Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Labeled Datasets”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 253 records · Page 14

Gradient-Based Novelty Detection Boosted by Self-Supervised Binary Classification

Novelty detection aims to automatically identify out-of-distribution (OOD) data, without any prior knowledge of them. It is a critical step in data monitoring, behavior analysis and other applications, helping enable continual learning in the field. Conventional methods of OOD detection perform multi-variate analysis on an ensemble of data or features, and usually resort to the supervision with OOD data to improve the accuracy. In reality, such supervision is impractical as one cannot anticipate the anomalous data. In this paper, we propose a novel, self-supervised approach that does not rely on any pre-defined OOD data: (1) The new method evaluates the Mahalanobis distance of the gradients between the in-distribution and OOD data. (2) It is assisted by a self-supervised binary classifier to guide the label selection to generate the gradients, and maximize the Mahalanobis distance. In the evaluation with multiple datasets, such as CIFAR-10, CIFAR-100, SVHN and TinyImageNet, the proposed approach consistently outperforms state-of-the-art supervised and unsupervised methods in the area under the receiver operating characteristic (AUROC) and area under the precision-recall curve (AUPR) metrics. We further demonstrate that this detector is able to accurately learn one OOD class in continual learning.

Sun, Jingbo↗

FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics

Large language models have revolutionized artificial intelligence by enabling large, generalizable models trained through self-supervision. This paradigm has inspired the development of scientific foundation models (FMs). However, applying this capability to experimental particle physics is challenging due to the sparse, spatially distributed nature of detector data, which differs dramatically from natural language. This work addresses if an FM for particle physics can scale and generalize across diverse tasks. We introduce a new dataset with more than 11 million particle collision events and a suite of downstream tasks and labeled data for evaluation. We propose a novel self-supervised training method for detector data and demonstrate its neural scalability with models that feature up to 188 million parameters. With frozen weights and task-specific adapters, this FM consistently outperforms baseline models across all downstream tasks. The performance also exhibits robust data-efficient adaptation. Further analysis reveals that the representations extracted by the FM are task-agnostic but can be specialized via a single linear mapping for different downstream tasks.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Multi-Level Anomaly Detection on Time-Varying Graph Data

This work presents a novel modeling and analysis framework for graph sequences which addresses the challenge of detecting and contextualizing anomalies in labelled, streaming graph data. We introduce a generalization of the BTER model of Seshadhri et al. by adding flexibility to community structure, and use this model to perform multi-scale graph anomaly detection. Specifically, probability models describing coarse subgraphs are built by aggregating probabilities at finer levels, and these closely related hierarchical models simultaneously detect deviations from expectation. This technique provides insight into a graph's structure and internal context that may shed light on a detected event. Additionally, this multi-scale analysis facilitates intuitive visualizations by allowing users to narrow focus from an anomalous graph to particular subgraphs or nodes causing the anomaly. For evaluation, two hierarchical anomaly detectors are tested against a baseline Gaussian method on a series of sampled graphs. We demonstrate that our graph statistics-based approach outperforms both a distribution-based detector and the baseline in a labeled setting with community structure, and it accurately detects anomalies in synthetic and real-world datasets at the node, subgraph, and graph levels. To illustrate the accessibility of information made possible via this technique, the anomaly detector and an associated interactive visualization tool are tested on NCAA football data, where teams and conferences that moved within the league are identified with perfect recall, and precision greater than 0.786.

anomaly detection↗

Simulating metagenomic stable isotope probing datasets with MetaSIPSim

DNA-stable isotope probing (DNA-SIP) links microorganisms to their in-situ function in diverse environmental samples. Combining DNA-SIP and metagenomics (metagenomic-SIP) allows us to link genomes from complex communities to their specific functions and improves the assembly and binning of these targeted genomes. However, empirical development of metagenomic-SIP methods is hindered by the complexity and cost of these studies. We developed a toolkit, ‘MetaSIPSim,’ to simulate sequencing read libraries for metagenomic-SIP experiments. MetaSIPSim is intended to generate datasets for method development and testing. To this end, we used MetaSIPSim generated data to demonstrate the advantages of metagenomic-SIP over a conventional shotgun metagenomic sequencing experiment. Through simulation we show that metagenomic-SIP improves the assembly and binning of isotopically labeled genomes relative to a conventional metagenomic approach. Improvements were dependent on experimental parameters and on sequencing depth. Community level G + C content impacted the assembly of labeled genomes and subsequent binning, where high community G + C generally reduced the benefits of metagenomic-SIP. Furthermore, when a high proportion of the community is isotopically labeled, the benefits of metagenomic-SIP decline. Finally, the choice of gradient fractions to sequence greatly influences method performance. Metagenomic-SIP is a valuable method for recovering isotopically labeled genomes from complex communities. We show that metagenomic-SIP performance depends on optimization of experimental parameters. MetaSIPSim allows for simulation of metagenomic-SIP datasets which facilitates the optimization and development of metagenomic-SIP experiments and analytical approaches for dealing with these data.

