Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Dataset Annotations”

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 37 records · Page 2

Automatic Crack Segmentation and Feature Extraction in Electroluminescence Images of Solar Modules

The effect of cracks in solar cells on the long-term degradation of photovoltaic (PV) modules remains to be determined. To investigate this effect in future studies, it is necessary to quantitatively describe the crack features (e.g., length) and correlate them with module power loss. Electroluminescence (EL) imaging is a common technique for identifying cracks. However, it is currently challenging and time-consuming to identify cracks in a large number of EL images and quantify complex crack features by human inspection. This article introduces a fast semantic segmentation method (~0.18 s/cell) to automatically segment cracks from EL images and algorithms to extract crack features. Here we fine-tuned a UNet neural network model using pretrained VGG16 as the encoder and obtained an average F1 score of 0.875 and an intersection over union score of 0.782 on the testing set. With cracks and busbars segmented, we developed algorithms for extracting crack features, including the crack-isolated area, the brightness inside the isolated area, and the crack length. We also developed an automatic preprocessing tool for cropping individual cell images from EL images of PV modules (~0.72 s/module). Our codes are published as open-source an software, and our annotated dataset composed of various types of cells is published as a benchmark for crack segmentation in EL images.

14 SOLAR ENERGY↗

SpaceNet 9—Cross-Sensor Alignment of Optical and SAR Imagery

Precise registration of high-resolution synthetic aperture radar (SAR) and optical imagery is necessary for realizing the full potential and benefits of multimodal image analysis. However, two significant challenges presently exist. First, there is a lack of annotated datasets and benchmarks available for high-resolution SAR–optical image registration. Second, an assessment of efficient and reliable image registration methods that can precisely align these modalities is lacking. Here, we present a holistic description of the SpaceNet 9 Challenge and its results. We present a description of the dataset and baseline algorithm along with the results of the challenge, including a description of the winning algorithms. We release the SpaceNet 9 dataset along with open-sourcing the winning algorithms and baseline. The objective of SpaceNet 9 was to compute a dense displacement map that indicates the shift needed to align pixels in an optical image to the pixels in a SAR image. The challenge launched in April 2025 and was active for approximately two months. The top five solutions reduced image alignment error from approximately 34 m to under 13 m for public and private test data, with the best results obtaining a registration error of only 8.5 and 6.7 m on the public testing and private testing dataset, respectively. Usage of pretrained image matching models, robust outlier rejection with RANSAC, and estimating local displacement were common among the top solutions. The results of this challenge provide insight into high-resolution SAR–optical image registration and offer opportunities for future benchmarking in this domain. The baseline algorithm, winning solutions, and datasets are available at https://spacenet.ai/sn9-challenge/.

benchmark datasets↗

Diagnosing Malaria Patients with Plasmodium falciparum and vivax Using Deep Learning for Thick Smear Images

We propose a new framework, PlasmodiumVF-Net, to analyze thick smear microscopy images for a malaria diagnosis on both image and patient-level. Our framework detects whether a patient is infected, and in case of a malarial infection, reports whether the patient is infected by Plasmodium falciparum or Plasmodium vivax. PlasmodiumVF-Net first detects candidates for Plasmodium parasites using a Mask Regional-Convolutional Neural Network (Mask R-CNN), filters out false positives using a ResNet50 classifier, and then follows a new approach to recognize parasite species based on a score obtained from the number of detected patches and their aggregated probabilities for all of the patient images. Reporting a patient-level decision is highly challenging, and therefore reported less often in the literature, due to the small size of detected parasites, the similarity to staining artifacts, the similarity of species in different development stages, and illumination or color variations on patient-level. We use a manually annotated dataset consisting of 350 patients, with about 6000 images, which we make publicly available together with this manuscript. Our framework achieves an overall accuracy above 90% on image and patient-level.

