High-resolution Unmanned Aerial Vehicle Research Articles

Enhancing cotton seedling recognition: a method for high-resolution UAV remote sensing images

ABSTRACT Timely and accurate acquisition of cotton seedling information is crucial for seedling replenishment, yield increase, and effective cotton field management. This study proposes a cotton seedling recognition model and weed removal method for high-resolution UAV (Unmanned Aerial Vehicle) remote sensing images. First, the image information of cotton seedlings is enhanced by calculating the remote sensing index and using threshold segmentation. Second, the recognition model is constructed based on the spatial morphological characteristics of cotton seedlings, with weed interference removed using the straight-line method. Finally, the growth condition of cotton seedlings in farmland is comprehensively evaluated by calculating the seedling emergence rate and coverage rate. The experimental results show that: (1) the combination of the GLI (Green Leaf Index) and Otsu threshold segmentation algorithm effectively distinguishes cotton seedlings from the soil background and accurately portrays cotton seedling contours, further enhancing image information; (2) the proposed recognition model achieves an average overall accuracy of 95.75%, making it a practical method for cotton seedling recognition; and (3) the overall growth condition of cotton seedlings in the study area is good, enabling farmers to apply appropriate seedling replenishment and fertilizer based on the comprehensive cotton evaluation map to further improve the final cotton yield. This study aims to provide new research ideas for crop seedling identification and serve as a reference for the practical application of UAVs in crop monitoring.

PVF-10: A high-resolution unmanned aerial vehicle thermal infrared image dataset for fine-grained photovoltaic fault classification

Accurate identification of faulty photovoltaic (PV) modules is crucial for the effective operation and maintenance of PV systems. Deep learning (DL) algorithms exhibit promising potential for classifying PV fault (PVF) from thermal infrared (TIR) images captured by unmanned aerial vehicle (UAV), contingent upon the availability of extensive and high-quality labeled data. However, existing TIR PVF datasets are limited by low image resolution and incomplete coverage of fault types. This study proposes a high-resolution TIR PVF dataset with 10 classes, named PVF-10, comprising 5579 cropped images of PV panels collected from 8 PV power plants. These classes are further categorized into two groups according to the repairability of PVF, with 5 repairable and 5 irreparable classes each. Additionally, the circuit mechanisms underlying the TIR image features of typical PVF types are analyzed, supported by high-resolution images, thereby providing comprehensive information for PV operators. Finally, five state-of-the-art DL algorithms are trained and validated based on the PVF-10 dataset using three levels of resampling strategy. The results show that the overall accuracy (OA) of these algorithms exceeds 83%, with the highest OA reaching 93.32%. Moreover, the preprocessing procedure involving resampling and padding strategies are beneficial for improving PVF classification accuracy using PVF-10 datasets. The developed PVF-10 dataset is expected to stimulate further research and innovation in PVF classification.

Classifying Stand Compositions in Clover Grass Based on High-Resolution Multispectral UAV Images

In organic farming, clover is an important basis for green manure in crop rotation systems due to its nitrogen-fixing effect. However, clover is often sown in mixtures with grass to achieve a yield-increasing effect. In order to determine the quantity and distribution of clover and its influence on the subsequent crops, clover plants must be identified at the individual plant level and spatially differentiated from grass plants. In practice, this is usually done by visual estimation or extensive field sampling. High-resolution unmanned aerial vehicles (UAVs) offer a more efficient alternative. In the present study, clover and grass plants were classified based on spectral information from high-resolution UAV multispectral images and texture features using a random forest classifier. Three different timestamps were observed in order to depict the phenological development of clover and grass distributions. To reduce data redundancy and processing time, relevant texture features were selected based on a wrapper analysis and combined with the original bands. Including these texture features, a significant improvement in classification accuracy of up to 8% was achieved compared to a classification based on the original bands only. Depending on the phenological stage observed, this resulted in overall accuracies between 86% and 91%. Subsequently, high-resolution UAV imagery data allow for precise management recommendations for precision agriculture with site-specific fertilization measures.

