35180 research outputs found
Sort by
MOBILE VISION-BASED TOOLS FOR YIELD ESTIMATION IN WINE GRAPES USING RGB AND 3D IMAGING
Get PDFEarly season precision yield estimation is a critical challenge in modern viticulture, influencing vineyard operation management, resource allocation, and productivity. Traditional methods, such as manual cluster and berry counting, are labor-intensive, prone to sampling bias, affected by vine-to-vine variability, and limited in accuracy due to occlusions and environmental conditions. These factors contribute to inconsistent and low throughput yield estimation, limiting their practical use. Therefore, this study developed a smartphone-based machine vision system that integrates RGB imaging, 3D point cloud analysis, and deep learning models to improve both early- and harvest-season yield prediction.This study first investigated early-season yield estimation by automating lag-phase detection, a critical stage in berry development when growth slows, and berry size reaches approximately half of its final weight. Using RGB images and a Mask Region-based Convolutional Neural Network (Mask R-CNN) model, the system detected and measured berry size throughout the season. The model achieved a Root Mean Square Error (RMSE) of 0.473 mm and an R2 of 0.837 against ground truth berry measurements. The system successfully tracked berry growth and identified the onset of the lag-phase with a Mean Absolute Error (MAE) of seven days. Automating this traditionally labor-intensive process allows growers to predict the onset of lag phase leading to estimating current crop and predict harvest yield. Such estimates can be used to make timely production management decisions including crop thinning, irrigation, and fertigation, ultimately optimizing yield and berry quality. Early forecasting/prediction of harvest yield also helps growers with effective harvest and post-harvest planning.This dissertation also investigated on ways the 3D point cloud data could be used to extract canopy and cluster volumes to improve harvest yield estimation. It involved segmenting canopy and cluster structures from 3D point cloud data acquired using a mobile device. Canopy volume was found to be a critical feature for yield prediction, which was segmented with 98% accuracy using a Gradient Boosting Classifier. For grape cluster segmentation, a YOLO11 deep learning model with 0.98 mean Average Precision (mAP) was employed to detect and isolate clusters from RGB images before generating 3D models. The detected cluster images were processed using Structure-from-Motion (SfM) to reconstruct dense 3D point clouds, providing segmented cluster point clouds necessary for volume estimation. Once segmentation was completed, surface reconstruction techniques were applied to compute canopy and cluster volumes. An alpha shape reconstruction algorithm was then used to generate a watertight 3D surface from the segmented point clouds, enabling precise volume calculations. The reconstructed canopy surface, when compared to ground truth, demonstrated a relative error of 21.68%, confirming its accuracy in representing canopy structure and maintaining measurement reliability. For cluster volume estimation, the SfM-derived 3D models were analyzed against ground-truth volume measurements obtained from manually measured clusters. However, occlusions and irregular cluster geometries introduced challenges, leading to an RMSE of 46.47 cm3 (44.72% relative error) for partially occluded clusters. When only fully visible clusters were analyzed, RMSE improved to 24.50 cm3 (22.28% relative error), highlighting the importance of reducing occlusions and improving reconstruction accuracy for more precise cluster volume estimation.Finally, a comprehensive yield estimation model was developed combining RGB imagery features and 3D canopy volume data. Deep learning models effectively detected key vineyard attributes, with the YOLO11 model achieving a mAP of 0.76 for grape cluster detection and segmentation and 0.65 for shoot detection, despite challenges posed by overlapping and occluded clusters and shoots. The yield estimation model, developed using Linear Regression, demonstrated moderate predictive performance at the individual vine level, achieving an R2 of 0.375 and an RMSE of 1.35 kg per vine. The low R2 suggests that yield at the individual vine level was influenced by factors beyond the model’s predictors, such as variability in vine vigor, soil properties, and microclimate effects. However, when aggregated at the vineyard scale, the model’s performance improved significantly. The total harvest yield prediction error at the vineyard level decreased from 4.21% to 1.61% as the sampling percentage increased from 15% to 40% of vines. These results suggest that errors and variabilities at the individual vine level may be balanced across the vineyard, making the system practical for vineyard management operations such as harvest logistics and planning. By leveraging both 2D and 3D imaging techniques, the model successfully addressed challenges such as occlusion, lighting variability, and irregular cluster structures, demonstrating its high accuracy, scalability, and adaptability across diverse vineyard conditions. The use of smartphones to collect RGB and 3D images ensures that this system is highly practical, enabling growers to implement precision agriculture techniques without the need for specialized equipment.Overall, this dissertation demonstrated the feasibility of an automated machine vision system for both early-season (lag-phase-based) and harvest-season yield prediction. By addressing key challenges in early and harvest season yield estimation, these systems enhance the precision of vineyard management operations, providing a foundation for improved sustainability, reduced resource waste, and higher productivity