60 APPLIED LIFE SCIENCES↗

AI Foundation Models for Science: An Open Collaborative Initiative

Foundation Models (FMs), AI models designed to replace task-specific models, are increasingly being recognized for their versatility across numerous downstream applications. These models, trained using self-supervised techniques on any type of sequence data, circumvent the need for large annotated datasets, a major bottleneck in traditional AI model development. FMs can be applied to downstream tasks using few-shot learning and fine-tuning, significantly reducing the need for large labeled training datasets and computational resources. However, the development of FMs requires substantial resources, including access to data and compute power, expertise in the latest models, and specialized scientific knowledge for systematic evaluation. It is challenging for a single group to possess all these capabilities. To address this, NASA IMPACT has initiated an open collaborative effort, leveraging partnerships with the private sector and other groups within and outside NASA, to jointly build FMs. The overarching goal is to develop a consistent and collaborative approach to building FMs for high-value science datasets. This initiative has fostered collaboration within NASA and with external partners, including IBM Research, Clark University, DOE’s ORNL, ESA, and USGS. The effort focuses on identifying key datasets with a wide range of downstream applications, pretraining and building FMs using modified transformer architectures, evaluating compute infrastructure needs, and sharing models, pretraining and fine-tuning code, and data with the community. Furthermore, it aims to train the Earth science community to fine-tune these models for various downstream applications. Our initial effort resulted in the creation of a 100 million parameter HLS Geospatial Model within six months, which was released on HuggingFace. We are now expanding our scope to include data from weather and climate models and investigating multimodal models. We invite those interested in participating in this effort to join us by sharing their use cases, expertise, or data.

Rahul Ramachandran↗

A traffic accident dataset for Chattanooga, Tennessee

This publication presents an annotated accident dataset which fuses traffic data from radar detection sensors, weather condition data, and light condition data with traffic accident data (as illustrated in Fig. 1) in a format that is easy to process using machine learning tools, databases, or data workflows. The purpose of this data is to analyze, predict, and detect traffic patterns when accidents occur. Each file contains a timeseries of traffic speeds, flows, and occupancies at the sensor nearest to the accident, as well as 5 neighboring sensors upstream and downstream. It also contains information about the accident type, date, and time. In addition to the accident data, we provide baseline data for typical traffic patterns during a given time of day. Overall, the dataset contains 6 months of annotated traffic data from November 2020 to April 2021. During this timeframe, and 361 accidents occurred in the monitored area around Chattanooga, Tennessee. This dataset served as the basis for a study on topology-aware automated accident detection for a companion publication [1].

97 MATHEMATICS AND COMPUTING↗

Dataset: Occupancy Detection, Tracking, and Estimation Using a Vertically Mounted Depth Sensor

Occupancy detection, tracking, and estimation has a wide range of applications including improving building energy efficiency, safety, and security of the occupants. As depth sensors are getting cheaper, they offer a viable solution to estimate occupancy accurately in a non-privacy invasive manner. Even though there are publicly available depth datasets, they do not consider placing the sensor in the ceiling looking downwards to estimate occupancy. We deployed four Kinect for XBOX One in four CMU classrooms and conference rooms for a period of four weeks in 2017 and collected over 6 TB of depth data. We annotate this huge dataset by labelling bounding boxes around occupants and release the annotated dataset.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Dataset: Occupancy Detection, Tracking, and Estimation Using a Vertically Mounted Depth Sensor

Occupancy detection, tracking, and estimation has a wide range of applications including improving building energy efficiency, safety, and security of the occupants. As depth sensors are getting cheaper, they offer a viable solution to estimate occupancy accurately in a non-privacy invasive manner. Even though there are publicly available depth datasets, they do not consider placing the sensor in the ceiling looking downwards to estimate occupancy. We deployed four Kinect for XBOX One in four CMU classrooms and conference rooms for a period of four weeks in 2017 and collected over 6 TB of depth data. We annotate this huge dataset by labelling bounding boxes around occupants and release the annotated dataset.