An Improved Instance Segmentation Method for Fast Assessment of Damaged Buildings Based on Post-Earthquake UAV Images.

Quickly and accurately assessing the damage level of buildings is a challenging task for post-disaster emergency response. Most of the existing research mainly adopts semantic segmentation and object detection methods, which have yielded good results. However, for high-resolution Unmanned Aerial Vehicle (UAV) imagery, these methods may result in the problem of various damage categories within a building and fail to accurately extract building edges, thus hindering post-disaster rescue and fine-grained assessment. To address this issue, we proposed an improved instance segmentation model that enhances classification accuracy by incorporating a Mixed Local Channel Attention (MLCA) mechanism in the backbone and improving small object segmentation accuracy by refining the Neck part. The method was tested on the Yangbi earthquake UVA images. The experimental results indicated that the modified model outperformed the original model by 1.07% and 1.11% in the two mean Average Precision (mAP) evaluation metrics, mAPbbox50 and mAPseg50, respectively. Importantly, the classification accuracy of the intact category was improved by 2.73% and 2.73%, respectively, while the collapse category saw an improvement of 2.58% and 2.14%. In addition, the proposed method was also compared with state-of-the-art instance segmentation models, e.g., Mask-R-CNN and YOLO V9-Seg. The results demonstrated that the proposed model exhibits advantages in both accuracy and efficiency. Specifically, the efficiency of the proposed model is three times faster than other models with similar accuracy. The proposed method can provide a valuable solution for fine-grained building damage evaluation.

Decomposition of Submesoscale Ocean Wave and Current Derived from UAV-Based Observation

The consecutive submesoscale sea surface processes observed by an unmanned aerial vehicle (UAV) were used to decompose into spatial waves and current features. For the image decomposition, the Fast and Adaptive Multidimensional Empirical Mode Decomposition (FA-MEMD) method was employed to disintegrate multicomponent signals identified in sea surface optical images into modulated signals characterized by their amplitudes and frequencies. These signals, referred to as Bidimensional Intrinsic Mode Functions (BIMFs), represent the inherent two-dimensional oscillatory patterns within sea surface optical data. The BIMFs, separated into seven modes and a residual component, were subsequently reconstructed based on the physical frequencies. A two-dimensional Fast Fourier Transform (2D FFT) for each high-frequency mode was used for surface wave analysis to illustrate the wave characteristics. Wavenumbers (Kx, Ky) ranging between 0.01–0.1 radm−1 and wave directions predominantly in the northeastward direction were identified from the spectral peak ranges. The Optical Flow (OF) algorithm was applied to the remaining consecutive low-frequency modes as the current signal under 0.1 Hz for surface current analysis and to estimate a current field with a 1 m spatial resolution. The accuracy of currents in the overall region was validated with in situ drifter measurements, showing an R-squared (R2) value of 0.80 and an average root-mean-square error (RMSE) of 0.03 ms−1. This study proposes a novel framework for analyzing individual sea surface dynamical processes acquired from high-resolution UAV imagery using a multidimensional signal decomposition method specialized in nonlinear and nonstationary data analysis.

Drone-based early detection of bark beetle infested spruce trees differs in endemic and epidemic populations

