This thesis focuses on uncertainties in remotely sensed image segmentation and their visualisation. The first part describes a visualisation tool, allowing interaction with the parameters of a fuzzy classification algorithm by visually adjusting fuzzy membership functions of classes in a 3D feature space plot. Its purpose is to improve insight into fuzzy classification of remotely sensed imagery and related uncertainty. Additionally, alpha-shapes are used to visualise irregular shaped class clusters. The second part of the thesis describes segmentation techniques for identification of objects and quantification of their uncertainties. The Local Binary Pattern (LBP) operator is used to model texture. A multivariate extension of the standard univariate LBP operator is proposed to describe texture in multiple bands. Texture-based image segmentation, provides good results yielding valuable information about object uncertainty at transition zones. Visualisation methods described in the first part and segmentation techniques described in the second part are combined and extended to visualise object uncertainty. An object is visualised in 3D feature space and in geographic space based on a user-defined uncertainty threshold. Isosurfaces provide a visualisation technique for fast interaction facilitating visualisation of the relation between uncertainty in the spatial extent of objects and their thematic uncertainty.

No description available.

No description available.

No description available.

<div>Sorghum, which is grown internationally as a cereal crop that is robust to heat, drought, and disease, has numerous applications for food, forage, and biofuels. When monitoring the growth stages of sorghum, or phenotyping specific traits for plant breeding, it is important to identify and monitor the panicles in the field due to their impact relative to grain production. Several studies have focused on detecting panicles based on data acquired by RGB and multispectral remote sensing technologies. However, few experiments have included hyperspectral data because of its high dimensionality and computational requirements, even though the data provide abundant spectral information. Relative to analysis approaches, machine learning, and specifically deep learning models have the potential of accommodating the complexity of these data. In order to detect panicles in the field with different physical characteristics, such as colors and shapes, very high spectral and spatial resolution hyperspectral data were collected with a wheeled-based platform, processed, and analyzed with multiple extensions of the VGG-16 Fully Convolutional Network (FCN) semantic segmentation model.</div><div><br></div><div>In order to have correct positioning, orthorectification experiments were also conducted in the study to obtain the proper positioning of the image data acquired by the pushbroom hyperspectral camera at near range. The scale of the DSM derived from LiDAR that was used for orthorectification of the hyperspectral data was determined to be a critical issue, and the application of the Savitzky-Golay filter to the original DSM data was shown to contribute to the improved quality of the orthorectified imagery.</div><div><br></div><div>Three tuned versions of the VGG-16 FCN Deep Learning architecture were modified to accommodate the hyperspectral data: PCA&FCN, 2D-FCN, and 3D-FCN. It was concluded that all the three models can detect the late season panicles included in this study, but the end-to-end models performed better in terms of precision, recall, and the F-score metrics . Future work should focus on improving annotation strategies and the model architecture to detect different panicle varieties and to separate overlapping panicles based on an adequate quantities of training data acquired during the flowering stage.</div>