47 OTHER INSTRUMENTATION↗

Improvement of eukaryotic protein predictions from soil metagenomes

During the last decades, metagenomics has highlighted the diversity of microorganisms from environmental or host-associated samples. Most metagenomics public repositories use annotation pipelines tailored for prokaryotes regardless of the taxonomic origin of contigs. Consequently, eukaryotic contigs with intrinsically different gene features, are not optimally annotated. Using a bioinformatics pipeline, we have filtered 7.9 billion contigs from 6,872 soil metagenomes in the JGI’s IMG/M database to identify eukaryotic contigs. We have re-annotated genes using eukaryote-tailored methods, yielding 8 million eukaryotic proteins and over 300,000 orphan proteins lacking homology in public databases. Comparing the gene predictions we made with initial JGI ones on the same contigs, we confirmed our pipeline improves eukaryotic proteins completeness and contiguity in soil metagenomes. The improved quality of eukaryotic proteins combined with a more comprehensive assignment method yielded more reliable taxonomic annotation. This dataset of eukaryotic soil proteins with improved completeness, quality and taxonomic annotation reliability is of interest for any scientist aiming at studying the composition, biological functions and gene flux in soil communities involving eukaryotes.

54 ENVIRONMENTAL SCIENCES↗

BioADAPT-MRC: adversarial learning-based domain adaptation improves biomedical machine reading comprehension task

ABSTRACT Motivation Biomedical machine reading comprehension (biomedical-MRC) aims to comprehend complex biomedical narratives and assist healthcare professionals in retrieving information from them. The high performance of modern neural network-based MRC systems depends on high-quality, large-scale, human-annotated training datasets. In the biomedical domain, a crucial challenge in creating such datasets is the requirement for domain knowledge, inducing the scarcity of labeled data and the need for transfer learning from the labeled general-purpose (source) domain to the biomedical (target) domain. However, there is a discrepancy in marginal distributions between the general-purpose and biomedical domains due to the variances in topics. Therefore, direct-transferring of learned representations from a model trained on a general-purpose domain to the biomedical domain can hurt the model’s performance. Results We present an adversarial learning-based domain adaptation framework for the biomedical machine reading comprehension task (BioADAPT-MRC), a neural network-based method to address the discrepancies in the marginal distributions between the general and biomedical domain datasets. BioADAPT-MRC relaxes the need for generating pseudo labels for training a well-performing biomedical-MRC model. We extensively evaluate the performance of BioADAPT-MRC by comparing it with the best existing methods on three widely used benchmark biomedical-MRC datasets—BioASQ-7b, BioASQ-8b and BioASQ-9b. Our results suggest that without using any synthetic or human-annotated data from the biomedical domain, BioADAPT-MRC can achieve state-of-the-art performance on these datasets. Availability and implementation BioADAPT-MRC is freely available as an open-source project at https://github.com/mmahbub/BioADAPT-MRC. Supplementary information Supplementary data are available at Bioinformatics online.

60 APPLIED LIFE SCIENCES↗

ScienceSearch: Enabling Search through Automatic Metadata Generation

© 2018 IEEE. Scientific facilities are increasingly generating and handling large amounts of data from experiments and simulations. Next-generation scientific discoveries rely on insights derived from data, especially across domain boundaries. Search capabilities are critical to enable scientists to discover datasets of interest. However, scientific datasets often lack the signals or metadata required for effective searches. Thus, we need formalized methods and systems to automatically annotate scientific datasets from the data and its surrounding context. Additionally, a search infrastructure needs to account for the scale and rate of application data volumes. In this paper, we present ScienceSearch, a system infrastructure that uses machine learning techniques to capture and learn the knowledge, context, and surrounding artifacts from data to generate metadata and enable search. Our current implementation is focused on a dataset from the National Center for Electron Microscopy (NCEM), an electron microscopy facility at Lawrence Berkeley National Laboratory sponsored by the Department of Energy which supports hundreds of users and stores millions of micrographs. In this paper, we describe a) our search infrastructure and model, b) methods for generating metadata using machine learning techniques, and c) optimizations to improve search latency, and deployment on an HPC system. We demonstrate that ScienceSearch is capable of producing valid metadata for NCEM's dataset and providing low-latency good quality search results over a scientific dataset.

Rodrigo, Gonzalo P.↗

Livewire: Automatic Annotations

Diogenes processes datasets to provide data quality metrics for the Livewire platform and creates standardized data dictionaries from data annotations. Diogenes needs data annotations that clearly outline thenformat and organization of the data. It also relies on the type, class, and unit of each data piece for comprehensive analysis, which it cannot determine independently. The Annotation Tool significantly reduces the time needed to create annotations for Diogenes by generating data annotations with the correct formatting and content. It also employs machine learning and hard-coded models to automatically annotate data class, quality type, and data units.