59 BASIC BIOLOGICAL SCIENCES↗

Dataset 1: A National and City Dataset on Human Factors in Pooled Rideshare, 2021

Dataset 1: A National and City Dataset on Human Factors in Pooled Rideshare, 2021. Dataset Description: Pooled Rideshare Acceptance Survey - Phase 1 (2021, N = 5,385). This dataset captures responses from a nationally representative sample of 5,385 adults across the United States to understand public acceptance, preferences, and behavioral intentions related to pooled rideshare (PR) services. The primary objective of this research is to provide actionable insights to inform the design, deployment, and policy development of sustainable shared mobility systems. Data was collected via an online survey administered through a national panel provider. Participants ranged in age from 18 to 95 years, and representation from all U.S. regions. The survey instrument was designed to explore numerous dimensions related to PR adoption including demographic traits, current travel habits, rideshare familiarity, trust, safety, environmental attitudes, and user experience preferences. Both rideshare users and non-users were included, offering a diverse range of perspectives. - Phase_1_Final - The dataset includes survey items developed from literature reviews, and prior field studies. Each row represents an individual respondent, and each column corresponds to a variable such as willingness to use pooled rideshare, attitudes toward specific service features, and sociodemographic data. The data is available in both .CSV and .SAV formats. - Phase_1_Final_MapFile - The accompanying data dictionary explains all variable labels, response scales, and codes. An .XLSX format of the full survey instrument is also included to support interpretation and reuse of the dataset.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Southwest Pacific tropical cyclone development classification utilizing machine learning and synoptic composites

This study evaluates the ability of machine learning algorithms to classify tropical depressions (TDs) and tropical storms (TSs) in the western region of the southwest Pacific Ocean (SWPO). Decision rules are generated to predict the environment required for a depression to fully develop into a mature storm, and the most influential predictors in the classification decision are ranked. TD and TS are discriminated based on a maximum sustained wind speed threshold (≥17 ms -1 ). Various aerosol, thermodynamic, and dynamic parameters are extracted closest to the initiation point of each non-developing and developing sample. The covariates associated with each labelled sample are used to train a decision tree and random forest model. Results using a testing dataset suggest the random forest approach more accurately distinguishes between non-developing and developing samples. The classification accuracy of the decision tree and random forest are 72% and 91%, respectively. Random forest outperformed the decision tree by providing higher accuracy in test data. The most important variables for binary classification are sea salt aerosol optical depth (AOD), 1,000 mb relative humidity, and sea surface temperature. AOD is a quantitative estimate of the aerosols presents in the air through the extinction of a ray of light as it passes through the atmosphere. Mean composite maps constructed in an unsupervised manner have been created for the most important variables identified by the random forest classifier during TD and TS events to highlight the difference in geophysical and aerosol variables' climatology during the two different classifications. This work will advance the risk management strategies for northeastern Australia and other SWPO basin islands to control their tropical cyclone related losses through prioritizing forecasting variables that are the strongest predictors of the strengthening of tropical depressions into tropical cyclones.

54 ENVIRONMENTAL SCIENCES↗

Latent Laplacian Maximum Entropy Discrimination for Detection of High-Utility Anomalies

Data-driven anomaly detection methods suffer from the drawback of detecting all instances that are statistically rare, irrespective of whether the detected instances have realworld significance or not. In this paper, we are interested in the problem of specifically detecting anomalous instances that are known to have high real-world utility, while ignoring the low-utility statistically anomalous instances. To this end, we propose a novel method called Latent Laplacian Maximum Entropy Discrimination (LatLapMED) as a potential solution. This method uses the EM algorithm to simultaneously incorporate the Geometric Entropy Minimization principle for identifying statistical anomalies, and the Maximum Entropy Discrimination principle to incorporate utility labels, in order to detect highutility anomalies. Here, we apply our method in both simulated and real datasets to demonstrate that it has superior performance over existing alternatives that independently pre-process with unsupervised anomaly detection algorithms before classifying.

