Search

ABSTRACTAccording to climate statistics, clouds cover more than a third of the Earth’s land surface on average. This cloud coverage obstructs optical satellite imagery, resulting in a loss of essential information concerning the surfaces beneath the clouds. Thus, various studies have been conducted to develop techniques for removing clouds from images. However, most of these studies have relied on non-cloud images as reference data, and radiometric differences have been identified as a factor that degrades the quality and accuracy of restoration images. To address these challenges, we propose the time-series spectral similarity group (TSSG) method for reconstructing cloud regions in satellite imagery using multi-temporal data. Prior to restoring the cloud areas in images, we employ Landsat quality assessment (QA) bands for cloud detection using redefined criteria. Subsequently, the TSSG algorithm uses time-series data to select reference images, overcoming the limitations of previous restoration methods by identifying pixels with locations of similar spectral values in the reference images. Using multi-temporal reference images, the TSSG method extracts pixels based on similar spectral values. This extraction enables the estimation of the true ground surface values by using the corresponding locations of similarity group pixels in cloud images. Simulated cloud images were used to assess the performance of the proposed cloud region restoration method. According to quantitative evaluations, the structural similarity index (SSIM) values representing image quality indicate that the proposed algorithm exhibits a higher restoration accuracy, with an average of 0.89, compared to the previous research having an average of 0.52. To further evaluate the visual impact of our approach, we applied the TSSG method to real cloud images. In addition, we conducted an in-depth analysis of various algorithm parameters to identify the optimal factors that yielded the most effective results. We focused on three key variables in terms of image quality indices: the number of reference images, the proportion of clouds within these references, and the proportion of similarity group pixels within the entire image. This study indicates the potential to enhance the utility of satellite imagery significantly by providing reliable data for time-series analysis, thus expanding the range of image applicability.

This study introduces an automated, open-source workflow for large-scale 2D and 3D building mapping using airborne LiDAR data. Uniquely, our workflow operates entirely unsupervised, eliminating the need for any training procedures. We have integrated a specially tailored digital terrain model generation algorithm into our workflow to prevent errors in complex urban landscapes, especially around highways and overpasses. Through fine rasterization of LiDAR point clouds, we have enhanced building-tree differentiation. In addition, we have reduced errors near water bodies and augmented computational efficiency by introducing a new planarity calculation. Our workflow offers a practical and scalable solution for the mass production of rasterized 2D and 3D building maps from raw airborne LiDAR data. Our method's robustness has been rigorously validated across a diverse dataset in comparison with deep learning-based and hand-digitized products. Through these extensive comparisons, we provide a valuable analysis of building maps generated via different methodologies. We anticipate that our highly scalable building mapping workflow will facilitate the production of reliable 2D and 3D building maps, fostering advances in large-scale urban analysis.

Forests play a critical role in the provision of ecosystem services, and understanding their compositions, especially tree species, is essential for effective ecosystem management and conservation. However, identifying tree species is challenging and time-consuming. Recently, unmanned aerial vehicles (UAVs) equipped with various sensors have emerged as a promising technology for species identification due to their relatively low cost and high spatial and temporal resolutions. Moreover, the advancement of various deep learning models makes remote sensing based species identification more a reality. However, three questions remain to be answered: first, which of the state-of-the-art models performs best for this task; second, which is the optimal season for tree species classification in a temperate forest; and third, whether a model trained in one season can be effectively transferred to another season. To address these questions, we focus on tree species classification by using five state-of-the-art deep learning models on UAV-based RGB images, and we explored the model transferability between seasons. Utilizing UAV images taken in the summer and fall, we captured 8799 crown images of eight species. We trained five models using summer and fall images and compared their performance on the same dataset. All models achieved high performances in species classification, with the best performance on summer images, with an average F1-score was 0.96. For the fall images, Vision Transformer (ViT), EfficientNetB0, and YOLOv5 achieved F1-scores greater than 0.9, outperforming both ResNet18 and DenseNet. On average, across the two seasons, ViT achieved the best accuracy. This study demonstrates the capability of deep learning models in forest inventory, particularly for tree species classification. While the choice of certain models may not significantly affect performance when using summer images, the advanced models prove to be a better choice for fall images. Given the limited transferability from one season to another, further research is required to overcome the challenge associated with transferability across seasons.

