Unmanned Aerial Vehicle Imagery Research Articles

Estimation of Maize Foliar Temperature and Stomatal Conductance as Indicators of Water Stress Based on Optical and Thermal Imagery Acquired Using an Unmanned Aerial Vehicle (UAV) Platform

Climatic variability and extreme weather events impact agricultural production, especially in sub-Saharan smallholder cropping systems, which are commonly rainfed. Hence, the development of early warning systems regarding moisture availability can facilitate planning, mitigate losses and optimise yields through moisture augmentation. Precision agricultural practices, facilitated by unmanned aerial vehicles (UAVs) with very high-resolution cameras, are useful for monitoring farm-scale dynamics at near-real-time and have become an important agricultural management tool. Considering these developments, we evaluated the utility of optical and thermal infrared UAV imagery, in combination with a random forest machine-learning algorithm, to estimate the maize foliar temperature and stomatal conductance as indicators of potential crop water stress and moisture content over the entire phenological cycle. The results illustrated that the thermal infrared waveband was the most influential variable during vegetative growth stages, whereas the red-edge and near-infrared derived vegetation indices were fundamental during the reproductive growth stages for both temperature and stomatal conductance. The results also suggested mild water stress during vegetative growth stages and after a hailstorm during the mid-reproductive stage. Furthermore, the random forest model optimally estimated the maize crop temperature and stomatal conductance over the various phenological stages. Specifically, maize foliar temperature was best predicted during the mid-vegetative growth stage and stomatal conductance was best predicted during the early reproductive growth stage. Resultant maps of the modelled maize growth stages captured the spatial heterogeneity of maize foliar temperature and stomatal conductance within the maize field. Overall, the findings of the study demonstrated that the use of UAV optical and thermal imagery, in concert with prediction-based machine learning, is a useful tool, available to smallholder farmers to help them make informed management decisions that include the optimal implementation of irrigation schedules.

Reliable Label-Supervised Pixel Attention Mechanism for Weakly Supervised Building Segmentation in UAV Imagery

Building segmentation for Unmanned Aerial Vehicle (UAV) imagery usually requires pixel-level labels, which are time-consuming and expensive to collect. Weakly supervised semantic segmentation methods for image-level labeling have recently achieved promising performance in natural scenes, but there have been few studies on UAV remote sensing imagery. In this paper, we propose a reliable label-supervised pixel attention mechanism for building segmentation in UAV imagery. Our method is based on the class activation map. However, classification networks tend to capture discriminative parts of the object and are insensitive to over-activation; therefore, class activation maps cannot directly guide segmentation network training. To overcome these challenges, we first design a Pixel Attention Module that captures rich contextual relationships, which can further mine more discriminative regions, in order to obtain a modified class activation map. Then, we use the initial seeds generated by the classification network to synthesize reliable labels. Finally, we design a reliable label loss, which is defined as the sum of the pixel-level differences between the reliable labels and the modified class activation map. Notably, the reliable label loss can handle over-activation. The preceding steps can significantly improve the quality of the pseudo-labels. Experiments on our home-made UAV data set indicate that our method can achieve 88.8% mIoU on the test set, outperforming previous state-of-the-art weakly supervised methods.

Olive trees cases classification based on deep convolutional neural network from unmanned aerial vehicle imagery

Unmanned aerial vehicles (UAVs) are one of the various aerial remote sensing platforms with ease of use and cost-effectiveness it can deliver high-resolution imaging, obtained using a variety of sensors. Photogrammetric data is derived by the use of unmanned aerial systems (UAS, which consists of a UAV, sensor(s), and base station). As a result of these types, vegetation monitoring is conceivable. Deep neural networks have had a lot of success with image classification tasks, especially in the remote sensing field. In this paper, we demonstrate how deep neural networks can be used to classify olive trees status from aerial images. We have addressed a multi-class classification problem. In this work five different neural network architectures: VGG16, ResNet50, MobileNet, Xception, and VGG19 had been compared. Transfer learning had been accomplished using training of the fully connected layer(s) at the end of the deep learning layers. We used metrics such as accuracy, precision, recall, and confusion metric to evaluate the results. With accuracy, our model achieves the best results using ResNet50 with an accuracy is (97.2%).