97 MATHEMATICS AND COMPUTING↗

Data Driven Commercial Building Energy Code Compliance and Technology Inventory for New York City

Building Performance Standards (BPS) are gaining national traction. A BPS will require new processes in the design, construction, and operation of buildings that take the occupants into account and enable predictive analysis to ensure compliance with current and future GHG emissions caps. In New York City, most buildings over 25,000 square feet will be regulated by a BPS starting in 2024, regardless of whether it is new construction permitted under current energy codes or an existing building. This research is one of the first to begin the evaluation of a long-term series of building policies in the context of an open data ecosystem, in cooperation with city agencies. Existing building policies enacted in NYC have ranged from building energy benchmarking and labeling to energy audits to the regulation of GHG emission in buildings. Through the development of a dataset related to building technologies and energy consumption, this project can help to evaluate if meaningful conclusions can be drawn for the data that has been largely self-reported in compliance with city regulations. This project will also provide lessons learned from a deep dive into these types of datasets to provide best practices for municipalities or states seeking to embark on policies like those enacted in NYC. In addition, a Building Automation System (BAS) Stretch Standard of Care (SSOC) for owners, designers, and building operators will enable the measurement and predictive analysis of energy consumption and GHG emissions at the plant, system, or component level, in anticipation of regulated GHG limits on buildings based on energy use. The SSOC is expected to be suitable for use on a national level. The primary feature of an SSOC is a standardized format for a set of BAS points that can be used to control and to gather data from individual plants, systems, or components that are related to building energy consumption. This project examined how measurements compare to prescriptive or simulation-based energy code targets, finding little correlation between predictive 8760-hour energy modeling and actual energy consumption for a small sample (n=27) of buildings constructed after 2015. Other analysis found that, while large multifamily housing (MFH) buildings showed a general trend similar to predicted reductions in energy use from the implementation of model commercial energy codes, this trend was not evident in the office, K-12 school, and hotel use groups in NYC. No upward or downward trends in energy consumption were found when buildings were grouped by size. Energy audit data were analyzed and it appears that there is bias by audit company on measures recommended to clients. Further research should be performed to cross-analyze this with other attributes, such as building size, vintage, and number of stories. Analysis found that for 281 buildings that were permitted and completed after 2015 and had submitted benchmarking data in 2022, between 81% and 96% (by use group) were found to be in compliance with the 2024 to 2029 NYC BPS emission caps, and between 55% and 89% were in compliance with the 2030-2034 caps. This work is beneficial to the public in helping policymakers and building stakeholders better understand the wide-ranging implications of a BPS.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Dynamic Networks Experiment 2: Measuring Associator Sensitivity to Signal Detection Errors

Using the Dynamic Networks Experiment 2 (DNE2) human-analyst event bulletin picks as a baseline signal detection dataset, we generate 47 additional datasets by gradually reducing their accuracy and completeness by randomly removing DNE2 picks, changing the initial phase labels from P to S and vice-versa, and injecting noise detections to simulate real-world signal detection algorithms.

58 GEOSCIENCES↗

Coincidence anomaly detection for unsupervised locating of edge localized modes in the DIII-D tokamak dataset

Using supervised learning to train a machine learning model to predict an on-coming edge localized mode (ELM) requires a large number of labeled samples. Creating an appropriate data set from the very large database of discharges at a long-running tokamak, such as DIII-D, would be a very time-consuming process for a human. Considering this need and difficulty, we use coincidence anomaly detection, an unsupervised learning technique, to train an ELM-identifier to identify and label ELMs in the DIII-D discharge database. This ELM-identifier shows, simultaneously, a precision of 0.68 and a recall of 0.63 (AUC is 0.73) on identifying ELMs in example time series pulled from thousands of discharges spanning five years. In a test set of 50 discharges, the algorithm finds over 26 thousand ELM candidates, more than 5 times the existing catalog of ELMs labeled by humans.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Generation of random geological models using multi-randomization for machine learning