It is necessary to extract and recognize the cloud regions presented in imagery to generate satellite imagery as analysis-ready data (ARD). In this manuscript, we proposed a new deep learning model to detect cloud areas in very-high-resolution (VHR) satellite imagery by fusing two deep learning architectures. The proposed UNet3+ model with a hybrid Swin Transformer and EfficientNet (UNet3+STE) was based on the structure of UNet3+, with the encoder sequentially combining EfficientNet based on mobile inverted bottleneck convolution (MBConv) and the Swin Transformer. By sequentially utilizing convolutional neural networks (CNNs) and transformer layers, the proposed algorithm aimed to extract the local and global information of cloud regions effectively. In addition, the decoder used MBConv to restore the spatial information of the feature map extracted by the encoder and adopted the deep supervision strategy of UNet3+ to enhance the model’s performance. The proposed model was trained using the open dataset derived from KOMPSAT-3 and 3A satellite imagery and conducted a comparative evaluation with the state-of-the-art (SOTA) methods on fourteen test datasets at the product level. The experimental results confirmed that the proposed UNet3+STE model outperformed the SOTA methods and demonstrated the most stable precision, recall, and F1 score values with fewer parameters and lower complexity.

Rising food demands require new techniques to achieve higher genetic gains for crop production, especially in regions where climate can negatively affect agriculture. Wheat is a staple crop that often encounters this challenge, and ideotype breeding with optimized canopy traits for grain yield, such as determinate tillering, synchronized flowering, and stay-green (SG), can potentially improve yield under terminal drought conditions. Among these traits, SG has emerged as a key factor for improving grain quality and yield by prolonging photosynthetic activity during reproductive stages. This study aims to highlight the importance of growth dynamics in a wheat mapping population by using multispectral images obtained from uncrewed aerial vehicles as a high-throughput phenotyping technique to assess the effectiveness of using such images for determining correlations between vegetation indices and grain yield, particularly regarding the SG trait. Results show that the determinate group exhibited a positive correlation between NDVI and grain yield, indicating the effectiveness of these traits in yield improvement. In contrast, the indeterminate group, characterized by excessive vegetative growth, showed no significant NDVI–grain yield relationship, suggesting that NDVI values in this group were influenced by sterile tillers rather than contributing to yield. These findings provide valuable insights for crop breeders by offering a non-destructive approach to enhancing genetic gains through the improved selection of resilient wheat genotypes.

Agricultural land parcels (ALPs) are essential for effective agricultural management, influencing activities ranging from crop yield estimation to policy development. However, traditional methods of ALP delineation are often labor-intensive and require frequent updates due to the dynamic nature of agricultural practices. Additionally, the significant variations across different regions and the seasonality of agriculture pose challenges to the automatic generation of accurate and timely ALP labels for extensive areas. This study introduces the cadastral-to-agricultural (Cad2Ag) framework, a novel approach that utilizes cadastral data as training labels to train deep learning models for the delineation of ALPs. Cadastral parcels, which are relatively widely available and stable elements in land management, serve as proxies for ALP delineation. Employing an adapted U-Net model, the framework automates the segmentation process using remote sensing images and geographic information system (GIS) data. This research evaluates the effectiveness of the proposed Cad2Ag framework in two U.S. regions—Indiana and California—characterized by diverse agricultural conditions. Through rigorous evaluation across multiple scenarios, the study explores diverse scenarios to enhance the accuracy and efficiency of ALP delineation. Notably, the framework demonstrates effective ALP delineation across different geographic contexts through transfer learning when supplemented with a small set of clean labels, achieving an F1-score of 0.80 and an Intersection over Union (IoU) of 0.67 using only 200 clean label samples. The Cad2Ag framework’s ability to leverage automatically generated, extensive, free training labels presents a promising solution for efficient ALP delineation, thereby facilitating effective management of agricultural land.

Crop Residue Cover (CRC) is crucial for enhancing soil quality and mitigating erosion in agricultural fields. Accurately estimating CRC in near real-time presents challenges due to the limitations of traditional and remote sensing methods. This study addresses the challenge of accurately estimating CRC using unsupervised algorithms on high-resolution Unmanned Aerial System (UAS) imagery. We employ two methods to perform CRC estimation: (1) K-means unsupervised algorithm and (2) Principal Component Analysis (PCA) along with the Otsu thresholding technique. The advantages of these methods lie in their independence from human intervention for any supervised training stage. Additionally, these methods are rapid and suitable for near real-time estimation of CRC as a decision-making support in agricultural management. Our analysis reveals that the K-means method, with an R2=0.79, achieves superior accuracy in CRC estimation over the PCA-Otsu method with an R2=0.46. The accuracy of CRC estimation for both corn and soybean crops is significantly higher in winter than in spring, attributable to the more weathered state of crop residue. Furthermore, CRC estimations in corn fields exhibit a stronger correlation, likely due to the larger size of corn residue which enhances detectability in images. Nevertheless, the variance in CRC estimation accuracy between corn and soybean fields is minimal. Furthermore, CRC estimation achieves the highest correlation in no-till fields, while the lowest correlation is observed in conventionally tilled fields, a difference likely due to the soil disturbance during plowing in conventional tillage practices.