This study advances understanding of how the changes in ecosystem structure and function associated with woody shrub encroachment in semi-arid grasslands alter ecosystem carbon (C) dynamics. In terms of both magnitude and dynamism, dryland ecosystems represent a major component of the global C cycle. Woody shrub encroachment is a widespread phenomenon globally, which is known to substantially alter ecosystem structure and function, with resultant impacts on C dynamics. A series of focal sites were studied at the Sevilleta National Wildlife Refuge in central New Mexico, USA. A space-for-time analogue was used to identify how landscape structure and function change at four stages over a grassland to shrubland transition. The research had three key threads: 1. Soil-associated carbon: Stocks of organic and inorganic C in the near-surface soil, and the redistribution of these C stocks by erosion during high-intensity rainfall events were quantified using hillslope-scale monitoring plots. Coarse (>2 mm) clasts were found to account for a substantial proportion of the organic and inorganic C in these calcareous soils, and the erosional effluxes of both inorganic and organic C increased substantially across the vegetation ecotone. Eroded sediment was found to be significantly enriched in organic C relative to the contributing soil with systematic changes in OC enrichment across the vegetation transition. The OC enrichment dynamics observed were inconsistent with existing understanding (derived largely from reductionist, laboratory-based experiments) that OC enrichment is largely insignificant in the erosional redistribution of C. 2. Plant biomass: Cutting-edge proximal remote sensing approaches, using a remotely piloted lightweight multirotor drone combined with structure-from-motion (SfM) photogrammetry were developed and used to quantify biomass carbon stocks at the focal field sites. In such spatially heterogeneous and temporally dynamic ecosystems existing measurement techniques (e.g. on-the-ground observations or satellite- or aircraft-based remote sensing) struggle to capture the complexity of fine-grained vegetation structure, which is crucial for accurately estimating biomass. The data products available from the novel SfM approach developed for this research quantified plants just 15 mm high, achieving a fidelity nearly two orders of magnitude finer than previous implementations of the method. The approach developed here will revolutionise the study of biomass dynamics in short-sward ecogeomorphic systems. 3. Ecohydrological modelling: Understanding the effects of water-mediated degradation processes on ecosystem carbon dynamics over greater than observable spatio-temporal scales is complicated by significant scale-dependencies and thus requires detailed mechanistic understanding. A process-based, spatially-explicit ecohydrological modelling approach (MAHLERAN - Model for Assessing Hillslope to Landscape Erosion, Runoff and Nutrients) was therefore comprehensively evaluated against a large assemblage of rainfall runoff events. This evaluation highlighted both areas of strength in the current model structure, and also areas of weakness for further development. The research has improved understanding of ecosystem degradation processes in semi-arid rangelands, and demonstrates that woody shrub encroachment may lead to a long-term reduction in ecosystem C storage, which is contrary to the widely promulgated view that woody shrub encroachment increases C storage in terrestrial ecosystems.

Measurements of soil moisture are a crucial component for understanding the global water and carbon cycle, weather forecasting, climate models, drought prediction, and agriculture production. Active and passive microwave radar instruments are currently in use for remote sensing of soil moisture. Signals of Opportunity (SoOp) based remote sensing has recently emerged as a complementary method for soil moisture remote sensing. SoOp reuses general digital communication signals allowing the reuse of allocated wireless communication signal bands for science measurements. This thesis developed a tower based SoOp instrument implementing frequencies in the P-Band and S-Band. Two field campaigns were conducted using this new instrument during the summers of 2017 and 2018 at Purdue's Agronomy Center for Research and Education.

<p>High-throughput phenotyping using high spatial, spectral, and temporal resolution remote sensing (RS) data has become a critical part of the plant breeding chain focused on reducing the time and cost of the selection process for the “best” genotypes with respect to the trait(s) of interest. In this study, the potential of accurate and reliable sorghum biomass prediction using hyperspectral and LiDAR data acquired by sensors mounted on UAV platforms is investigated. Experiments comprised multiple varieties of grain and forage sorghum, including some photoperiod sensitive varieties, providing an opportunity to evaluate a wide range of genotypes and phenotypes. </p><p>Feature extraction is investigated, where various novel features, as well as traditional features, are extracted directly from the hyperspectral imagery and LiDAR point cloud data and input to classical machine learning (ML) regression based models. Predictive models are developed for multiple experiments conducted during the 2017, 2018, and 2019 growing seasons at the Agronomy Center for Research and Education (ACRE) at Purdue University. The impact of the regression method, data source, timing of RS and field-based biomass reference data acquisition, and number of samples on the prediction results are investigated. R2 values for end-of-season biomass ranged from 0.64 to 0.89 for different experiments when features from all the data sources were included. Using geometric based features derived from the LiDAR point cloud and the chemistry-based features extracted from hyperspectral data provided the most accurate predictions. The analysis of variance (ANOVA) of the accuracies of the predictive models showed that both the data source and regression method are important factors for a reliable prediction; however, the data source was more important with 69% significance, versus 28% significance for the regression method. The characteristics of the experiments, including the number of samples and the type of sorghum genotypes in the experiment also impacted prediction accuracy. </p><p>Including the genomic information and weather data in the “multi-year” predictive models is also investigated for prediction of the end of season biomass. Models based on one and two years of data are used to predict the biomass yield for the future years. The results show the high potential of the models for biomass and biomass rank predictions. While models developed using one year of data are able to predict biomass rank, using two years of data resulted in more accurate models, especially when RS data, which encode the environmental variation, are included. Also, the possibility of developing predictive models using the RS data collected until mid-season, rather than the full season, is investigated. The results show that using the RS data until 60 days after sowing (DAS) in the models can predict the rank of biomass with R2 values of around 0.65-0.70. This not only reduces the time required for phenotyping by avoiding the manual sampling process, but also decreases the time and the cost of the RS data collections and the associated challenges of time-consuming processing and analysis of large data sets, and particularly for hyperspectral imaging data.</p><p>In addition to extracting features from the hyperspectral and LiDAR data and developing classical ML based predictive models, supervised and unsupervised feature learning based on fully connected, convolutional, and recurrent neural networks is also investigated. For hyperspectral data, supervised feature extraction provides more accurate predictions, while the features extracted from LiDAR data in an unsupervised training yield more accurate prediction. </p><p>Predictive models based on Recurrent Neural Networks (RNNs) are designed and implemented to accommodate high dimensional, multi-modal, multi-temporal data. RS data and weather data are incorporated in the RNN models. Results from multiple experiments focused on high throughput phenotyping of sorghum for biomass predictions are provided and evaluated. Using proposed RNNs for training on one experiment and predicting biomass for other experiments with different types of sorghum varieties illustrates the potential of the network for biomass prediction, and the challenges relative to small sample sizes, including weather and sensitivity to the associated ground reference information.</p>

