Documenting customary land boundaries using unmanned aerial vehicle imagery and artificial intelligence

Hyperspectral imagery from an Unmanned Aerial Vehicle (UAV) has been used to predict soil organic carbon (SOC) content. This study compared UAV hyperspectral imagery with Sentinel-2A multispectral imagery and soil spectra in the laboratory to predict the SOC content in Suaeda salsa (S. salsa) coastal wetland. Characteristic variables were determined through correlation analysis comparing observed SOC with constructed spectral indices of SOC, index variables, and spectral bands. Predict Models of SOC were constructed by multilayer perceptron (MLP), the prediction accuracy was evaluated and the spatial maps of SOC were produced. The results showed that estimating SOC using UAV, Sentinel imagery, and laboratory spectra has great potential when using MLP algorithm. The SOC content can be accurately predicted by the laboratory spectra without spatial map. The accuracy of SOC estimation was better when based on Sentinel-2A imagery than UAV imagery, with the model determination coefficients (R 2) of 0.838 and 0.638, the root mean square error (RMSE) of 0.398 and 0.596 g kg−1, ratio of performance to deviations (RPD) of 2.485 and 1.660, and bias of − 0.028 and 0.085 g kg−1, respectively. The SOC estimation model using Sentinel-2A imagery demonstrated better stability than that using UAV imagery, with the mean uncertainty standard deviations (SD) of 0.226 and 0.291 g kg−1, respectively. The spatial SOC maps generated from both UAV and Sentinel-2A imagery presented more subtle variation information when compared with the spatial interpolation results. Although the prediction accuracy of SOC from UAV was lower than that from Sentinel-2A, UAV imagery could disclose more fine distribution information of SOC at the local level. The UAV imagery can be used to estimate SOC in coastal wetlands but with a higher cost than freely available Sentinel images. We recommend using sentinel imagery to predict SOC at regional, national, or even global levels.

Wild birds are monitored with the important objectives of identifying their habitats and estimating the size of their populations. Especially in the case of migratory bird, they are significantly recorded during specific periods of time to forecast any possible spread of animal disease such as avian influenza. This study led to the construction of deep-learning-based object-detection models with the aid of aerial photographs collected by an unmanned aerial vehicle (UAV). The dataset containing the aerial photographs includes diverse images of birds in various bird habitats and in the vicinity of lakes and on farmland. In addition, aerial images of bird decoys are captured to achieve various bird patterns and more accurate bird information. Bird detection models such as Faster Region-based Convolutional Neural Network (R-CNN), Region-based Fully Convolutional Network (R-FCN), Single Shot MultiBox Detector (SSD), Retinanet, and You Only Look Once (YOLO) were created and the performance of all models was estimated by comparing their computing speed and average precision. The test results show Faster R-CNN to be the most accurate and YOLO to be the fastest among the models. The combined results demonstrate that the use of deep-learning-based detection methods in combination with UAV aerial imagery is fairly suitable for bird detection in various environments.

Unmanned aerial vehicle (UAV) imagery is considered the best approach to acquire high-resolution data over a large area in a short time, such as in the case of coastal wetlands comprising small-sized patches of perennial plants. However, to the best of our knowledge, no previous studies have used UAV imagery to quantitatively estimate blue carbon in coastal wetlands. Therefore, this study aimed to estimate the area-wide carbon stocks accumulated in halophyte habitats through the case studies of the southern coastal wetlands in South Korea using UAV imagery. Accordingly, object-based image analysis was used to identify the spatial distribution of major halophytes based on the spectral reflectance derived from multispectral UAV images. Elemental analysis (EA) was performed to estimate the content of organic carbon accumulated in individual halophytes, and the halophyte communities identified in the case study area were confirmed to have accumulated 20.80 MgC of blue carbon. These results demonstrate that applying EA to UAV images can be an innovative method to solve the spatial uncertainty encountered in the monitoring of blue carbon in existing field-based surveys and traditional remote sensing.