Abstract The measurement of forest structure has evolved steadily due to advances in technology, methodology, and theory. Such advances have greatly increased our capacity to describe key forest structural elements and resulted in a range of measurement approaches from traditional analog tools such as measurement tapes to highly derived and computationally intensive methods such as advanced remote sensing tools (e.g., lidar, radar). This assortment of measurement approaches results in structural metrics unique to each method, with the caveat that metrics may be biased or constrained by the measurement approach taken. While forest structural diversity (FSD) metrics foster novel research opportunities, understanding how they are measured or derived, limitations of the measurement approach taken, as well as their biological interpretation is crucial for proper application. We review the measurement of forest structure and structural diversity—an umbrella term that includes quantification of the distribution of functional and biotic components of forests. We consider how and where these approaches can be used, the role of technology in measuring structure, how measurement impacts extend beyond research, and current limitations and potential opportunities for future research.

Not all building labels for training improve the performance of the deep learning model. Some labels can be falsely labeled or too ambiguous to represent their ground truths, resulting in poor performance of the model. For example, building labels in OpenStreetMap (OSM) and Microsoft Building Footprints (MBF) are publicly available training sources that have great potential to train deep models, but directly using those labels for training can limit the model's performance as their labels are incomplete and inaccurate, called noisy labels. This article presents self-filtered learning (SFL) that helps a deep model learn well with noisy labels for building extraction in remote sensing images. SFL iteratively filters out noisy labels during the training process based on loss of samples. Through a multiround manner, SFL makes a deep model learn progressively more on refined samples from which the noisy labels have been removed. Extensive experiments with the simulated noisy map as well as real-world noisy maps, OSM and MBF, showed that SFL can improve the deep model's performance in diverse error types and different noise levels.

Cotton boll count is an important phenotypic trait that aids in a better understanding of the genetic and physiological mechanisms of cotton growth. Several computer vision technologies are available for cotton boll segmentation. However, estimating the number of cotton bolls in a segmented cluster of cotton bolls is a challenging task due to the complex shapes of cotton bolls. This study proposed a combination of spectral-spatial and supervised machine learning based methods for cotton boll candidate recognition and counting from high resolution RGB images obtained from unmanned aerial vehicles (UAVs). An algorithm consisting of machine vision, band-mean filter, Otsu thresholding, red/blue band ratio filter, and geometrical characteristics-based error removal techniques, was employed to detect open cotton boll pixels under several environmental settings. In addition, a support vector machine (SVM) based encoding method was developed using geometric features of cotton boll candidates to predict the number of cotton bolls from the segmented cotton boll candidates. This algorithm was implemented over three experiment sites with three cotton varieties, two tillage practices, seven cover crop treatments, two irrigation regimes (irrigated and rainfed), 26 irrigation levels, and two sensors (DJI FC6310 RGB and MicaSense Rededge) capturing images at two spatial resolutions (0.75 cm and 1.07 cm) over two growing seasons (2019 and 2021). These different experimental settings allowed the proposed approaches to be validated against a variety of complex backgrounds. A visual inspection of 1000 randomly selected pixels revealed that the proposed cotton boll candidate recognition approach was highly effective in segmenting cotton bolls and background pixels, with high classification accuracy (> 95%) and a low number of falsely classified pixels (precision > 0.96; recall > 0.93). A high correlation between ground truth observations and predicted cotton boll count indicated that the use of geometric features of segmented candidates as predictors in association with the SVM model demonstrated a good performance in estimating boll count from recognized cotton boll candidates. Furthermore, linear regression analyses revealed that both boll count and candidate area are potential predictors of lint yield, with boll count being a better predictor than candidate area. Overall, the study demonstrated that machine vision/learning techniques can be potentially used on UAV images to count the number of cotton bolls and predict lint yield over large acreages with reasonable accuracy.

Abstract Unmanned aerial systems (UASs) have increased our capacity for collecting finer spatiotemporal resolution data that were previously unobtainable through conventional methods. The use of UAS for obtaining high‐throughput phenotyping (HTP) data in plant breeding programs has gained popularity in recent years. The integrity and quality of the raw data are essential for ensuring the accuracy of predictive tools and proper interpretation of the data. This paper summarizes the standard operation procedures for high‐quality UAS data collection, processing, and analysis for UAS‐based HTP (UAS‐HTP). Plant breeders can follow these procedures to implement a UAS‐HTP system in their germplasm enhancement and cultivar development programs.