To support the growing human population, plant phenotyping technologies must innovate to rapidly interpret hyperspectral (HS) data into genetic inferences for plant breeders and managers. While pigment and nutrient concentrations within canopies are known to be vertically non-uniform, these chemical distributions as sources of HS noise are not universally addressed in scaling leaf information to canopy data nor in detecting spectral plant health traits. <br>In this project, soybeans (Glycine Max, cultivar Williams 82) were imaged with a Spectra Vista Corporation (SVC) HR-1024 spectroradiometer (350-2500 nm) at the highest five node positions. The samples were subjected to nitrogen and drought stress in factorial design (n=12) that was validated via relative water content (RWC) and PLS Regression of photopigments (chlorophyll a, chlorophyll b, lutein, neoxanthin, violaxanthin, and zeaxanthin in mg/g DW) and N concentration (%) for each imaged tissue. Welch’s ANOVA and Tamhane’s T2 post-hoc testing quantified spectral heterogeneity with respect to treatments and node positions through spectral angle measurements (SAMs) and percent NDVI difference. Drought-stressed samples had the lowest SAM between node positions compared to other treatments, and SAM node comparisons were greatest when including the highest sampled tissues. Taking ratios of NDVI between node positions proved more statistically effective at discerning between all factorial treatments than individual leaf NDVI values. Finally, intra-canopy spectral heterogeneity was exploited by training Linear Discriminant Analysis (LDA) classifiers on relative reflectance between node positions, tuning for the F1-Score. A classifier built on Node 1 vs. Node 3 reflectance outperformed in class-specific accuracies compared to analogous models trained on point-view data. Accounting for intra-canopy spectral variability is an opportunity to develop more comprehensive phenotyping tools for plant breeders in a world with rapidly rising agricultural demand.<br><br>

<p>The spaceborne lidar Ice, Cloud, and land Elevation Satellite (ICESat)-2 provides the ATL13 data product for inland water bodies. However, its quality characteristics are not yet fully understood. This study presents a robust method for extracting lake water level data and makes a comprehensive evaluation on the determined water levels. The selected study areas are Lake Huron and Lake Superior, which are part of the Great Lakes. The extracted water levels from ATL13 over a period of four years are validated by using the field measurements at the closest NOAA hydrological stations. The evaluation is carried out in terms of data specifications, wind speed, frozen precipitation, distance of photon segments to hydrological stations, data acquisition time, and beam intensity. The determined water levels are then further used for seasonal monitoring and modeling of water surface. This work demonstrates the critical need for outlier removal and the capability of the ATL13 data. A total bias of 9 - 10 cm is found in the ATL13 product. It is found that frozen precipitation can lead to an overestimation (~ 5 cm) of the water level. However, the uncertainty of water level determination is not found to be significantly related to the laser beam intensity and data acquisition time. We expect that these findings will be valuable for users employing the ATL13 inland water body product and for developers producing future versions of the ATL13 product.</p>