IntroductionEuropean forests face increasing threats due to climate change-induced stressors, which create the perfect conditions for bark beetle outbreaks. The most important spruce forest pest in Europe is the European Spruce Bark Beetle (Ips typographus L.). Effective management of I. typographus outbreaks necessitates the timely detection of recently attacked spruce trees, which is challenging given the difficulty in spotting symptoms on infested tree crowns. Bark beetle population density is one of many factors that can affect infestation rate and symptoms development. This study compares the appearance of early symptoms in endemic and epidemic bark beetle populations using highresolution Unmanned Aerial Vehicles (UAV) multispectral imagery.MethodsIn spring of 2022, host colonization by bark beetles was induced on groups of spruce trees growing in 10 sites in the Southern Alps, characterized by different population density (5 epidemic and 5 endemic). A multispectral sensor mounted on a drone captured images once every 2 weeks, from May to August 2022. The analyses of a set of vegetational indices allowed the actual infested trees’ reflectance features and symptoms appearance to be observed at each site, comparing them with those of unattacked trees.ResultsResults show that high bark beetles population density triggers a more rapid and intense response regarding the emergence of symptoms. Infested trees were detected at least 1 month before symptoms became evident to the human eye (red phase) in epidemic sites, while this was not possible in endemic sites. Key performing vegetation indices included NDVI (Normalized Difference Vegetation Index), SAVI (Soil Adjust Vegetation Index, with a correction factor of 0.44), and NDRE (Normalized Difference Red Edge index).DiscussionThis early-detection approach could allow automatic diagnosis of bark beetles’ infestations and provide useful guidance for the management of areas suffering pest outbreaks.

Efficient Structure from Motion for Large-Size Videos from an Open Outdoor UAV Dataset.

Modern UAVs (unmanned aerial vehicles) equipped with video cameras can provide large-scale high-resolution video data. This poses significant challenges for structure from motion (SfM) and simultaneous localization and mapping (SLAM) algorithms, as most of them are developed for relatively small-scale and low-resolution scenes. In this paper, we present a video-based SfM method specifically designed for high-resolution large-size UAV videos. Despite the wide range of applications for SfM, performing mainstream SfM methods on such videos poses challenges due to their high computational cost. Our method consists of three main steps. Firstly, we employ a visual SLAM (VSLAM) system to efficiently extract keyframes, keypoints, initial camera poses, and sparse structures from downsampled videos. Next, we propose a novel two-step keypoint adjustment method. Instead of matching new points in the original videos, our method effectively and efficiently adjusts the existing keypoints at the original scale. Finally, we refine the poses and structures using a rotation-averaging constrained global bundle adjustment (BA) technique, incorporating the adjusted keypoints. To enrich the resources available for SLAM or SfM studies, we provide a large-size (3840 × 2160) outdoor video dataset with millimeter-level-accuracy ground control points, which supplements the current relatively low-resolution video datasets. Experiments demonstrate that, compared with other SLAM or SfM methods, our method achieves an average efficiency improvement of 100% on our collected dataset and 45% on the EuRoc dataset. Our method also demonstrates superior localization accuracy when compared with state-of-the-art SLAM or SfM methods.

Spatial mapping of soil moisture content using very-high resolution UAV-based multispectral image analytics

Assessing soil moisture content (SMC) is necessary for managing water at a spatial scale. Remote sensing technologies provide a robust approach for detecting the spatial-temporal fluctuations of SMC. The aim of this study was to estimate SMC at different soil depths using very high-resolution unmanned aerial vehicle (UAV)-based multispectral (MS) images and machine learning algorithms and generate spatial maps of SMC using the best-performed machine learning (ML) algorithm. The UAV-based multispectral images of bare soil were captured at 40 m altitude with a very high spatial resolution (2.89 cm) during the rabi 2021/22 season. At the same time, the soil samples were collected from different soil depths, and the gravimetric SMC was measured. Five machine-learning algorithms (Linear Regression (LR), K-Nearest Neighbors (KNN), Random Forest (RF), Decision Tree (DT), and Support Vector Regression (SVR)) were used to train the model between SMC and MS data (MS band reflectance and vegetation indices). The soil with high SMC has low spectral reflectance and soil with low SMC shows high spectral reflectance. For the prediction of surface SMC, the linear regression (R2 = 0.89; RMSE = 2.80 %) and 5 cm depth SMC, the SVR (R2 = 0.64; RMSE = 3.03 %) were performed well compared to other ML algorithms. For surface SMC, blue band reflectance, and 5 cm depth SMC, the Ratio Vegetation Index (RVI) correlated well compared to others. All models failed to predict the SMC at the deeper soil depths. The spatial SMC mapping described the visual color variations in SMC within the field. Crop irrigation scheduling can be significantly improved through the insights this spatial SMC estimation approach provides, making it a valuable tool for farmers and irrigation planners.