Generating high-fidelity geological models is essential for advancing machine learning (ML) methods in automated seismic interpretation. For instance, seismic images paired with corresponding fault labels are foundational for ML-based fault detection from seismic migration sections. While several open-access datasets of random geological models exist, open-source tools specifically designed to produce large volumes of such models for ML applications remain scarce. To address this gap, we present RGM (Random Geological Model), an open-source software package for efficiently generating 2D and 3D synthetic geological models tailored for ML workflows. RGM supports the creation of diverse model components, including medium property distributions (P-/S-wave velocities and density), seismic reflectivity images (i.e., synthetic migration sections), relative geological time, and discrete fault attributes such as probability, dip, strike, rake, and displacement. It also accommodates the creation of complex geological features such as salt bodies and unconformities. The model generation algorithm employs a multi-randomization strategy, yielding an effectively infinite-dimensional model space that encompasses a wide range of geological scenarios and associated seismic features. Furthermore, RGM incorporates a method to generate synthetic elastic migration images using analytical elastic reflection coefficients combined with frequency-dependent scaling. This functionality enables the creation of training datasets for ML models that leverage elastic seismic images. RGM is implemented in modern object-oriented Fortran, allowing users to flexibly control statistical parameters governing model variability. We demonstrate the capability, performance, and geological realism of the package through comprehensive 2D and 3D examples.

58 GEOSCIENCES↗

Aftershock Identification Using a Paired Neural Network Applied to Constructed Data

This report is intended to detail the findings of our investigation of the applicability of machine learning to the task of aftershock identification. The ability to automatically identify nuisance aftershock events to reduce analyst workload when searching for events of interest is an important step in improving nuclear monitoring capabilities and while waveform cross - correlation methods have proven successful, they have limitations (e.g., difficulties with spike artifacts, multiple aftershocks in the same window) that machine learning may be able to overcome. Here we apply a Paired Neural Network (PNN) to a dataset consisting of real, high quality signals added to real seismic noises in order to work with controlled, labeled data and establish a baseline of the PNN's capability to identify aftershocks. We compare to waveform cross - correlation and find that the PNN performs well, outperforming waveform cross - correlation when classifying similar waveform pairs, i.e., aftershocks.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

An Agenda for Multimodal Foundation Models for Earth Observation

Archives of remote sensing (RS) data are increasing swiftly as new sensing modalities with enhanced spatiotemporal resolution become operational. While promising new breakthroughs, the sheer volume of RS archives stretches the limits of human analysts and existing AI tools, as most models are: i) limited to single data modalities; ii) task-specific; iii) heavily reliant on labeled data. The emerging Foundation Models (FMs) have the potential to address these limitations. Trained on vast unlabeled datasets through self-supervised learning, FMs enable generic feature extraction that facilitate specialization to a wide variety of downstream tasks. This paper describes a vision towards an FM for multimodal Earth Observation data (FM4EO), discussing key building blocks and open challenges. We put particular emphasis on multimodal reasoning, a topic underexplored in EO. Our ultimate goal is a practical path toward FM4EO with capacity to unlock breakthroughs in few-shot learning scenarios, multimodal geographic knowledge integration, synthesis, and hypothesis generation.

Ambrozio Dias, Philipe↗

Leaf-scale carbon assimilation simulated by the multi-assumption architecture & testbed for process sensitivity analysis

Raw and processed output from the Multi-Assumption Architecture & Testbed (MAAT, https://github.com/walkeranthonyp/MAAT, tag: v1.2.1_walker2020; Walker et al., 2018). The dataset contains leaf-scale carbon assimilation data from three MAAT simulations and their post-processing and plotting scripts. Simulation 1, labelled "FvC_4proc_proc," is a four-process, process sensitivity analysis (Dai et al., 2017) of carbon assimilation based on the Farquhar et al. (1980) photosynthesis model, and its variant proposed by Collatz et al. (1991). Simulation 2, labelled "FvC_4proc_salt," is a 14 parameter sensitivity analysis (Saltelli et al., 2010) of carbon assimilation based on the Farquhar and Collatz photosynthesis models. Simulation 3, labelled "FvC," is a factorial combination of the alternative process representations and parameter values presented in those original papers across a factorial combination of incident photosynthetically active radiation (50-2000 µmol m-2 s-1, in increments of 50) and atmospheric CO2 concentration (50-1500 µmol m-2 s-1, in increments of 50). Additional scripts/functions for post-processing and plotting are part of the MAAT repository. These data were generated to evaluate and compare the Farquhar and Collatz models of leaf-scale photosynthesis across a range of CO2 and incident light conditions. Analysis of results can be found in Walker et al. (2020). This dataset is comprised of R scripts to process data and produce figures; .RDS files, a binary format specific to R and read with the 'readRDS' function, to store raw MAAT output; R and bash scripts (.bs) that help to run a MAAT simulation, .XML files that are output from MAAT describing a run, and .csv files which contain the calculated sensitivity indexes published in Walker et al. (2020).