An assessment of background removal approaches for improved estimation of rice leaf nitrogen concentration with unmanned aerial vehicle multispectral imagery at various observation times

Background effect is a crucial limitation for the monitoring of leaf nitrogen concentration (LNC) in crops with unmanned aerial vehicle (UAV) multispectral imagery. Some background removal approaches have been developed for improve the estimation of LNC, but their performances are not compared in one study and it is unclear whether they are sensitive to the observation time of UAV imagery. This study evaluated three background removal approaches, i.e., the soil-adjusted vegetation index (SAVI) approach, the green pixel vegetation index approach (GPVI) and abundance adjusted vegetation index (AAVI), for estimating rice LNC from UAV-based multispectral imagery at individual and across growth stages as well as different observation times of the day. The red edge chlorophyll index (CIre) was chosen as the common basis for the last two approaches. In particular, the AAVI approach was refined with a higher number of endmembers and automated endmember extraction, and further evaluated for assessing the effect of separating sunlit components from shaded components of the canopy. Our results demonstrated that the vegetation indices (VIs) for off-noon observation times showed better relationships with LNC than those for noon at individual and across growth stages. Compared to both SAVI and CIre-green, the AACIre for all pixels (AACIre-all) exhibited the weakest sensitivity to observation time and yielded the best relationships for single-stage (jointing: r2=0.70, booting: r2=0.76, heading: r2=0.70) and across-stage (r2=0.66) models. Among the AAVIs derived from three categories of pixels, the AACIre-sunlit (R2 =0.90, RMSE=0.17%, Bias=0.03%) outperformed AACIre-all (R2 =0.85, RMSE=0.23%, Bias=0.08%) and then AACIre-shaded (R2 =0.38, RMSE=0.49%, Bias=0.40%) remarkably for the estimation accuracy of LNC. This study suggests that the refined AAVI approach has great value in reducing the background effect for more accurate monitoring of growth parameters and could be extended to other crops and regions for improved precision crop management and field-based high-throughput phenotyping.

Comparison of Deep Learning Methods for Detecting and Counting Sorghum Heads in UAV Imagery

With the rapid development of remote sensing with small, lightweight unmanned aerial vehicles (UAV), efficient and accurate crop spike counting, and yield estimation methods based on deep learning (DL) methods have begun to emerge, greatly reducing labor costs and enabling fast and accurate counting of sorghum spikes. However, there has not been a systematic, comprehensive evaluation of their applicability in cereal crop spike identification in UAV images, especially in sorghum head counting. To this end, this paper conducts a comparative study of the performance of three common DL algorithms, EfficientDet, Single Shot MultiBox Detector (SSD), and You Only Look Once (YOLOv4), for sorghum head detection based on lightweight UAV remote sensing data. The paper explores the effects of overlap ratio, confidence, and intersection over union (IoU) parameters, using the evaluation metrics of precision P, recall R, average precision AP, F1 score, computational efficiency, and the number of detected positive/negative samples (Objects detected consistent/inconsistent with real samples). The experiment results show the following. (1) The detection results of the three methods under dense coverage conditions were better than those under medium and sparse conditions. YOLOv4 had the most accurate detection under different coverage conditions; on the contrary, EfficientDet was the worst. While SSD obtained better detection results under dense conditions, the number of over-detections was larger. (2) It was concluded that although EfficientDet had a good positive sample detection rate, it detected the fewest samples, had the smallest R and F1, and its actual precision was poor, while its training time, although medium, had the lowest detection efficiency, and the detection time per image was 2.82-times that of SSD. SSD had medium values for P, AP, and the number of detected samples, but had the highest training and detection efficiency. YOLOv4 detected the largest number of positive samples, and its values for R, AP, and F1 were the highest among the three methods. Although the training time was the slowest, the detection efficiency was better than EfficientDet. (3) With an increase in the overlap ratios, both positive and negative samples tended to increase, and when the threshold value was 0.3, all three methods had better detection results. With an increase in the confidence value, the number of positive and negative samples significantly decreased, and when the threshold value was 0.3, it balanced the numbers for sample detection and detection accuracy. An increase in IoU was accompanied by a gradual decrease in the number of positive samples and a gradual increase in the number of negative samples. When the threshold value was 0.3, better detection was achieved. The research findings can provide a methodological basis for accurately detecting and counting sorghum heads using UAV.