Non‐native species maps are important tools for understanding and managing biological invasions. We demonstrate a novel approach to extend presence modeling to map fractional cover (FC) of non‐native yellow sweet clover Melilotus officinalis in the Northern Great Plains, USA. We used ensembles of MaxEnt models to map FC across landscapes from satellite imagery trained from regional aerial imagery that was trained by local unmanned aerial vehicle (UAV) imagery. Clover cover from field surveys and classified UAV imagery were nearly identical (n = 22, R2 = 0.99). Two classified UAV images provided training data to map clover presence with MaxEnt and National Agricultural Imagery Program (NAIP) aerial imagery. We binned cover predictions from NAIP imagery within each Sentinel‐2 pixel into eight cover classes to create pure (100%) and FC (20%–95%) training data and modeled each class separately using MaxEnt and Sentinel‐2 imagery. We mapped pure clover with one classification threshold and compared its performance to 15 candidate maps that included FC predictions outside pure predictions. Each FC map represented alternative combinations of five MaxEnt thresholds and three approaches to assign cover to pixels with multiple predictions from the FC ensemble. Evaluations of performance with independent datasets revealed maps including FC corresponded to field (n = 32, R2 range: 0.39–0.68) and UAV (n = 20, R2 range: 0.61–0.84) data better than pure clover maps (R2 = 0.15 and 0.31, respectively). Overall, the pure clover map predicted 3.2% cover, whereas the three best performing FC maps predicted 6.6%–8.0% cover. Including FC predictions increased accuracy and cover predictions which can improve ecological understanding of invasions. Our method allows efficient FC mapping for vegetative species discernible in UAV imagery and may be especially useful for mapping rare, irruptive or patchily distributed species with poor representation in field data, which challenges landscape‐level mapping.

Accurate traffic prediction is crucial to alleviating traffic congestion in cities. Existing physical sensor-based traffic data acquisition methods have high transmission costs, serious traffic information redundancy, and large calculation volumes for spatiotemporal data processing, thus making it difficult to ensure accuracy and real-time traffic prediction. With the increasing resolution of UAV imagery, the use of unmanned aerial vehicles (UAV) imagery to obtain traffic information has become a hot spot. Still, analyzing and predicting traffic status after extracting traffic information is neglected. We develop a framework for traffic speed extraction and prediction based on UAV imagery processing, which consists of two parts: a traffic information extraction module based on UAV imagery recognition and a traffic speed prediction module based on deep learning. First, we use deep learning methods to automate the extraction of road information, implement vehicle recognition using convolutional neural networks and calculate the average speed of road sections based on panchromatic and multispectral image matching to construct a traffic prediction dataset. Then, we propose an attention-enhanced traffic speed prediction module that considers the spatiotemporal characteristics of traffic data and increases the weights of key roads by extracting important fine-grained spatiotemporal features twice to improve the prediction accuracy of the target roads. Finally, we validate the effectiveness of the proposed method on real data. Compared with the baseline algorithm, our algorithm achieves the best prediction performance regarding accuracy and stability.

Unmanned Aerial Vehicle (UAV) imagery are being assessed for analyzing within field spatial variability for agricultural precision management, because unmanned aerial vehicle imagery may be acquired quickly during critical periods of rapid crop growth. The objective of this study was to evaluate the use of unmanned aerial vehicle for the monitoring barley growth. Unmanned aerial vehicle imagery obtained from middle February to late June in Wanju, Jeollabuk-do. Unmanned aerial vehicle imagery corrected geometrically and atmospherically to calculate normalized difference vegetation index (NDVI). We analyzed the relationships between NDVI UAV of barley and biophysical measurements such as plant height, number of tiller and shoot dry weight over an entire barley growth period. The similar trend between NDVI UAV and growth parameters was shown. Correlation analysis between NDVI UAV and barley growth parameters revealed that NDVI UAV was highly correlated with shoot dry weight (r=0.932) and plant height (r=0.879). According to the relationship among growth parameters and NDVI UAV , the temporal variation of NDVIUAV was significant to interpret barley growth. The spatial distribution map of barley growth was in strong agreement with the field measurements in terms of geographical variation and relative numerical values when NDVI UAV was applied to power function. From these results, NDVI UAV can be used as a new tool for monitoring barley growth.