IntroductionTar spot is a high-profile disease, causing various degrees of yield losses on corn (Zea mays L.) in several countries throughout the Americas. Disease symptoms usually appear at the lower canopy in corn fields with a history of tar spot infection, making it difficult to monitor the disease with unmanned aircraft systems (UAS) because of occlusion.MethodsUAS-based multispectral imaging and machine learning were used to monitor tar spot at different canopy and temporal levels and extract epidemiological parameters from multiple treatments. Disease severity was assessed visually at three canopy levels within micro-plots, while aerial images were gathered by UASs equipped with multispectral cameras. Both disease severity and multispectral images were collected from five to eleven time points each year for two years. Image-based features, such as single-band reflectance, vegetation indices (VIs), and their statistics, were extracted from ortho-mosaic images and used as inputs for machine learning to develop disease quantification models.Results and discussionThe developed models showed encouraging performance in estimating disease severity at different canopy levels in both years (coefficient of determination up to 0.93 and Lin’s concordance correlation coefficient up to 0.97). Epidemiological parameters, including initial disease severity or y0 and area under the disease progress curve, were modeled using data derived from multispectral imaging. In addition, results illustrated that digital phenotyping technologies could be used to monitor the onset of tar spot when disease severity is relatively low (< 1%) and evaluate the efficacy of disease management tactics under micro-plot conditions. Further studies are required to apply and validate our methods to large corn fields.

Differencing digital terrain models (DTMs) generated from multitemporal airborne light detection and ranging (lidar) data provide accurate and detailed information about three-dimensional (3D) changes on the Earth. However, noticeable spurious errors along flight paths are often included in the differencing results, hindering the accurate analysis of the topographic changes. This paper proposes a new scalable method to alleviate the problematic systematic errors with a high degree of automation in consideration of the practical limitations raised when processing the rapidly increasing amount of large-scale lidar datasets. The proposed method focused on estimating the displacements caused by vertical positioning errors, which are the most critical error source, and adjusting the DTMs already produced as basic lidar products without access to the point cloud and raw data from the laser scanner. The feasibility and effectiveness of the proposed method were evaluated with experiments with county-level multitemporal airborne lidar datasets in Indiana, USA. The experimental results demonstrated that the proposed method could estimate the vertical displacement reasonably along the flight paths and improve the county-level lidar differencing results by reducing the problematic errors and increasing consistency across the flight paths. The improved differencing results presented in this paper are expected to provide more consistent information about topographic changes in Indiana. In addition, the proposed method can be a feasible solution to upcoming problems induced by rapidly increasing large-scale multitemporal lidar given recent active government-driven lidar data acquisition programs, such as the U.S. Geological Survey (USGS) 3D Elevation Program (3DEP).

Digital terrain model (DTM) creation is a modeling process that represents the Earth’s surface. An aptly designed DTM generation method tailored for intended study can significantly streamline ensuing processes and assist in managing errors and uncertainties, particularly in large-area projects. However, existing methods often exhibit inconsistent and inexplicable results, struggle to clearly define what an object is, and often fail to filter large objects due to their locally confined operations. We introduce a new DTM generation method that performs object-based ground filtering, which is particularly beneficial for urban topography. This method defines objects as areas fully enclosed by steep slopes and grounds as smoothly connected areas, enabling reliable “object-based” segmentation and filtering, extending beyond the local context. Our primary operation, controlled by a slope threshold parameter, simplifies tuning and ensures predictable results, thereby reducing uncertainties in large-area modeling. Uniquely, our method considers surface water bodies in modeling and treats connected artificial terrains (e.g., overpasses) as ground. This contrasts with conventional methods, which often create noise near water bodies and behave inconsistently around overpasses and bridges, making our approach particularly beneficial for large-area 3D urban mapping. Examined on extensive and diverse datasets, our method offers unique features and high accuracy, and we have thoroughly assessed potential artifacts to guide potential users.

In general, remote sensing studies assessing cover crop growth are species nonspecific, use imagery from satellites or modified unmanned aerial vehicles (UAVs), and rely on multispectral vegetation indexes (VIs). However, using RGB imagery and visible-spectrum VIs from commercial off-the-shelf (COTS) UAVs to assess species specific cover crop growth is limited in the current scientific literature. Thus, this study evaluated RGB imagery and visible-spectrum VIs from COTS UAVs for suitability to estimate concentration (%) and content (kg ha−1) based cereal rye (CR) biomass, carbon (C), nitrogen (N), phosphorus (P), potassium (K), and sulfur (S). UAV surveys were conducted at two fields in Indiana and evaluated five visible-spectrum VIs—Visible Atmospherically Resistant Index (VARI), Green Leaf Index (GLI), Modified Green Red Vegetation Index (MGRVI), Red Green Blue Vegetation Index (RGBVI), and Excess of Greenness (ExG). This study utilized simple linear regression (VI only) and stepwise multiple regression (VI with weather and geographic data) to produce individual models for estimating CR biomass, C, N, P, K, and S concentration and content. The goodness-of-fit statistics were generated using repeated K-fold cross-validation to compare individual model performance. In general, the models developed using simple linear regression were inferior to those developed using the multiple stepwise regression method. Furthermore, for models developed using the multiple stepwise regression method all five VIs performed similarly when estimating concentration-based CR variables; however, when estimating content-based CR variables the models developed with GLI, MGRVI, and RGBVI performed similarly explaining 74–81% of the variation in CR data, and outperformed VARI and ExG. However, on an individual field basis, MGRVI consistently outperformed GLI and RGBVI for all CR characteristics. This study demonstrates the potential to utilize COTS UAVs for estimating in-field CR characteristics; however, the models generated in this study need further development to expand geographic scope and incorporate additional abiotic factors.