Estimation of River Discharge Using Unmanned Aerial Vehicle (UAV) Based on Manning Formula for an Ungauged Alpine River in the Eastern Qilian Mountains

River discharge is crucial to water resources development and ecological protection. However, in some arid areas of northwest China, it is still difficult to measure discharge accurately. In this study, unmanned aerial vehicle (UAV) imagery has been used to estimate river discharge at two river sections in the upper reaches of the Shiyang River in the eastern part of the Qilian Mountains based on the Manning formula. The estimated discharges at those two sections are 1.16 m3/s and 3.11 m3/s, respectively. Taking the discharges measured by an acoustic Doppler current profiler (ADCP) as the reference, the relative error of the estimates is below 5%, which is accurate enough for water resources management in mountain basin regions. Multiple high-resolution satellite images were also used to calculate water discharges at the two sections, which were in good agreement with the discharges estimated from UAVs. This study demonstrates the feasibility of using UAVs to estimate river discharge, which is of great significance for future regional-scale water resource assessments.

Using an unmanned aerial vehicle (UAV) for lake management: ecological status, lake regime shift and stratification processes in a small Mediterranean karstic lake

High-resolution remote sensing imagery by unmanned aerial vehicles (UAVs) has been used as a tool for the environmental management of natural resources. Monitoring programmes that evaluate the ecological status of water bodies according to the Water Framework Directive (WFD) involve significant costs and sampling efforts that can be reduced by using UAVs. UAV imagery was used to measure some metrics of the “macrophytes and phytobenthos” biological quality element (BQE), which is required to assess the ecological status of European lakes; e.g. the percentage cover of hydrophytes and helophytes. Eight UAV flights took place during an annual cycle (July 2016 to July 2017) in a small karstic lake located in southeast Spain. Limnolog¬ical surveys of physicochemical (temperature, conductivity, dissolved oxygen, pH) and biological (pigments) parameters were simultaneously performed to correctly interpret the UAV images. For each flight performance, an orthomosaic of georeferenced RGB images was obtained, and the different features of interest were monitored and quantified by an automated identification and classification system (the LAIC software). The UAV images allowed us to not only evaluate the lake’s ecological status by measuring macrophyte metrics, but to also detect relevant ecological events for environmental management. A gradual burial process of charophyte meadows by the proliferation of periphytic cyanobacterial was detected in an early state by UAV images. Stratification processes, such as hypolimnetic sulphur bacteria blooms or metalimnetic white colloidal layers, were also ob¬served by UAV imaging. We conclude that UAV imagery is a useful tool for environmental lake management.

Using Synthetic Images to Evaluate and Improve Object Detection Neural Network Performance on Aerial Image Datasets

Machine learning-based models for object detection heavily rely on large datasets of labeled images. When models trained on these datasets are applied to Unmanned Aerial Vehicle (UAV) imagery, the problem arises that the conditions under which the training images were created (lighting, altitude, angle) may be different to the UAVs applied conditions, leading to misclassifications. This problem becomes even more pressing in safety critical applications where failures can have huge negative impacts and constitute obstacles for certification of cognitive UAV components. Along a case study on car detection in low-altitude aerial imagery, we show that using, both, artificial and real images for model training has a positive effect on the performance of object detection algorithms when the trained model is applied on images from another domain. Additionally, we show that weak points from object detection neural networks trained on real-world images transfer to synthetic images and that synthetic data can be used to evaluate neural networks trained on real-world data. Since simulated images are easy to create and object labels are inherently given, the presented approaches show a promising direction for scenarios where adequate datasets are difficult to obtain, as well as for the targeted exploration of weak points of object detection algorithms.