Application of an improved U-Net with image-to-image translation and transfer learning in peach orchard segmentation

Peach cultivation holds a significant economic importance, and obtaining the spatial distribution of peach orchards is helpful for yield prediction and precision agriculture. In this study, we introduce a new U-Net semantic segmentation model, utilizing ResNet50 as a backbone network, augmented with an Efficient Multi-Scale Attention (EMA) mechanism module and a LayerScale adaptive scaling parameter. To address style differences between images from Unmanned Aerial Vehicle (UAV), Google Earth, and Sentinel-2 satellite, we incorporate Cycle-Consistent Generative Adversarial Networks (CycleGAN). This synthesis ensures that UAV images conform to a comparable style found in Google Earth and Sentinel-2 images, while feature details of high spatial resolution UAV images are transferred to Google Earth and Sentinel-2 images through transfer learning. The results demonstrate that using ResNet50 as a backbone network for the U-Net model yields higher accuracy compared to using VGG16 for the U-Net model. Specifically, the Mean Intersection over Union (MIoU) values for UAV and Sentinel-2 images are higher by 0.49 % and 0.95 %, respectively. The MIoU values for UAV, Google Earth, and Sentinel-2 images increased by 0.87 %, 1.71 %, and 1.74 %, respectively, with the introduction of EMA. Additionally, with the introduction of LayerScale adaptive scaling parameters, the MIoU values increased by 0.31 %, 0.33 %, and 1.44 %, respectively, further enhancing the segmentation accuracy of the model. After applying CycleGAN and transfer learning, the MIoU increased by 1.02 %, 0.15 %, and 1.57 % for UAV, Google Earth, and Sentinel-2 images, respectively, resulting in MIoU values of 97.39 %, 92.08 %, and 84.54 %. The comparative analysis with DeepLabV3+, PSPNet, and HRNet models demonstrates the superior mapping performance of the proposed method. Moreover, the method exhibits good generalization and mapping speed across six test sites in the research area. Overall, this approach ensures high precision and efficiency in peach orchard mapping, accommodating various spatial resolutions, and holds potential for addressing diverse requirements in peach orchard mapping applications.

Monitoring the spatial–temporal distribution of invasive plant in urban water using deep learning and remote sensing technology

Despite water ecosystems being capable of sustaining biodiversity and enhancing the overall resilience of the urban environment, they are highly susceptible to biological invasions. Invasive aquatic plants (IAPs) threaten the natural environment by reducing the diversity of native aquatic plants and animal communities. Detecting IAPs and mapping their distribution is crucial for the protection of urban water ecosystems. This study is the first of its kind to use high-resolution unmanned aerial vehicle (UAV) imagery and deep learning approaches to monitor the expansion of Pistia stratiotes (water lettuce). A DJI Matrice 300 RTK was utilized to capture time-series images with a resolution of 0.018 m on the Guanyin Lake, Cangshan District, Fuzhzou, China, during an outbreak and subsequent rapid growth stage of water lettuce. Three deep learning model architectures and three backbones were combined to detect water lettuce. Model performance and the ability to generalize the model were evaluated to determine the optimal model for water lettuce detection from time-series high-resolution UAV imagery. Results show that the DeepLabv3 + model with ResNet-34 achieved superior performance in detecting water lettuce from time-series imagery, yielding an average accuracy of 90.24 % (85.33 %≤F1_score ≤ 96.54 %) for water lettuce detection on five different dates. For the UAV image acquired on September 26th, the U-Net model with ResNet-18 yielded the highest accuracy (F1_score = 92.46 %), but it was not the optimal model for multi-temporal water lettuce detection on subsequent dates. The distribution of water lettuce can have large variations at different times, with an average change rate of 49.50 % every two days and the highest change rate up to 60.33 %. The study demonstrates that the combination of UAV imagery and a deep learning model can achieve excellent accuracy for water lettuce monitoring and provide a method to map IAPs in dynamic urban water systems over time.