Tree height is one of the key parameters in forest plantations that plays a crucial role in estimation of above-ground biomass (AGB) of trees and stands. The parameter may be obtained by different methods from airborne remotely sensed datasets such shadow length of each tree individual or crown height models (CHMs). However, tree height estimation based on shadow length might be biased considering diverse topography of forest sites. Therefore, this study aims to develop a reliable method to estimate tree height in a plantation forest using shadow length on UAV imagery. First, heights of 151 pine (Pinus eldarica) trees were precisely measured in Pardisan Park, North Khorasan province, Iran. Additionally, a collection of images was captured by a Phantom 4Pro UAV in order to illustrate the study area. Then, two different approaches were considered to estimate the height of trees. In the first approach, tree heights were estimated based on shadow length on the UAV orthomosaic and correcting the effect of slope. The second approach considered the UAV-based CHM and height estimation using CHM segmentation and local maximum filtering. The results showed that tree heights estimated by the first approach were not significantly different from the in-situ data (p=0.298). Furthermore, the heights estimated by the slope corrected shadow length showed higher precision compared to the heights estimated by the shadow length without slope correction (Relative root mean squared error, RRMSE 5.6% and 8.2% respectively). However, the heights obtained from the first approach were less precise than the second approach (RRMSE 5.6% and 4.2% respectively). In general, it was concluded that height estimation of pine trees based on shadow length after correction of slope effects can be considered as a reliable approach, although CHM is more efficient in estimating the tree heights. The findings of this study are applicable for height estimation of pine trees within plantation forests on UAV imagery.

Multi-scale adversarial network for vehicle detection in UAV imagery

<p>Detailed information on rock formations and assessment of their geological structures, such as joints, faults, shears and bedding planes, are required for evaluation of rock integrity and stability. Our research demonstrates a comprehensive approach for producing high spatial resolution 3D information of rock formations from unmanned aerial vehicle (UAV) imagery for rock joint identification, and presents an innovative technique for using Terrestrial Laser Scanning (TLS) data to derive ground control points (GCPs) for geo-referencing of UAV imagery of vertical rock walls. UAV imagery was collected from a freestanding 90 m tall rock formation with a 3.5 km perimeter, via oblique rock façade scans and also at-nadir, covering a ground and rock facade area of approximately 0.32 and 0.25 km<sup>2</sup>, respectively. Seventy-two GCPs were distributed around the rock for geo-referencing of the UAV imagery. As GCPs could not be deployed on vertical rock walls, a TLS system was used for identification of 93 distinct natural features as pseudo-GCPs on the rock walls. Forty scans were collected and geo-referenced from triplets of GCPs placed on the ground near the TLS system for each scan. A Real-Time Kinematic (RTK) Global Navigation Satellite System (GNSS) survey was performed on the GCPs, using a base station and a rover. Continuously Operating Reference Stations (CORS) data were used to fix the position of the base station and improve the absolute geometric position to an average and lowest accuracy of 3.3 and 18.6 mm for 177 GCPs. A total of 44 façade scans, as well as 14 low and 5 high altitude nadir-viewing UAV flights, were undertaken with a Zenmuse X3 RGB camera mounted to a DJI Matrice 100 platform, resulting in a collection of nearly 17,000 photos. The five high altitude flights were designed to include a larger area around the rock to incorporate additional GCPs, while the low altitude flights were to increase the spatial resolution of the imaged rock. Flight planning was undertaken with the Universal Ground Control Station Client application. Façade scans were flown horizontally and parallel to the rock walls at a distance of 20-30 m and at heights between 10-80 m with sidelaps >70% between horizontal flight lines and >80% forward overlap along flight lines. Façade scans were collected with a 4° viewing angle to ensure the base of the rock was included. The Agisoft MetaShape software was initially used to generate a sparse point cloud using all façade scans and nadir-viewing imagery. Geo-referencing of the UAV imagery was based on 136 GCPs, which produced an accuracy of 0.1558 m, with an addition 100 control points kept aside for independent evaluation, yielding an accuracy of 0.2018 m. Subsequent image processing was split into 14 evenly sized “chunks” to enable more efficient processing of a dense point cloud (4.33 billion points) and 3D model (mesh with 129.5 million faces and texture layer) for the entire area. The produced 3D model was found suitable for identification of rock joints larger than 1 m in length.</p>

Can reproducible comparisons of cereal genotypes be generated in field experiments based on UAV imagery using RGB cameras?

Small object detection is a challenging computer vision problem due to their low feature representation in the images and factors such as occlusions and noise. In images captured from a camera mounted on an unmanned aerial vehicle (UAV), objects are usually acquired in small sizes depending on the UAV flight altitude. The state-of-the-art object detectors often have lower detection accuracy with small objects. New approaches of combining features at multi-levels in the network helps in improving the object detection performance. In this paper, we propose a multi-scale approach of low-level feature combinations with deconvolutional modules on a single shot multibox detection (SSD) object detector to improve the small object detection in images acquired from a UAV. The proposed SSD based architecture is evaluated on UAV datasets to compare its performance with the state-of-the-art detectors.