Recent Advances in Forest Insect Pests and Diseases Monitoring Using UAV-Based Data: A Systematic Review

Unmanned aerial vehicles (UAVs) are platforms that have been increasingly used over the last decade to collect data for forest insect pest and disease (FIPD) monitoring. These machines provide flexibility, cost efficiency, and a high temporal and spatial resolution of remotely sensed data. The purpose of this review is to summarize recent contributions and to identify knowledge gaps in UAV remote sensing for FIPD monitoring. A systematic review was performed using the preferred reporting items for systematic reviews and meta-analysis (PRISMA) protocol. We reviewed the full text of 49 studies published between 2015 and 2021. The parameters examined were the taxonomic characteristics, the type of UAV and sensor, data collection and pre-processing, processing and analytical methods, and software used. We found that the number of papers on this topic has increased in recent years, with most being studies located in China and Europe. The main FIPDs studied were pine wilt disease (PWD) and bark beetles (BB) using UAV multirotor architectures. Among the sensor types, multispectral and red–green–blue (RGB) bands were preferred for the monitoring tasks. Regarding the analytical methods, random forest (RF) and deep learning (DL) classifiers were the most frequently applied in UAV imagery processing. This paper discusses the advantages and limitations associated with the use of UAVs and the processing methods for FIPDs, and research gaps and challenges are presented.

Corn planting quality assessment in very high-resolution RGB UAV imagery using Yolov5 and Python

Abstract. Uniform plant spacing along crop rows is a primary concern in maximising yield in precision agriculture, and research has shown that variation in this spacing uniformity has a detrimental effect on productive potential. This irregularity needs to be evaluated as early and efficiently as possible to facilitate effective decision-making. Traditionally, variation in seedling spacing is sampled manually on site, however recent technological developments have made it possible to refine, scale and automate this process. Using machine-learning (ML) object detection techniques, plants can be detected in very high-resolution RGB (redgreen-blue) imagery acquired by an unmanned aerial vehicle (UAV), and after processing and geometric analysis of the results a measurement of the variability in intra-row plant distances can be obtained. This proposed technique is superior to traditional methods since the sampling can be made over more area in less time, and the results are more representative and objective. The main benefits are speed, accuracy and cost reduction. This work aims to demonstrate the feasibility of automatically assessing sowing quality in any number of images, using ML object detection and the Shapely Python library for geometrical analysis. The prototype model can detect 99.35% of corn plants in test data from the same field, but also detects 1.89% false positives. Our geometric analysis algorithm has been shown to be robust in finding planting rows orientation and interplant lines in test cases. The result is a coefficient of variation (CV) calculated per sample image, which can be visualised geographically to support decision-making.

Estimating pasture aboveground biomass under an integrated crop-livestock system based on spectral and texture measures derived from UAV images

Pasture production under integrated crop-livestock systems (ICLS) is a key element to support grazing management decisions. Therefore, further experimentation is required to produce reliable estimates of pasture productivity. An unmanned aerial vehicle (UAV) is a viable tool to obtain fast and accurate aboveground biomass (AGB) estimates in pastures under ICLS. We tested several datasets of variables composed of original spectral bands (RGB-NIR), vegetation indices, and gray-level cooccurrence matrix (GLCM) textures to estimate pasture AGB using the random forest (RF) algorithm and feature selection methods in a commercial ICLS farm in western São Paulo State, Brazil. Field measures of pasture AGB were carried out in three field campaigns over five months to capture the spatiotemporal variability of the pasture fields. Most tested models reached similar results on pasture AGB estimates (R2: 0.60 to 0.70), while the number of variables selected for the RF algorithm differed among models (from 6 to 160 variables). The three more accurate models used few variables, two of which used only texture measures, while the third also employed combined spectral bands and vegetation indices. The best model (R2 = 0.70) used only 10 texture measures. Texture measures that represented images’ inner patterns, calculated from NIR, red-edge, and triangular greenness index data, were the most relevant variables for the main models. The texture contribution indicates important information to be considered to estimate pasture AGB when using UAV imagery. The results achieved by UAV estimates allow the design of multitemporal AGB pasture maps, which may be useful for building more reliable spatiotemporal data.