In-Season Cotton Yield Prediction with Scale-Aware Convolutional Neural Network Models and Unmanned Aerial Vehicle RGB Imagery.

In the pursuit of sustainable agriculture, efficient water management remains crucial, with growers relying on advanced techniques for informed decision-making. Cotton yield prediction, a critical aspect of agricultural planning, benefits from cutting-edge technologies. However, traditional methods often struggle to capture the nuanced complexities of crop health and growth. This study introduces a novel approach to cotton yield prediction, leveraging the synergy between Unmanned Aerial Vehicles (UAVs) and scale-aware convolutional neural networks (CNNs). The proposed model seeks to harness the spatiotemporal dynamics inherent in high-resolution UAV imagery to improve the accuracy of the cotton yield prediction. The CNN component adeptly extracts spatial features from UAV-derived imagery, capturing intricate details related to crop health and growth, modeling temporal dependencies, and facilitating the recognition of trends and patterns over time. Research experiments were carried out in a cotton field at the USDA-ARS Cropping Systems Research Laboratory (CSRL) in Lubbock, Texas, with three replications evaluating four irrigation treatments (rainfed, full irrigation, percent deficit of full irrigation, and time delay of full irrigation) on cotton yield. The prediction revealed that the proposed CNN regression models outperformed conventional CNN models, such as AlexNet, CNN-3D, CNN-LSTM, ResNet. The proposed CNN model showed state-of-art performance at different image scales, with the R2 exceeding 0.9. At the cotton row level, the mean absolute error (MAE) and mean absolute percentage error (MAPE) were 3.08 pounds per row and 7.76%, respectively. At the cotton grid level, the MAE and MAPE were 0.05 pounds and 10%, respectively. This shows the proposed model's adaptability to the dynamic interplay between spatial and temporal factors that affect cotton yield. The authors conclude that integrating UAV-derived imagery and CNN regression models is a potent strategy for advancing precision agriculture, providing growers with a powerful tool to optimize cultivation practices and enhance overall cotton productivity.

Inversion of winter wheat leaf area index from UAV multispectral images: classical vs. deep learning approaches.

Precise and timely leaf area index (LAI) estimation for winter wheat is crucial for precision agriculture. The emergence of high-resolution unmanned aerial vehicle (UAV) data and machine learning techniques offers a revolutionary approach for fine-scale estimation of wheat LAI at the low cost. While machine learning has proven valuable for LAI estimation, there are still model limitations and variations that impede accurate and efficient LAI inversion. This study explores the potential of classical machine learning models and deep learning model for estimating winter wheat LAI using multispectral images acquired by drones. Initially, the texture features and vegetation indices served as inputs for the partial least squares regression (PLSR) model and random forest (RF) model. Then, the ground-measured LAI data were combined to invert winter wheat LAI. In contrast, this study also employed a convolutional neural network (CNN) model that solely utilizes the cropped original image for LAI estimation. The results show that vegetation indices outperform the texture features in terms of correlation analysis with LAI and estimation accuracy. However, the highest accuracy is achieved by combining both vegetation indices and texture features to invert LAI in both conventional machine learning methods. Among the three models, the CNN approach yielded the highest LAI estimation accuracy (R 2 = 0.83), followed by the RF model (R 2 = 0.82), with the PLSR model exhibited the lowest accuracy (R 2 = 0.78). The spatial distribution and values of the estimated results for the RF and CNN models are similar, whereas the PLSR model differs significantly from the first two models. This study achieves rapid and accurate winter wheat LAI estimation using classical machine learning and deep learning methods. The findings can serve as a reference for real-time wheat growth monitoring and field management practices.