Construction projects should be planned and executed in a way that minimizes the inconvenience to the local community. For that, it is crucial to incorporate public opinion by engaging them in the decision-making process. However, the public is generally involved indirectly in the planning of infrastructure projects through information-sharing reports and meetings, which have not shown to be very effective. This paper presents the findings of a case study as a hands-on experience for graduate engineering students toward engaging the public in the feasibility assessment of a real-world rehabilitation project. The case study involves the application of a simple additive weighting (SAW) multi-criteria decision-making (MCDM) approach to the assessment of various dimensions of the proposed rehabilitation alternatives. As a part of the MCDM framework, public opinion is sought to determine the relative importance of various criteria in making the final decision. The steps and processes of the case study are summarized and proposed in the form of a framework for engaging both students and the community members in the planning of construction projects. The case study and the framework serve as a structured introductory exercise for raising awareness in the students about the impact of public opinion on the planning of construction projects, and the existence of methods that can help them articulate participatory processes. This structured exercise is replicable for future researchers. It is expected that the application of the approach pursued in this study will help promote a culture of accommodating public engagement among engineering students as future engineers in the long term.

Terrestrial laser scanners, unmanned aerial LiDAR, and unmanned aerial photogrammetry are increasingly becoming the go-to methods for forest analysis and mapping. The three-dimensionality of the point clouds generated by these technologies is ideal for capturing the structural features of trees such as trunk diameter, canopy volume, and biomass. A prerequisite for extracting these features from point clouds is tree segmentation. This paper introduces an unsupervised method for segmenting individual trees from point clouds. Our novel, canopy-to-root, least-cost routing method segments trees in a single routine, accomplishing stem location and tree segmentation simultaneously without needing prior knowledge of tree stem locations. Testing on benchmark terrestrial-laser-scanned datasets shows that we achieve state-of-the-art performances in individual tree segmentation and stem-mapping accuracy on boreal and temperate hardwood forests regardless of forest complexity. To support mapping at scale, we test on unmanned aerial photogrammetric and LiDAR point clouds and achieve similar results. The proposed algorithm’s independence from a specific data modality, along with its robust performance in simple and complex forest environments and accurate segmentation results, make it a promising step towards achieving reliable stem-mapping capabilities and, ultimately, towards building automatic forest inventory procedures.

Abstract Unoccupied aerial systems (UAS) were used to phenotype growth trajectories of inbred maize populations under field conditions. Three recombinant inbred line populations were surveyed on a weekly basis collecting RGB images across two irrigation regimens (irrigated and non‐irrigated/rain fed). Plant height, estimated by the 95th percentile (P95) height from UAS generated 3D point clouds, exceeded 70% correlation (r) to manual ground truth measurements and 51% of experimental variance was explained by genetics. The Weibull sigmoidal function accurately modeled plant growth (R2: >99%; RMSE:

Forest canopy height model (CHM) is useful for analyzing forest stocking and its spatiotemporal variations. However, high-resolution CHM with regional coverage is commonly unavailable due to the high cost of LiDAR data acquisition and computational cost associated with data processing. We present a CHM generation method using U.S. Geological Survey (USGS) 3D Elevation Program (3DEP) LiDAR data for tree height measurement capabilities for entire state of Indiana, USA. The accuracy of height measurement was investigated in relation to LiDAR point density, inventory height, and the timing of data collection. A simple data exploratory analysis (DEA) was conducted to identify problematic input data. Our CHM model has high accuracy compared to field-based height measurement (R2 = 0.85) on plots with relatively accurate GPS locations. Our study provides an easy-to-follow workflow for 3DEP LiDAR based CHM generation in a parallel processing environment for a large geographic area. In addition, the resulting CHM can serve as critical baseline information for monitoring and management decisions, as well as the calculation of other key forest metrics such as biomass and carbon storage.