A Survey Bias Index Based on Unmanned Aerial Vehicle Imagery to Review the Accuracy of Rural Surveys

Field surveys and questionnaires are a cornerstone of rural socioeconomic research, providing invaluable firsthand data regarding on-the-ground situations. However, cost-effective and efficient methods for validating the accuracy of self-reported data in such questionnaires are lacking. Biased data are likely to lead to incorrect conclusions. In this study, we propose a new index, the survey bias index (SBI), for evaluating the degree of survey bias in field surveys. This index was obtained by comparing the data recorded in questionnaires with those from portable unmanned aerial vehicles (UAVs). In a case study, we employed SBI to reveal the degree of survey bias of questionnaires in field surveys on rural homesteads. The SBI of self-reported areas of rural homesteads reached 0.439, implying that 43.9% of data were significantly different from those collected using UAVs. A greater SBI was obtained in the pre-urban zone (0.515) than in the pure rural zone (0.258). These results indicate that homestead areas in the pre-urban zone have more incentive to expand than those in the pure rural zone. UAV remote sensing can strongly support research in the field of social economy, which reveals key information hidden in field surveys and questionnaires.

Multi-sensor and multi-platform consistency and interoperability between UAV, Planet CubeSat, Sentinel-2, and Landsat reflectance data

ABSTRACT Unmanned aerial vehicle (UAV) and satellite data have considerable complementarity for platform inter-operability, data fusion studies, calibration and validation efforts, and various multiscale analyses. To optimize cross-platform synergies between field-deployable UAV and space-based satellite systems, an understanding of spectral characteristics and compatibility is required. Here, we present the assessment of spectral consistency, undertaking a pixel-to-pixel similarity assessment of co-registered reflectance maps using corresponding spectral bands from UAV and satellite multispectral imagery. A high-resolution centimeter-scale UAV-mounted MicaSense RedEdge-MX sensor is intercompared against variable-resolution multi-spectral sensors on-board PlanetScope, Sentinel-2 and Landsat 8 platforms. Sampling from within an urban environment that covers a range of both natural and man-made surfaces, we employ ground-based spectroradiometer data to evaluate pixel-level responses, using regression analysis and measurements of relative root mean square error (rRMSE) to assess for factors such as spatial and spectral misalignment. Using two radiometric correction approaches for the UAV data, we found that a vicarious radiometric correction was more accurate than a linear empirical line method, with the former improving rRMSE by between 1.6% and 20.11% when assessed against spectroradiometer measurements. Spectral band misalignment between the UAV and satellite sensors affected their spectral consistency, causing different reflectance values for the same object in the corresponding UAV and satellite bands, with the issue amplified over specific land-cover classes (e.g. grass in the red edge part of the spectrum). Using the standard deviation of a UAV-derived normalized difference vegetation index (NDVI) as a metric of spatial heterogeneity, larger differences between the UAV and satellite-based NDVI were observed for different ground features in response to both land-cover boundary and shadow effects. Interestingly, higher spatial heterogeneity did not necessarily lead to higher spectral inconsistencies. It was also determined that as spatial scale differences between the UAV and satellite platforms increased, the lower was the impact of geometric misregistration on their consistency. Indeed, the rRMSE between the reflectance values of the UAV-based spectral bands and the corresponding satellite imagery was smaller at lower resolution (e.g. Landsat 8) than higher resolution (e.g. PlanetScope). Overall, the study provides insight into the collective effect of spectral and spatial misalignments on the degree of spectral consistency that can be expected between UAV and satellite data, guiding robust radiometric intercalibration efforts and the potential for improved synergy and interoperability between UAV and satellite data.

Mapping built environments from UAV imagery: a tutorial on mixed methods of deep learning and GIS

Evidence has suggested that built environments are significantly associated with residents’ health and the conditions of built environments vary between neighborhoods. Recently, there have been remarkable technological advancements in using deep learning to detect built environments on fine spatial scale remotely sensed images. However, integrating the extracted built environment information by deep learning with geographic information systems (GIS) is still rare in existing literature. This method paper presents how we harnessed deep leaning techniques to extract built environments and then further utilized the extracted information as input data for analysis and visualization in a GIS environment. Informative guidelines on data collection with an unmanned aerial vehicle (UAV), greenspace extraction using a deep learning model (specifically U-Net for image segmentation), and mapping spatial distributions of greenspace and sidewalks in a GIS environment are offered. The novelty of this paper lies in the integration of deep learning into the GIS decision-making system to identify the spatial distribution of built environments at the neighborhood scale.