Examining the Role of Frost Damage on Leaf Composition to Unravel the Mystery of the "Frost" Taint Wine Phenomenon
Get PDFThe volatile organic composition of Cabernet Sauvignon leaves subjected to freeze drying (FD) and natural freezing (NF) was measured using untargeted headspace solid-phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS) to establish a chemical relationship between grapevine leaf composition and the sensory characteristics of wines affected by frost damage, specifically the "frost" or "rose" taint phenomenon. This study thoroughly investigated the influence of freeze events by comparing the volatile organic compounds (VOCs) present in NF, FD, and control (fresh) Cabernet Sauvignon grapevine leaves. Results confirmed 6-methyl-5-hepten-2-ol and p-menth-1-en-9-al as important chemical markers associated with herbal and coriander-like aroma attributes. These volatile changes were confirmed in wines produced with varied dosages of freeze-damaged leaf material (0.0 g/kg, 0.9 g/kg, 3.6 g/kg, and 8.0 g/kg). Sensory analysis provided partial validation of these chemical changes, although establishing definitive sensory thresholds for frost taint remains an ongoing area of research. Additionally, this study explored the effectiveness of freeze-drying as an alternative method to replicate naturally occurring frost damage, revealing limitations in accurately reproducing the sensory characteristics associated with natural frost events
CORDON SANITAIRE EMPIRE AND DISEASE PREVENTION IN BRITISH AND AMERICAN OVERSEAS COLONIES IN EAST AND SOUTHEAST ASIA, 1870-1910
Get PDFThis study explores the progress of medical modernization of British and American colonial territories in East and Southeast Asia, including the British Straits Settlements, British Hong Kong, the American Philippines and treaty ports of Qing China. In doing so, it draws on Epistemic Network, Epidemic Network and Actor Network theories. With the development of understanding of the causation of epidemic diseases, the construction of sanitary works, and the improvement of sanitary measures through statistics collecting and information recording and compiling, the colonial medical experts observed epidemics and provided suggestions to colonial authorities in order to meet the goal of the control of infectious diseases, especially cholera and plague, in the late nineteenth and early twentieth century British and American overseas formal and informal territories in East and Southeast Asia. Focusing on the development of the transition from miasma to germ theories of epidemics in formal and informal colonies and their metropoles, this dissertation argues that the developments in Western medicine also affected the colonial and Qing government’s attitudes to epidemic policies and contributed to conflicts between colonists and colonial subjects in late nineteenth and early twentieth century East and Southeast Asia. This study argues that the colonial governments imposed strict sanitary surveillance on local populations in order to meet the goal of disease control. Simultaneously, colonial medical experts used empirical approaches, clinical observation, investigation, field survey, and laboratory works, to develop an epistemic network because the existing sanitary measures were not effective which created a breakthrough in the understanding of epidemics in the early twentieth century European medical world and Asian colonies. Examination of the collaboration between human and non-human agents, such as medical experts, colonial and Qing officials, patients, different ethnic groups, colonial governments, governmental institutions, hospitals, vectors, even microbes reveals that these different agents converged, intertwined to develop the understanding of the nature of epidemics and reached the goal of the control of epidemic diseases. This study indicates that empirical approaches not only developed the understanding of the causation of epidemics but also shaped the colonial civilization and sanitary modernization in the late nineteen and early twentieth century American and British East and Southeast Asian colonies. Behind the development of the understanding of epidemics and sanitary modernization, it reveals the strict social control and sanitary surveillance imposed by colonial governments and colonial sanitary authorities on colonial subjects
USING MARK-RELEASE-RECAPTURE AND DISTANCE SAMPLING METHODS TO ESTIMATE OREGON SILVERSPOT BUTTERFLY POPULATION SIZE AND DEMOGRAPHY
Get PDFButterflies are in decline wordwide. In the face of habitat loss and fragmentation, understanding population dynamics and dispersal behavior is critical for implementing effective conservation strategies. However, for many at-risk species, basic biological information is lacking, which limits our understanding of species’ needs and the success of recovery efforts. Additionally, commonly used monitoring strategies often fail to provide the information needed to adapt management approaches. This thesis addresses key knowledge gaps for the federally threatened Oregon silverspot butterfly (Argynnis = Speyeria zerene hippolyta), focusing on population size, adult demography, and dispersal. In Chapter One, we use two monitoring methods – distance sampling and mark-release-recapture (MRR) – to estimate population size and assess the relative reliability of current index count-based monitoring. We also investigate adult demography across three occupied sites and compare survival rates between captive-reared individuals released for population augmentation with wild origin individuals. We found that counts did not consistently scale with population size and that distance sampling and index counts showed opposing population trends, emphasizing the importance of accounting for detection probability to improve the reliability of population estimates. MRR also revealed substantial