54 ENVIRONMENTAL SCIENCES↗

Site and endmember spectra of terrestrial vegetation and soils for the Colorado Headwaters Ecological Spectroscopy Study, June-July 2025

This dataset provides site and endmember spectra collected during the 2025 Colorado Headwaters Ecological Spectroscopy Study (CHESS) campaign. The site spectra were collected to help validate airborne hyperspectral data acquired by the National Ecological Observatory Network's aerial observation platform (NEON AOP). Endmember spectra were collected to augment existing spectral libraries with additional samples of bare surfaces and non-photosynthetic vegetation. All measurements were acquired with an Analytical Spectral Devices (ASD) FieldSpec4 Hi-Res NG (Next Generation) spectroradiometer, which records radiance at 1nm (nanometer) intervals from the ultraviolet to the short-wave infrared (350-2500 nm). The dataset includes spectra measured at meadow sites where the CHESS team also collected vegetation samples for trait analyses. The site spectra were collected with the ASD FieldSpec4 palm grip attachment using an 8° field-of-view foreoptic. Site spectra are integrated measurements of the entire surface within the foreoptic’s field of view. For site-level spectra, the sun is the illumination source. A Spectralon panel mounted on a tripod was used for instrument optimization and white reference measurements for all site spectra. Site spectra were acquired within two hours of solar noon and within 48 hours of a NEON AOP overflight. Site spectra are labeled by date, sampling area, and site number according to the naming conventions of the CHESS campaign’s data management plan. The dataset also contains endmember spectra in the following categories: photosynthetic vegetation (PV), non-photosynthetic vegetation (NPV), bare (soil/rock), and flowers. Endmember measurements were acquired using either the contact probe or the leaf clip attachments of the ASD FieldSpec4. In these configurations, the bulb inside the spectrometer provides the light source for the measurements. The spectrometer was optimized and white reference measurements were recorded using the circular white pucks attached to the contact probe and leaf clip. Because they do not rely on solar illumination, contact probe and leaf clip measurements were collected during a broader time frame than the palm grip site spectra. Some endmembers were measured at CHESS meadow sites, while others were collected within the larger sampling area or in nearby locations (e.g. Gothic Townsite) with similar characteristics. Radiance, reflectance, and metadata files are split into three subfolders according to measurement type: proximal/palm grip (prx), contact probe (cp), and leaf clip (lc). Radiance spectra are provided in ASD file format (.asd file extension). All ASD files can be opened using the provided scripts. Metadata is provided in two formats: CSV file format (no geolocation) and GEOJSON file format (includes geolocation for each spectra). The dataset includes a set of pre-processed reflectance spectra as CSV files (yyyymmdd_rfl.csv). The python scripts and jupyter notebook used to calculate reflectance spectra from the ASD radiance data is included here and was previously published at: https://doi.org/10.3334/ORNLDAAC/2446. There is also a folder of JPEG photographs corresponding to selected spectra. We include a protocol document with detailed steps for ASD FieldSpec4 assembly and operations. This data additionally contains a file level metadata (flmd.csv) and data dictionary (dd.csv) file. Geospatial information: Geospatial data for mapping measurement site locations are in the files CHESS_polygons_lai_UTM.geojson, CHESS_polygons_shrub_UTM.geojson, and CHESS_polygons_meadow_UTM.geojson in the companion geospatial package for the 2025 CHESS campaign, ‘CHESS 2025: Location data for field observations and sampling’ (Henderson et al., 2026). CHESS Project Description: The Colorado Headwaters Ecological Spectroscopy Study (CHESS) comprised a multi-week airborne remote sensing and field observation campaign in the Upper Gunnison Basin, Colorado, conducted in June and July of 2025. Airborne remote sensing was conducted by the National Ecological Observatory Network Airborne Observation Platform (NEON AOP), concurrent with a field campaign run by the Rocky Mountain Biological Laboratory (RMBL), the Lawrence Berkeley National Laboratory (LBNL) and SLAC National Accelerator Laboratory Watershed Function Science Focus Area (SFA), and NASA-JPL (Jet Propulsion Laboratory) Earth Surface Mineral Dust Source Investigation (EMIT) program. Between June 10 and July 18, 2025, the NEON AOP flight team collected high-resolution aerial imaging spectroscopy and Light Detection and Ranging (LiDAR) data over three domains: the Upper East River (CRBU), Almont Triangle (ALMO), and the Upper Taylor Basin (UPTA). In coordination with the flights, a field campaign acquired ground-truth observations, including observations of vegetation composition, foliar traits, forest demography, and subsurface properties in 18 core sampling areas within the domains. Additional surface water observations were taken at over 380 point locations. All CHESS campaign datasets can be found within the CHESS ESS-DIVE data portal: https://data.ess-dive.lbl.gov/portals/chess. Funding Acknowledgment: This research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004) and was funded by EMIT Extended Mission Phase E Science.