Thanks to sensor developments, unmanned aircraft system (UAS) are the most promising modern technologies used to collect imagery datasets that can be utilized to develop agricultural applications in these days. UAS imagery datasets can grow exponentially due to the ultrafine spatial and high temporal resolution capabilities of UAS and sensors. One of the main obstacles to processing UAS data is the intensive computational resource requirements. The structure from motion (SfM) is the most popular algorithm to generate 3D point clouds, orthomosaic images, and digital elevation models (DEMs) in agricultural applications. Recently, the SfM algorithm has been implemented in parallel computing to process big UAS data faster for certain applications. This study evaluated the performance of parallel SfM processing on public cloud computing and on-premise cluster systems. The UAS datasets collected over cropping fields were used for performance evaluation. We used multiple computing nodes and centralized network storage with different network environments for the SfM workflow. In single-node processing, an instance with the most computing power in the cloud computing system performed approximately 20 and 35 percent faster than in the most powerful machine in the on-premises cluster. The parallel processing results showed that the cloud-based system performed better in speed-up and efficiency metrics for scalability, although the absolute processing time was faster in the on-premise cluster. The experimental results also showed that the public cloud computing system could be a good alternative computing environment in UAS data processing for agricultural applications.

Tar spot is a foliar disease of corn characterized by fungal fruiting bodies that resemble tar spots. The disease emerged in the U.S. in 2015, and severe outbreaks in 2018 caused an economic impact on corn yields throughout the Midwest. Adequate epidemiological surveillance and disease quantification are necessary to develop immediate and long-term management strategies. This study presents a measurement framework that evaluates the disease severity of tar spot using unmanned aircraft systems (UAS)-based plant phenotyping and regression techniques. UAS-based plant phenotypic information, such as canopy cover, canopy volume, and vegetation indices, were used as explanatory variables. Visual estimations of disease severity were performed by expert plant pathologists per experiment plot basis and used as response variables. Three regression methods, namely ordinary least squares (OLS), support vector regression (SVR), and multilayer perceptron (MLP), were used to determine an optimal regression method for UAS-based tar spot measurement. The cross-validation results showed that the regression model based on MLP provides the highest accuracy of disease measurements. By training and testing the model with spatially separated datasets, the proposed regression model achieved a Lin’s concordance correlation coefficient (ρc) of 0.82 and a root mean square error (RMSE) of 6.42. This study demonstrated that we could use the proposed UAS-based method for the disease quantification of tar spot, which shows a gradual spectral response as the disease develops.

Core Ideas High‐throughput imaging and genomic information can be combined to optimize marker development. Genome‐wide association studies identified loci associated with plant growth traits. We identified candidate genes associated with plant growth and development. Despite advances in sequencing for genotyping, the lack of rapid, accurate, and reproducible phenotyping platforms has hampered efforts to use genetic analysis to predict traits of interest. Therefore, the use of high‐throughput systems to phenotype traits related to crop growth, yield, quality, and resistance to biotic and abiotic stresses has become a major asset for breeding. Here, we assessed the efficacy of unmanned aircraft system (UAS)‐based high‐throughput phenotyping to obtain data for molecular marker development for spinach (Spinacia oleracea L.) improvement. We used a UAS equipped with a red–green–blue sensor to capture raw images of 284 spinach accessions throughout the crop cycle. Processed images generated orthomosaic and digital surface models for estimating canopy cover, canopy volume, and excess greenness index models. In addition, we manually recorded the number of days to bolting. Genome‐wide association studies against a single‐nucleotide polymorphism (SNP) panel obtained by ddRADseq identified 99 SNPs significantly associated with growth parameters. Some of these SNPs are in transcription factor and stress‐response genes with possible roles in plant growth and development. The results underscore the utility of combining aerial imaging and genomic data analysis to optimize marker development. This study lays the foundation for the use of UAS‐based high‐throughput phenotyping for the molecular breeding of spinach.

Drought significantly limits wheat productivity across the temporal and spatial domains. Unmanned Aerial Systems (UAS) has become an indispensable tool to collect refined spatial and high temporal resolution imagery data. A 2-year field study was conducted in 2018 and 2019 to determine the temporal effects of drought on canopy growth of winter wheat. Weekly UAS data were collected using red, green, and blue (RGB) and multispectral (MS) sensors over a yield trial consisting of 22 winter wheat cultivars in both irrigated and dryland environments. Raw-images were processed to compute canopy features such as canopy cover (CC) and canopy height (CH), and vegetation indices (VIs) such as Normalized Difference Vegetation Index (NDVI), Excess Green Index (ExG), and Normalized Difference Red-edge Index (NDRE). The drought was more severe in 2018 than in 2019 and the effects of growth differences across years and irrigation levels were visible in the UAS measurements. CC, CH, and VIs, measured during grain filling, were positively correlated with grain yield (r = 0.4–0.7, p < 0.05) in the dryland in both years. Yield was positively correlated with VIs in 2018 (r = 0.45–0.55, p < 0.05) in the irrigated environment, but the correlations were non-significant in 2019 (r = 0.1 to −0.4), except for CH. The study shows that high-throughput UAS data can be used to monitor the drought effects on wheat growth and productivity across the temporal and spatial domains.