variation in lifespan across sites and lower survival rates for captive-reared individuals compared to wild individuals. These findings inform the development of a long-term monitoring strategy using distance sampling to reliably estimate population size and evaluate the ongoing effectiveness of augmentation. In Chapter Two, we used MRR data to assess adult dispersal in Oregon silverspot metapopulations occupying patchy landscapes. We compared total movement distances and patch-to-patch movements with a literature review of 18 other at-risk Nymphalid species. We found that Oregon silverspot butterflies have moderately strong dispersal capacity – similar to or greater than other medium-sized fritillaries – with total movement distances of several kilometers and frequent patch-to-patch movements. These findings suggest that management should focus on expanding the overall size of habitat networks to support larger population sizes. Together, these chapters emphasize the importance of species-specific data on movement, demography, and population size for informing habitat restoration and long-term monitoring. By using field-based research to support conservation management, this work provides key insights to advance evidence-based conservation of Oregon silverspot butterflies and other at-risk species in fragmented ecosystems
Identity and Materiality in Diaspora
Get PDFThe Indian diaspora in the Dallas Fort Worth Metroplex is one of the largest in the United States. The first swaths of Indian immigrants began arriving in Texas in the 1960 and the growth in recent years has been exponential. Using a mixed methodological approach, I set out to understand the relationship between the Indian diaspora and public/private material representation, consumption, and identity. What I found was a unique identity with a few key factors differentiating the DFW Indian diaspora from other large diasporas in Texas and the US. These primary distinctions include an intentionality of engagement, a pattern of accelerated integration, and a generalized experience of authenticity apart from an individual’s personal and or regional representation. The first and second are closely related though distinct. The second calls into question what constitutes authenticity. Finally, the utility of the findings is explored in the realms of education, civic participation, and the building of community cohesion
BOUNDARY-ENFORCED PHYSICS-INFORMED NEURAL NETWORKS FOR MICROFLUIDIC DEVICE PERFORMANCE IN EARLY CANCER DETECTION
Get PDFDeterministic Lateral Displacement (DLD) devices are vital tools in microfluidics, enabling size-based, label-free separation of cells and particles. These devices play an essential role in cancer diagnostics by effectively isolating circulating tumor cells (CTCs) from blood samples. However, traditional methods used to evaluate and optimize DLD devices, such as computational fluid dynamics (CFD) simulations, are often costly, complex, and very time-consuming. While machine learning (ML) methods, particularly deep learning, offer potential improvements, current models typically require extensive modifications to physical datasets and domain restructuring, limiting their accuracy and ability to generalize to new scenarios. This research introduces an advanced Physics-Informed Deep Neural Network (PIDNN) that significantly enhances the prediction of velocity fields within DLD devices. Unlike conventional ML methods, PIDNN uniquely integrates essential physics principles directly into its architecture, enforcing critical boundary conditions and initial condition. This integration ensures physically accurate predictions, substantially improving the reliability of the model. The PIDNN is trained using detailed velocity field data generated by COMSOL Multiphysics simulations. Model inputs include critical parameters such as non-dimensional diameter (F), period number (N), Reynolds number (Re), and spatial coordinates. Furthermore, an innovative data sampling technique is introduced to enhance data density near crucial device boundaries, effectively capturing essential flow features. The application of PIDNN drastically reduces the time needed for evaluating device performance, enabling rapid selection of optimal design parameters that greatly enhance the effectiveness of particle and cell separation. When combined with a particle trajectory solver, the PIDNN can accurately predict particle trajectories and critical non-dimensional diameters (Dc) with average error less than 5% , essential for efficient cell separation. This innovative approach streamlines the development of versatile, high-performance DLD devices, significantly advancing their practical applications in cancer diagnostics
A COMMUNITY COLLEGE CO-REQUISITE MATHEMATICS CLASS THROUGH THE LENS OF MATHEMATICS ANXIETY STUDENT PERCEPTIONS AND EXPERIENCES
Get PDFStudents at community colleges, particularly those taking developmental courses, have among the highest concentrations of mathematics anxiety at any educational level. Most students referred to pre-requisite remedial courses do not successfully complete them. Consequently, many community colleges have changed to the co-requisite model for mathematics support to shorten the path to mathematics course completion. With little research focused on community colleges and the relatively new co-requisite model, there is a lack of research in these areas.This study amplified student voices to describe students' emotional and academic experiences with mathematics anxiety while enrolled in a developmental (co-requisite) statistics course at a community college. Data collection included interviews, class observations, and the Abbreviated Math Anxiety Survey (AMAS). Findings indicated that students expressed a reduction in their stress and math anxiety due to instructional decisions and relationship-building by the professor. However, the greatest influence on student success was not mathematics anxiety but their complicated lives. Based on these findings, I suggest several recommendations for students, faculty, and administrators and offer suggestions for further research