Mapping invasive noxious weed species in the alpine grassland ecosystems using very high spatial resolution UAV hyperspectral imagery and a novel deep learning model

ABSTRACT The term “invasive noxious weed species” (INWS), which refers to noxious weed plants that invade native alpine grasslands, has increasingly become an ecological and economic threat in the alpine grassland ecosystem of the Qinghai-Tibetan Plateau (QTP). Both the INWS and native grass species are small in physical size and share a habitat. Using remote sensing data to distinguish INWS from native alpine grass species remains a challenge. High spatial resolution hyperspectral imagery provides an alternative for addressing this problem. Here, we explored the use of unmanned aerial vehicle (UAV) hyperspectral imagery and deep learning methods with a small sample size for mapping the INWS in mixed alpine grasslands. To assess the method, UAV hyperspectral data with a very high spatial resolution of 2 cm were collected from the study site, and a novel convolutional neural network (CNN) model called 3D&2D-INWS-CNN was developed to take full advantage of the rich information provided by the imagery. The results indicate that the proposed 3D&2D-INWS-CNN model applied to the collected imagery for mapping INWS and native species with small ground truth training samples is robust and sufficient, with an overall classification accuracy exceeding 95% and a kappa value of 98.67%. The F1 score for each native species and INWS ranged from 92% to 99%. In conclusion, our results highlight the potential of using very high spatial resolution UAV hyperspectral data combined with a state-of-the-art deep learning model for INWS mapping even with small training samples in degraded alpine grassland ecosystems. Studies such as ours can aid the development of invasive species management practices and provide more data for decision-making in controlling the spread of invasive species in similar grassland ecosystems or, more widely, in terrestrial ecosystems.

Surface Heat Monitoring with High-Resolution UAV Thermal Imaging: Assessing Accuracy and Applications in Urban Environments

The urban heat island (UHI) effect, where urban areas experience higher temperatures than surrounding rural regions, necessitates effective monitoring to estimate and address its diverse impacts. Many existing studies on urban heat dynamics rely on satellite data with coarse resolutions, posing challenges in analyzing heterogeneous urban surfaces. Unmanned aerial vehicles (UAVs) offer a solution by providing thermal imagery at a resolution finer than 1 m. Despite UAV thermal imaging being extensively explored in agriculture, its application in urban environments, specifically for surface temperatures, remains underexplored. A pilot project conducted in Athens, Georgia, utilized a UAV with a FLIR Vue Pro R 640 thermal camera to collect thermal data from two neighborhoods. Ground data, obtained using a handheld FLIR E6-XT infrared imaging camera, were compared with UAV thermal imagery. The study aimed to assess the accuracy of the UAV camera and the handheld camera for urban monitoring. Initial testing revealed the handheld’s accuracy but tendency to underpredict, while UAV camera testing highlighted considerations for altitude in both the rjpg and tiff image pixel conversion models. Despite challenges, the study demonstrates the potential of UAV-derived thermal data for monitoring urban surface temperatures, emphasizing the need for careful model considerations in data interpretation.

Evapotranspiration partitioning by integrating eddy covariance, micro-lysimeter and unmanned aerial vehicle observations: A case study in the North China Plain

Partitioning of evapotranspiration (ET) into soil evaporation (E) and plant transpiration (T) poses a significant challenge. This study proposed a novel approach that combines eddy covariance, micro-lysimeter, and high-resolution unmanned aerial vehicle (UAV) images within the flux footprint to optimize the canopy and soil resistances of the Shuttleworth-Wallace model (S-W) using a multi-objective optimization scheme. A two-year experiment was conducted at the Yucheng and Luancheng sites to investigate the effectiveness of the proposed method for the summer maize-wheat winter system in the North China Plain. Our results showed that the S-W model with multi-objective optimization by using high-resolution UAV images within the flux footprint significantly improved the estimation accuracy of ET components compared with the traditional one- and two-objective optimization schemes. The estimation of ET, T, and E of our method had an average root mean square error of 0.67 mm day−1, 0.66 mm day−1, and 0.28 mm day−1, respectively. The optimized S-W model using high-resolution UAV data within the flux footprint produced better ET partitioning than that optimized using Moderate Resolution Imaging Spectroradiometer (MODIS) data. We then applied the proposed method to estimate ET and ET components in the North China Plain. This study highlighted the importance of utilizing multi-source data (especially high-resolution UAV images) and multi-objective optimization for accurate ET partitioning.