Using digital image processing and Unmanned Aerial Vehicle (UAV) for identifying ceramic cladding detachment in building facades

Abstract Facade maintenance actions are driven by results obtained in the inspection phase. Some methodological proposals aimedat optimizing the inspection process have been discussed, notablydigital image processing (DIP) techniques associated with unmanned aerial vehicle (UAV) imagery. Using UAV speeds up the access to the inspected area, and DIP techniques help to automate the identification of pathological manifestations. This article aims to apply DIP techniques to detect areas where the ceramic cladding on building facades is detaching. The methodology referred to herein starts with the creation of a database (images) captured by cell phone and UAV. The object detection algorithm YOLO (You Only Look Once) was applied to the database images. The results indicated these techniques are very promising, with a 94% precision level in the tests performed. The precision index obtained indicates that the model is applicable in practice and discussions about its limitationshelp improve the proposed methodology.

Shadow Removal from UAV Images Based on Color and Texture Equalization Compensation of Local Homogeneous Regions

Due to imaging and lighting directions, shadows are inevitably formed in unmanned aerial vehicle (UAV) images. This causes shadowed regions with missed and occluded information, such as color and texture details. Shadow detection and compensation from remote sensing images is essential for recovering the missed information contained in these images. Current methods are mainly aimed at processing shadows with simple scenes. For UAV remote sensing images with a complex background and multiple shadows, problems inevitably occur, such as color distortion or texture information loss in the shadow compensation result. In this paper, we propose a novel shadow removal algorithm from UAV remote sensing images based on color and texture equalization compensation of local homogeneous regions. Firstly, the UAV imagery is split into blocks by selecting the size of the sliding window. The shadow was enhanced by a new shadow detection index (SDI) and threshold segmentation was applied to obtain the shadow mask. Then, the homogeneous regions are extracted with LiDAR intensity and elevation information. Finally, the information of the non-shadow objects of the homogeneous regions is used to restore the missed information in the shadow objects of the regions. The results revealed that the average overall accuracy of shadow detection is 98.23% and the average F1 score is 95.84%. The average color difference is 1.891, the average shadow standard deviation index is 15.419, and the average gradient similarity is 0.726. The results have shown that the proposed method performs well in both subjective and objective evaluations.

SEX DISCRIMINATION OF INDIVIDUAL TREES USING UAV IMAGERY

Abstract. The sex ratio is the proportion of male to female trees, which has a substantial impact on reproductive success and conservation status. Appropriate sex-related differences in dioecious trees commonly result in leading to a robustly structured population. Fieldwork for sex discrimination is time-consuming and labor-required. Benefiting from the unmanned aerial vehicle (UAV) and SfM techniques, the present study aims to detect male and female Caspian poplar (Populus caspica) trees. In March 2021, a heterogeneous forest in Noor city located in Mazandaran province was photographed, then 3D point clouds were extracted from the images using structure from motion algorithm (SfM) to generate an orthomosaic and a point cloud. The field survey was carried out to record the species, sex, and position of the overstory trees which were identifiable on the orthomosaics. A random forest classification algorithm was applied using R software to classify the trees into male and female. By assessing the producer's accuracy, user's accuracy, and overall accuracy, the classification accuracy for identified trees was computed using 10-fold cross-validation. The results showed an accuracy of 83% for identifying Caspian poplar trees and 52% accuracy for Sex discrimination. Overall, our effort to evaluate sex discrimination of dioecious trees using UAV imagery represents a promising preliminary step in forest data collection.

Sugarcane yields prediction at the row level using a novel cross-validation approach to multi-year multispectral images

Early prediction of sugarcane crop yield would benefit sugarcane growers and policymakers by allowing for timely decisions. The primary objective of this study was to reduce reliance on satellite images and improve early prediction of sugarcane yield at row level by using high-resolution multispectral Unmanned Aerial Vehicle (UAV) imagery. To our knowledge, no previous study has evaluated the performance of multispectral UAV-derived vegetation indices in sugarcane crops at the crop row level. In this study, we used UAV mapping on 48 rows of sugarcane at three main growth stages (early, middle, and mature) over three growing seasons. A secondary objective was to predict future sugarcane yields at the earliest possible stage of growth. The results showed that the optimal growth stage for all 23 VIs varied, but the middle stage, from mid-March to early May, was the most prevalent. Further detailed analysis in the middle stage revealed that March was the best month for predicting future sugarcane yields when compared to April and May. This result is approximately a month earlier than previous studies in the same region. Following two stages of feature selection, such as Pearson correlation analysis and stepwise feature selection, a novel cross-validation methodology based on a generalized linear model trained and tested the yield prediction models on various combinations of the VIs. This novel methodology improves model accuracy by avoiding overfitting and over complexity caused by interdependent VIs, and then validates the model generality using previously unseen data. The best performance was achieved by combining the Normalized Difference RedEdge (NDRE) and the Green–Red Normalized Difference Vegetation Index (GRNDVI) at March. These results help growers and decision-makers benefit from early row-level yield forecast, six months before harvest, if UAV mapping is available.