Sorghum is one of the most important crops worldwide. An accurate and efficient high-throughput phenotyping method for individual sorghum panicles is needed for assessing genetic diversity, variety selection, and yield estimation. High-resolution imagery acquired using an unmanned aerial vehicle (UAV) provides a high-density 3D point cloud with color information. In this study, we developed a detecting and characterizing method for individual sorghum panicles using a 3D point cloud derived from UAV images. The RGB color ratio was used to filter non-panicle points out and select potential panicle points. Individual sorghum panicles were detected using the concept of tree identification. Panicle length and width were determined from potential panicle points. We proposed cylinder fitting and disk stacking to estimate individual panicle volumes, which are directly related to yield. The results showed that the correlation coefficient of the average panicle length and width between the UAV-based and ground measurements were 0.61 and 0.83, respectively. The UAV-derived panicle length and diameter were more highly correlated with the panicle weight than ground measurements. The cylinder fitting and disk stacking yielded R2 values of 0.77 and 0.67 with the actual panicle weight, respectively. The experimental results showed that the 3D point cloud derived from UAV imagery can provide reliable and consistent individual sorghum panicle parameters, which were highly correlated with ground measurements of panicle weight.

Citrus greening is a severe disease significantly affecting citrus production in the United States because the disease is not curable with currently available technologies. For this reason, monitoring citrus disease in orchards is critical to eradicate and replace infected trees before the spread of the disease. In this study, the canopy shape and vegetation indices of infected and healthy orange trees were compared to better understand their significant characteristics using unmanned aerial vehicle (UAV)-based multispectral images. Individual citrus trees were identified using thresholding and morphological filtering. The UAV-based phenotypes of each tree, such as tree height, crown diameter, and canopy volume, were calculated and evaluated with the corresponding ground measurements. The vegetation indices of infected and healthy trees were also compared to investigate their spectral differences. The results showed that correlation coefficients of tree height and crown diameter between the UAV-based and ground measurements were 0.7 and 0.8, respectively. The UAV-based canopy volume was also highly correlated with the ground measurements (R2 > 0.9). Four vegetation indices—normalized difference vegetation index (NDVI), normalized difference RedEdge index (NDRE), modified soil adjusted vegetation index (MSAVI), and chlorophyll index (CI)—were significantly higher in healthy trees than diseased trees. The RedEdge-related vegetation indices showed more capability for citrus disease monitoring. Additionally, the experimental results showed that the UAV-based flush ratio and canopy volume can be valuable indicators to differentiate trees with citrus greening disease.

Assessing plant population of cotton is important to make replanting decisions in low plant density areas, prone to yielding penalties. Since the measurement of plant population in the field is labor intensive and subject to error, in this study, a new approach of image-based plant counting is proposed, using unmanned aircraft systems (UAS; DJI Mavic 2 Pro, Shenzhen, China) data. The previously developed image-based techniques required a priori information of geometry or statistical characteristics of plant canopy features, while also limiting the versatility of the methods in variable field conditions. In this regard, a deep learning-based plant counting algorithm was proposed to reduce the number of input variables, and to remove requirements for acquiring geometric or statistical information. The object detection model named You Only Look Once version 3 (YOLOv3) and photogrammetry were utilized to separate, locate, and count cotton plants in the seedling stage. The proposed algorithm was tested with four different UAS datasets, containing variability in plant size, overall illumination, and background brightness. Root mean square error (RMSE) and R2 values of the optimal plant count results ranged from 0.50 to 0.60 plants per linear meter of row (number of plants within 1 m distance along the planting row direction) and 0.96 to 0.97, respectively. The object detection algorithm, trained with variable plant size, ground wetness, and lighting conditions generally resulted in a lower detection error, unless an observable difference of developmental stages of cotton existed. The proposed plant counting algorithm performed well with 0–14 plants per linear meter of row, when cotton plants are generally separable in the seedling stage. This study is expected to provide an automated methodology for in situ evaluation of plant emergence using UAS data.