Effect of high-resolution satellite and UAV imagery plot pixel resolution in wheat crop yield prediction

ABSTRACT Accurate crop performance assessment and yield prediction in plant breeding programmes can aid decision-making to improve productivity and product quality during crop selection and management. Grain yield is a complex trait, which is a function of the genotype-environment interaction. While using digital remote sensing traits to assess crop performance and predict yield, the characteristics of the sensing tools and approaches can influence prediction performance. In this study, two sensing scales, an unmanned aerial vehicle (UAV) equipped with a ten-band multispectral camera and high-resolution (~0.31 m) WorldView-3 satellite imagery, were used to monitor spring and winter wheat breeding trails in two growing seasons (2020 and 2021). The breeding plots were planted in three different plot sizes (about 1.5 × 5.0 m, 3.0 × 11.0 m, and 4.5 × 11.0 m in spring wheat, and about 1.5 × 3.0 m, 3.0 × 7.3 m, and 4.5 × 7.3 m in winter wheat), with each having 12 varieties and three replications per variety. The spectral and vegetation indices (VI) were extracted from the datasets, and machine learning models for yield prediction (partial least squares regression, least absolute shrinkage selector operator regression, and random forest regression) were evaluated. With multiscale approaches, a moderate to strong correlation of VI data between high-resolution satellite and UAV data (0.42 ≤ r ≤ 0.99, p < 0.01) was found in most cases. The yield prediction accuracies using the extracted data from the high-resolution satellite (6.26 ≤ RMSE% ≤ 25.49; 5.11 ≤ MAE% ≤ 20.95; 0.17 ≤ r ≤ 0.78) and UAV imagery (5.53 ≤ RMSE% ≤ 17.20; 4.28 ≤ MAE% ≤ 14.20; 0.43 ≤ r ≤ 0.92) were also comparable. These findings inform the applications of high-resolution satellite imagery in breeding programmes, considering that the plot size would influence yield prediction accuracies.

Mapping the Distribution of High-Value Broadleaf Tree Crowns through Unmanned Aerial Vehicle Image Analysis Using Deep Learning

High-value timber species with economic and ecological importance are usually distributed at very low densities, such that accurate knowledge of the location of these trees within a forest is critical for forest management practices. Recent technological developments integrating unmanned aerial vehicle (UAV) imagery and deep learning provide an efficient method for mapping forest attributes. In this study, we explored the applicability of high-resolution UAV imagery and a deep learning algorithm to predict the distribution of high-value deciduous broadleaf tree crowns of Japanese oak (Quercus crispula) in an uneven-aged mixed forest in Hokkaido, northern Japan. UAV images were collected in September and October 2022 before and after the color change of the leaves of Japanese oak to identify the optimal timing of UAV image collection. RGB information extracted from the UAV images was analyzed using a ResU-Net model (U-Net model with a Residual Network 101 (ResNet101), pre-trained on large ImageNet datasets, as backbone). Our results, confirmed using validation data, showed that reliable F1 scores (&gt;0.80) could be obtained with both UAV datasets. According to the overlay analyses of the segmentation results and all the annotated ground truth data, the best performance was that of the model with the October UAV dataset (F1 score of 0.95). Our case study highlights a potential methodology to offer a transferable approach to the management of high-value timber species in other regions.