IMPACT OF METRITIS AND ANTIMICROBIALS ON DAIRY INDUSTRY SUSTAINABILITY
Get PDFThe objective of this thesis was to review important topics related to uterine diseases in dairy cows and highlight a project conducted as part of the MS program that aimed to evaluate the differences in lactational performance associated with antimicrobial therapy and clinical cure of metritis in dairy cows. In this study, data from two randomized controlled trials consisting of 4,744 Holstein cows from 5 dairy farms in California, Florida, and Texas was used. Five groups composed this study, clinically cured cows that received ceftiofur (CEFC); not cured cows that received ceftiofur (CEFN); clinically cured cows that did not receive ceftiofur (NTC); not cured cows that did not receive ceftiofur (NTN); cows without metritis (NMET). Risk and time of pregnancy and culling, total milk production by 300 days in milk were assessed across groups. Overall, clinical cure of metritis was positively associated with milk production and reproduction, regardless of antimicrobial therapy, warranting further investigation regarding selective therapy of metritis
Evaluating a Regional Police Peer Support Program Using Interpretive Phenomenological Analysis to Improve Program Integration
Get PDFThis study evaluates a regional peer support program for three smaller police agencies in eastern Washington and northern Idaho. The goal was to explore cultural and practical factors that influence the implementation process. I conducted a mixed-methods evaluation, collecting quantitative and qualitative data from sworn and professional staff across the three agencies. Over 12 months, 547 survey responses from monthly assessments measured sleep quality (Pittsburgh Sleep Quality Index), perceived stress (Perceived Stress Scale), job satisfaction (Minnesota Job Satisfaction), and quality of life (World Health Organization Quality of Life). Additionally, 34 semi-structured interviews were conducted at two points during the program implementation process. Participants included sworn and professional staff, peer support team members, and non-team members. Secondary agency data was collected to assess organizational impacts, including use of force events, citizen complaints, overtime, sick and bereavement leave, training hours, and staffing ratios. I also conducted participant observations throughout the project, which served as an additional source of qualitative data. Quantitative analysis involved descriptive statistics, t-tests, ANOVA, and linear mixed-effects models. An Interpretive Phenomenological Analysis approach was used to analyze qualitative data, highlighting the lived experiences of agency members during the launch of the peer support team program. Findings are connected to existing policing theories, including camaraderie and masculinity, while emphasizing unique challenges faced by rural departments. This applied research, conducted in collaboration with participating agencies, provided real-time insights to support program development and adjustments. The results offer valuable guidance for regional peer support initiatives in small police agencies
Co-Simulation of Distributed Microgrid Control and Communication
Get PDFMuch of the potential for multi-microgrid Systems to provide improved resiliency and other benefits depends on effective collaboration between independent agents such as microgrid controllers (MGCs) and distribution system operators (DSOs). Therefore, proper design and implementation of the communication systems is essential. We cannot neglect to understand the interdependence within the cyber-physical multi-microgrid system and the communication infrastructure requirements to support the operations of smart multi-microgrid systems. To properly design the communication system for a smart multi-microgrid system we need to understand what applications and processes are needed as well as their communication requirements and be able to model and test the system for contingencies. At this time, no single tool exists that can model the power system, communication system, and control logic. In fact very little work has been done to simulate distributed algorithms with accurate communication models. In the course of this research, significant contributions were made in the study of distributed algorithms for multi-microgrid systems. First, the Cyber-Physical Multi-Agent Co-Simulation platform (CPMACS) platform is developed. CPMACS, enables power distribution systems, communications systems, and control agents to be simulated together using the software packages, GridLAB-D, NS-3, and Python, respectively. The timing and message passing between simulators is managed by HELICS. Second, the communication system requirements of Equivalent Network Approximation (ENApp) and Alternating Direction Method of Multipliers (ADMM) algorithms for distributed optimal power flow in normal operation are analyzed with CPMACS through a series of tests which force the algorithms to operate with, high communication delays, low bandwidth, high congestion, and data corruption. Third, the communication system requirements of a variety of methods for bulk power system voltage support are analyzed. The ENApp algorithm is adapted to bulk grid voltage support and analyzed on a variety of test systems with various communication restrictions and data corruptions as done for the normal mode. Additionally, an average consensus algorithm and a collaborative autonomy algorithm, which each depend on different data and connections, are analyzed and compared along with the ENApp algorithm