2018 NEON and 2025 CHESS Campaigns↗

Selecting a Classification Ensemble and Detecting Process Drift in an Evolving Data Stream

We characterize the commercial behavior of a group of companies in a common line of business using a small ensemble of classifiers on a stream of records containing commercial activity information. This approach is able to effectively find a subset of classifiers that can be used to predict company labels with reasonable accuracy. Performance of the ensemble, its error rate under stable conditions, can be characterized using an exponentially weighted moving average (EWMA) statistic. The behavior of the EWMA statistic can be used to monitor a record stream from the commercial network and determine when significant changes have occurred. Results indicate that larger classification ensembles may not necessarily be optimal, pointing to the need to search the combinatorial classifier space in a systematic way. Results also show that current and past performance of an ensemble can be used to detect when statistically significant changes in the activity of the network have occurred. The dataset used in this work contains tens of thousands of high level commercial activity records with continuous and categorical variables and hundreds of labels, making classification challenging.

Ensemble classifiers↗

Seismic signal augmentation to improve generalization of deep neural networks

Deep learning has emerged as an effective approach for seismic data processing in general, and for earthquake monitoring in particular. The ability of deep learning models to generalize beyond the training and validation data is important for comprehensive earthquake monitoring; this ability furthermore depends on the availability of a sufficiently large and complete training dataset. However, this requirement can prove challenging to meet due to significant effort and time for data collection and labeling. Data augmentation provides an efficient and effective approach for increasing the dimension of training samples and improving generalization to unseen samples. In this paper, we present augmentation methods appropriate for seismic waveforms and demonstrate their ability to reduce bias and increase performance. Furthermore, these augmentation methods can be applied to a wide range of deep learning applications designed for seismic data.

58 GEOSCIENCES↗

Evaluation of artificial neural network performance for classification of potato plants infected with potato virus Y using spectral data on multiple varieties and genotypes

Potato virus Y (Potyviridae, PVY) is a plant virus that poses a significant threat to potato producers on a global basis. The pathogen has disrupted seed potato supplies and negatively impacted yield and quality of commercial potato crops. The potato industry currently manages PVY infection levels via insecticide applications, regional seed certification programs that rely on field scouting to visually assess individual plants for infection status, and destructive and costly tissue sampling coupled with laboratory assays. Despite these efforts, PVY continues to confound potato industry stakeholders resulting in economic harm. Remote sensing and machine learning provide for the development of new tools to more accurately detect and spatially quantify PVY-infected plants versus the current state of the art. However, there is a need to understand how the occurrence of many different potato varieties impact the dynamics of developing models to detect potato plants impacted with PVY and their potential effectiveness. This study evaluates classification modelling outcomes using spectral datasets collected in different temporal and spatial environments (greenhouse and a production field) on multiple potato varieties consisting of labelled instances of plants infected with PVY and those not infected with the virus. A modelling framework was developed to support iterative modelling runs using artificial neural network (ANN) architectures configured as binary classifiers to develop sample populations to support statistical analysis on model performance using specific spectral subsets. When using spectral data to detect PVY-infected plants, ANN models achieved the highest mean accuracy of 0.894 on a single variety. Conversely, the same ANN model architecture only achieved a mean accuracy of 0.575 on a spectral data set representing 29 potato breeding lines. Additionally, statistical analysis indicates spectral regions including the red edge, near infrared and shortwave infrared contain more important spectral features for the ANN classifier introduced in this research.

60 APPLIED LIFE SCIENCES↗