Water Quality Inversion of a Typical Rural Small River in Southeastern China Based on UAV Multispectral Imagery: A Comparison of Multiple Machine Learning Algorithms

Remote sensing technology applications for water quality inversion in large rivers are common. However, their application to medium/small-sized water bodies within rural areas is limited due to the low spatial resolution of remote sensing images. In this work, a typical small rural river was selected, and high-resolution unmanned aerial vehicle (UAV) multispectral images and ground monitoring data of the river were obtained. Then, a comparative analysis of three univariate regression models and nine machine learning models (Ridge Regression (RR), Support Vector Regression (SVR), Grid Search Support Vector Regression (GS-SVR), Random Forest (RF), Grid Search Random Forest (GS-RF), eXtreme Gradient Boosting (XGBoost), Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), and Catboost Regression (CBR)) for their accuracy in the prediction of turbidity (TUB), total nitrogen (TN), and total phosphorus (TP) was performed. TUB can be achieved by simple statistical regression models. The CBR model exhibited the best performance for the three index inversions on the test set evaluation metrics: R2 (0.90~0.92), RMSE (7.57 × 10−3~1.59 mg/L), MAE (0.01~1.30 mg/L), RPD (3.21~3.56), and NSE (0.84~0.92). The water pollution of the study area was closely related to its land-use pattern, excessive and irrational fertilizer application, and distribution of pollutant outlets.

Review of Deep Learning Algorithms for Urban Remote Sensing Using Unmanned Aerial Vehicles (UAVs)

Abstract: This study conducts a comprehensive review of Deep Learning-based approaches for accurate object segmentation and detection in high-resolution imagery captured by Unmanned Aerial Vehicles (UAVs). The methodology employs three different existing algorithms tailored to detect roads, buildings, trees, and water bodies. These algorithms include Res-UNet for roads and buildings, DeepForest for trees, and WaterDetect for water bodies. To evaluate the effectiveness of this approach, the performance of each algorithm is compared with state-of-the-art (SOTA) models for each class. The results of the study demonstrate that the methodology outperforms SOTA models in all three classes, achieving an accuracy of 93% for roads and buildings using Res-U-Net, 95% for trees using DeepForest, and an impressive 98% for water bodies using WaterDetect. The paper utilizes a Deep Learning-based approach for accurate object segmentation and detection in high-resolution UAV imagery, achieving superior performance to SOTA models, with reduced overfitting and faster training by employing three smaller models for each task

Utilizing a single-temporal full polarimetric Gaofen-3 SAR image to map coseismic landslide inventory following the 2017 Mw 7.0 Jiuzhaigou earthquake (China)

On August 8, 2017, a magnitude 7.0 earthquake struck Jiuzhaigou County in Sichuan Province, triggering numerous coseismic landslides. The prompt identification of these landslides is imperative for emergency rescue efforts and post-earthquake hazard assessments. Optical satellite and unmanned aerial vehicle (UAV) images are often obstructed by cloud cover and fog following earthquakes. In contrast, polarimetric synthetic aperture radar (PolSAR), unaffected by adverse weather conditions, emerges as an indispensable tool. However, the utilization of spaceborne single-temporal SAR for mapping the inventory of coseismic landslides is infrequent and encounters constraints due to several limitations. In this study, we analyzed the amplitude feature and polarimetric decomposition of multiple ground categories in a full PolSAR image, and proposed an automated method to accurately identify coseismic landslides using a single-temporal full PolSAR image. The coseismic landslide inventory following the Jiuzhaigou Earthquake was mapped and validated using high-resolution UAV images. Detailed analysis was conducted to identify error sources leading to omissions and false positives. Additionally, we evaluated various machine learning models to compare their performance with our proposed method. Finally, we conducted a comprehensive discussion on the strengths and weaknesses of different data types (PolSAR, optical satellite, and UAV) for coseismic landslide identification. Our results indicate that the proposed PolSAR-based method achieves high accuracy in coseismic landslide inventory mapping, offering an effective solution for timely post-earthquake emergency responses in complex environments and all weather conditions in the future.