Data on forest parameters defining the structure, health and condition of a forest stand is essential for forest management and conservation. The increasing frequency of forest changes, such as those caused by climate change-related drought and heat events, highlight the importance of having a forest database with high spatial and temporal resolution. Automated forest parameter extraction based on unmanned aerial vehicle (UAV) imagery is a cost-effective way to address the need for accurate and up-to-date forest data.The aim of this project is to develop user-friendly tools based on optical data from UAVs that can be applied to accurately and efficiently conduct digital forest inventories. We are using spectral and geometric information from UAV data to create methods for automated derivation of forest parameters such as diameter at breast height (DBH), tree stem positions, individual tree crown delineation, and coarse wood debris. These methods are being designed with the practical needs of potential users from the forestry sector in mind. Different flight configurations, such as nadir and oblique camera angles, as well as different acquisition times, were combined to generate structure from motion (SfM) data products (dense 3D point clouds, orthomosaics and height models) containing both ground and canopy information. For a study site within the Hainich National Park, Germany, we analyzed how leaf-off and leaf-on data can be combined to improve the derivation of stand parameters, such as tree stem positions and individual tree crowns, using point- and raster-based algorithms. Additionally, DBH on an individual tree basis was derived for the same study site using the cast shadows of tree trunks. To do so, a deep learning model was trained to identify stem shadows based on an orthomosaic of only ground points acquired during sunny and leaf-off conditions.

Mangrove forests provide critical ecosystem services for coastal protection, biodiversity conservation, and carbon sequestration, yet monitoring their condition at crown level remains challenging using traditional field surveys alone. This study investigates the use of high-resolution unmanned aerial vehicle (UAV) imagery and deep learning for automated delineation of mangrove crowns in Lian, Batangas, Philippines. Orthomosaic RGB images acquired at 25 ~ m altitude (GSD ${\approx} 1.34 ~ \mathrm {cm} ^ { 2}$ pixel) were tiled into $256 \times 256$ px patches and manually annotated into three structural classes: Individual Crowns (IC), Crown Clusters (CC), and Canopy Gaps (CG). To reduce systematic confusion between gaps and low-canopy vegetation, a vegetation-focused dataset variant retaining only IC and CC labels was also constructed. We fine-tuned multiple Mask R-CNN variants (R101C4, R101DC5, R101FPN, X101-32 $\times$ 8d-FPN) pretrained on COCO and benchmarked them using Average Precision (AP) across object scales. The R101FPN backbone achieved the best trade-off between accuracy and efficiency (Scenario 1), with mean AP=17.71, $\mathrm{AP}_{50}=32.43$, and shortest training time per 1k iterations. Using this backbone, a second scenario incorporating an expanded, multi-site dataset and removal of the gap class further improved overall detection performance to $\mathrm{AP}=21.71$ and $\mathrm{AP}_{50}=46.24$, at substantially reduced training time. To quantify segmentation quality beyond box-level metrics, we computed region- and boundary-aware measures for the best model: mean Dice coefficient of 0.9273, Boundary IoU of 0.5181, and symmetric Hausdorff distance of 18.54 px over 295 validation images. These results indicate accurate recovery of crown extent, with moderate boundary imprecision in dense clusters and low-contrast conditions. Overall, the findings demonstrate that UAV-derived imagery coupled with Mask R-CNN provides a practical and effective framework for operational mangrove crown mapping in the Philippines, and highlight remaining challenges related to small crowns, crown clusters, and gap representation. -Mangrove forests, UAV imagery, Deep learning, Mask R-CNN, ResNet101, Object detection, Instance segmentation

Estimating mangrove leaf area index based on red-edge vegetation indices: A comparison among UAV, WorldView-2 and Sentinel-2 imagery

The paper aims to promote the growth of party building work in colleges and universities to improve school party organization, team management and strengthen party member ideological construction and overall party quality. We design intelligent party member business knowledge learning classrooms using deep learning to improve the quality of party members. First, we develop a convolutional neural network (CNN)-based classroom face recognition system and improve its loss function using the associated theory of the Visual Geometry Group 16 (VGG-16) model. Then, using the Single Shot Multi-Box Detector (SSD), we establish a classroom standing behavior identification system. The experimental results demonstrate that the accuracy rate of the conventional VGG-16 in the face recognition system is 93.5%, while the upgraded VGG-16 is 96.5%, with a 3.2% increase over the baseline models.