To meet future world food and fiber demands, plant breeders must increase the rate of genetic improvement of important agricultural crops. One of the biggest obstacles now facing crop scientists is a phenotyping bottleneck. To ease this burden, the emerging technology of unmanned aerial systems (UAS) presents an exciting opportunity. To assess the utility of UAS, it is important to investigate their application across multiple crop species. Terminal plant height is of great importance to maize ( L.) and sorghum [ (L.) Moench] breeders and has been hypothesized to be useful but has been logistically impractical to measure in the field. In this study, we statistically analyzed in depth the ability of UAS to estimate height in sorghum (advanced and early generation material) and maize (optimal and late material) and the application of these estimates in breeding programs. We found that UAS explain genotypic variation similarly to ground-truth methods and that the repeatability of the methodology is high ( = 0.61–0.99), indicating effective differentiation of genotypes. Additionally, correlations between ground truth and UAS measurements were moderate to high for all materials ( = 0.4–0.9). Finally, we present a novel application for the technology in the form of high-resolution temporal growth curves. Using these UAS-generated growth curves, new physiological insights can be obtained and new avenues of scientific investigation are possible.

Unmanned aerial vehicle (UAV) platforms with sensors covering the red-edge and near-infrared (NIR) bands to measure vegetation indices (VIs) have been recently introduced in agriculture research. Consequently, VIs originally developed for traditional airborne and spaceborne sensors have become applicable to UAV systems. In this study, we investigated the difference in tillage treatments for cotton and sorghum using various RGB and NIR VIs. Minimized tillage has been known to increase farm sustainability and potentially optimize productivity over time; however, repeated tillage is the most commonly-adopted management practice in agriculture. To this day, quantitative comparisons of plant growth patterns between conventional tillage (CT) and no tillage (NT) fields are often inconsistent. In this study, high-resolution and multi-temporal UAV data were used for the analysis of tillage effects on plant health and the performance of various vegetation indices investigated. Time series data over ten dates were acquired on a weekly basis by RGB and multispectral (MS) UAV platforms: a DJI Phantom 4 Pro and a DJI Matrice 100 with the SlantRange 3p sensor. Ground reflectance panels and an ambient illumination sensor were used for the radiometric calibration of RGB and MS orthomosaic images, respectively. Various RGB and NIR-based vegetation indices were then calculated for the comparison between CT and NT treatments. In addition, a one-tailed Z-test was conducted to check the significance of VIs’ difference between CT and NT treatments. The results showed distinct differences in VIs between tillage treatments during the whole growing season. NIR-based VIs showed better discrimination performance than RGB-based VIs. Out of 13 VIs, the modified soil adjusted vegetation index (MSAVI) and optimized soil adjusted vegetation index (OSAVI) showed better performance in terms of quantitative difference measurements and the Z-test between tillage treatments. The modified green red vegetation index (MGRVI) and excess green (ExG) showed reliable separability and can be an alternative for economic RGB UAV application.

Unmanned aerial vehicle (UAV) images have great potential for various agricultural applications. In particular, UAV systems facilitate timely and precise data collection in agriculture fields at high spatial and temporal resolutions. In this study, we propose an automatic open cotton boll detection algorithm using ultra-fine spatial resolution UAV images. Seed points for a region growing algorithm were generated hierarchically with a random base for computation efficiency. Cotton boll candidates were determined based on the spatial features of each region growing segment. Spectral threshold values that automatically separate cotton bolls from other non-target objects were derived based on input images for adaptive application. Finally, a binary cotton boll classification was performed using the derived threshold values and other morphological filters to reduce noise from the results. The open cotton boll classification results were validated using reference data and the results showed an accuracy higher than 88% in various evaluation measures. Moreover, the UAV-extracted cotton boll area and actual crop yield had a strong positive correlation (0.8). The proposed method leverages UAV characteristics such as high spatial resolution and accessibility by applying automatic and unsupervised procedures using images from a single date. Additionally, this study verified the extraction of target regions of interest from UAV images for direct yield estimation. Cotton yield estimation models had R2 values between 0.63 and 0.65 and RMSE values between 0.47 kg and 0.66 kg per plot grid.

Continuing population growth will result in increasing global demand for food and fiber for the foreseeable future. During the growing season, variability in the height of crops provides important information on plant health, growth, and response to environmental effects. This paper indicates the feasibility of using structure from motion (SfM) on images collected from 120 m above ground level (AGL) with a fixed-wing unmanned aerial vehicle (UAV) to estimate sorghum plant height with reasonable accuracy on a relatively large farm field. Correlations between UAV-based estimates and ground truth were strong on all dates (R2 > 0.80) but are clearly better on some dates than others. Furthermore, a new method for improving UAV-based plant height estimates with multi-level ground control points (GCPs) was found to lower the root mean square error (RMSE) by about 20%. These results indicate that GCP-based height calibration has a potential for future application where accuracy is particularly important. Lastly, the image blur appeared to have a significant impact on the accuracy of plant height estimation. A strong correlation (R2 = 0.85) was observed between image quality and plant height RMSE and the influence of wind was a challenge in obtaining high-quality plant height data. A strong relationship (R2 = 0.99) existed between wind speed and image blurriness.