Deep learning-based individual tree crown delineation in mangrove forests using very-high-resolution satellite imagery

Mangrove forests are vulnerable ecosystems that require broad-scale monitoring. Various solutions based on satellite imagery have emerged for this purpose but still suffer from the lack of methods to accurately delineate individual tree crowns (ITCs). Within-stand variability in crown size and shape, crown clumping and fragmentation, and understory vegetation hamper the delineation in these ecosystems. To cope with these factors, the proposed method combines a deep learning-based enhancement of ITCs with a marker-controlled watershed segmentation algorithm. The MT-EDv3 neural network is employed to compute the normalized Euclidean distance of crown pixels to treetops and a Laplacian of Gaussian filter is applied to the resulting image to enhance crown borders before segmentation. The method was applied to WorldView imagery over four mangrove sites worldwide and compared to previously published methods using standardized metrics. Accurate detection (Overall Accuracy ≥ 0.93 and Kappa ≥ 0.87) and area estimation (R2 ≥ 0.66, NRMSE ≤ 12%) of crowns was achieved for all sites using either the panchromatic band or a combination of the pan-sharpened visible-near-infrared bands. Based on Precision, Recall, and F1-score, the proposed method outperformed previous watershed segmentation and software-based algorithms of crown delineation, as well as the Mask R-CNN segmentation framework. The viewing geometry of images and the forest heterogeneity were identified as important contributors to the delineation accuracy. This study is the first to achieve accurate delineation of ITCs in mangrove forests across sites, opening perspectives of applications to satellite-based monitoring. The method shows promising transferability to other very-high-resolution satellite sensors as well as to aerial and unmanned aerial vehicle imagery and could be improved by including more spectral information and LiDAR-derived canopy height models.

Drone-Sensed and Sap Flux-Derived Leaf Phenology in a Cool Temperate Deciduous Forest: A Tree-Level Comparison of 17 Species

Understanding the relationship between leaf phenology and physiological properties has important implications for improving ecosystem models of biogeochemical cycling. However, previous studies have investigated such relationships only at the ecosystem level, limiting the biological interpretation and application of the observed relationships due to the complex vegetation structure of forest ecosystems. Additionally, studies focusing on transpiration are generally limited compared to those on photosynthesis. Thus, we investigated the relationship between stem sap flux density (SFD) and crown leaf phenology at the individual tree level using the heat dissipation method, unmanned aerial vehicle (UAV)-based observation, and ground-based visual observation across 17 species in a cool temperate forest in Japan, and assessed the potential of UAV-derived phenological metrics to track individual tree-level sap flow phenology. We computed five leaf phenological metrics (four from UAV imagery and one from ground observations) and evaluated the consistency of seasonality between the phenological metrics and SFD using Bayesian modelling. Although seasonal trajectories of the leaf phenological metrics differed markedly among the species, the daytime total SFD (SFDday) estimated by the phenological metrics was significantly correlated with the measured ones across the species, irrespective of the type of metric. Crown leaf cover derived from ground observations (CLCground) showed the highest ability to predict SFDday, suggesting that the seasonality of leaf amount rather than leaf color plays a predominant role in sap flow phenology in this ecosystem. Among the UAV metrics, Hue had a superior ability to predict SFDday compared with the other metrics because it showed seasonality similar to CLCground. However, all leaf phenological metrics showed earlier spring increases than did sap flow in more than half of the individuals. Our study revealed that UAV metrics could be used as predictors of sap flow phenology for deciduous species in cool, temperate forests. However, for a more accurate prediction, phenological metrics representing the spring development of sap flow must be explored.