous data results. Furthermore, the study identifies a maximum channel loss capacity of 26 dB for the UAV relay platform. This result is pivotal in setting realistic parameters for the deployment of UAV-based quantum communications and lays the foundation for practical implementation strategies in the field.

Garlic yield monitoring using vegetation indices and texture features derived from UAV multispectral imagery

Remote sensing and machine learning are widely used to estimate crop yield. The use of these technologies for yield estimation of bulbous vegetables is challenging because the yield is underground and can't be directly monitored by remote sensing images. Among the bulbous vegetables, garlic (Allium sativum L.) is one of the most widely cultivated in the world. The aim of this study was to develop an accurate and transferable machine learning model to monitor and to estimate garlic yield using unmanned aerial vehicle (UAV) multispectral images. Data were collected over three growing seasons (2021, 2022, and 2023) at four different garlic phenological phases (202, 405, 407, and 409 of BBCH). The random forest (RF) algorithm was used to estimate the garlic yield by comparing two different training feature sets: the vegetation indices (VIs) and the VIs with the addition of the texture features extracted from the UAV images. The most important VIs were selected using the recursive feature elimination algorithm. Two estimation methods were compared: a direct bulb estimation and an indirect bulb estimation using the aboveground biomass as a proxy. To evaluate the transferability of the RF models, two cross-validation strategies were compared: a nested leave-one-fold-out cross-validation (LOFOCV) and a leave-one-year-out cross-validation (LOYOCV). The best performance was achieved by the direct bulb estimation using the LOFOCV strategy. Regarding the transferability of the RF models between years (i.e. LOYOCV), the indirect estimation method showed a higher transferability than the direct estimation method. Finally, the addition of texture features improved the accuracy of the RF models, but in general, their contribution was poor. This study demonstrated that the yield of bulbous vegetables can be accurately estimated by remote sensing, and that UAVs are a suitable tool to provide rapid and reliable support for garlic yield monitoring.

Grouping Land Cover Using Orthophoto in Small Islands: An Application of Low-Cost UAV on Mansinam Island

Land cover is crucial for island management, but the lack of accessible and high-resolution remote sensing data has reduced investigations on small islands, including land cover identification. Therefore, this study aimed to investigate land cover using Unmanned Aerial Vehicle (UAV) technology, providing very high-resolution images. Classification and delineation were conducted using automatic segmentation followed by manual reinterpretation and visual verification. The results showed 14 cover classes, consisting of 8 vegetated and six non-vegetated categories. Forest cover on Mansinam island accounted for 75.5% or 302.4 ha, which was evenly distributed. Furthermore, primary forest covered 31.91% or 127.74 ha, and secondary covered 43.63% or 174.68 ha. The classification achieved an overall accuracy of 96% and a kappa coefficient of 0.94. Low-cost UAVs effectively produced high-resolution aerial images of small islands for land cover identification. Therefore, future studies were recommended to consider whether segmentation can reliably distinguish between primary and secondary forests, as well as assess the impact of flight altitude on segmentation accuracy using ground control points. The results were also expected to support spatial planning or sustainable forest and environment management on Mansinam Island.