33 - ADVANCED PROPULSION SYSTEMS↗

Annotation of DOM metabolomes with an ultrahigh resolution mass spectrometry molecular formula library

Current approaches to analyzing metabolomic data often rely on matching MS/MS fragmentation data to sparse libraries or databases. This approach results in limited identification of features, often with less than 10% of the dataset being annotated. A complementary approach is to assign molecular formula to features based on accurate mass measurements, but the platforms commonly used for metabolomics do not have the needed accuracy or resolving power to do this robustly, particularly for larger molecules. Using our newly modified analysis tool, CoreMS, we generated a library of molecular formula from pooled samples analyzed with LC-21T FT-ICR MS. This library successfully annotated approximately 53.2% of features identified from the exometabolome of marine diatom Phaeodactylum tricornutum – a nearly ten-fold increase over the 5.9% annotation rate achieved using a conventional MS/MS library matching approach. Using this FT-ICR MS library approach, we were able to differentiate differences in the exometabolome of P. tricornutum in iron replete and iron limited conditions, with 668 metabolites being differentially expressed (p < 0.05, 2 x intensity difference) under these conditions. The traditional MS/MS fragmentation-based annotation approach only annotated 61 of these metabolites, while our novel pipeline annotated 450 metabolites and revealed 12 metabolites that were significantly more abundant under low iron conditions. Our results demonstrate the utility of ultrahigh resolution mass spectrometry for generating more comprehensive and confident molecular annotations.

21T-FTICR-MS, CoreMS↗

FeGenie: a comprehensive tool for the identification of iron genes and iron gene neighborhoods in genomes and metagenome assemblies

Iron is a micronutrient for nearly all life on Earth. It can be used as an electron donor and electron acceptor by iron-oxidizing and iron-reducing microorganisms and is used in a variety of biological processes, including photosynthesis and respiration. While it is the fourth most abundant metal in the Earth’s crust, iron is often limiting for growth in oxic environments because it is readily oxidized and precipitated. Much of our understanding of how microorganisms compete for and utilize iron is based on laboratory experiments. However, the advent of next-generation sequencing and surge in publicly available sequence data has made it possible to probe the structure and function of microbial communities in the environment. To bridge the gap between our understanding of iron acquisition, iron redox cycling, iron storage, and magnetosome formation in model microorganisms and the plethora of sequence data available from environmental studies, we have created a comprehensive database of hidden Markov models (HMMs) based on genes related to iron acquisition, storage, and reduction/oxidation in Bacteria and Archaea. Along with this database, we present FeGenie, a bioinformatics tool that accepts genome and metagenome assemblies as input and uses our comprehensive HMM database to annotate provided datasets with respect to iron-related genes and gene neighborhood. An important contribution of this tool is the efficient identification of genes involved in iron oxidation and dissimilatory iron reduction, which have been largely overlooked by standard annotation pipelines. We validated FeGenie against a selected set of 28 isolate genomes and showcase its utility in exploring iron genes present in 27 metagenomes, 4 isolate genomes from human oral biofilms, and 17 genomes from candidate organisms, including members of the candidate phyla radiation. We show that FeGenie accurately identifies iron genes in isolates. Furthermore, analysis of metagenomes using FeGenie demonstrates that the iron gene repertoire and abundance of each environment is correlated with iron richness. While this tool will not replace the reliability of culture-dependent analyses of microbial physiology, it provides reliable predictions derived from the most up-to-date genetic markers. FeGenie’s database will be maintained and continually updated as new genes are discovered.

59 BASIC BIOLOGICAL SCIENCES↗

Self-Supervised Cloud Classification

Abstract Low-level marine clouds play a pivotal role in Earth’s weather and climate through their interactions with radiation, heat and moisture transport, and the hydrological cycle. These interactions depend on a range of dynamical and microphysical processes that result in a broad diversity of cloud types and spatial structures, and a comprehensive understanding of cloud morphology is critical for continued improvement of our atmospheric modeling and prediction capabilities moving forward. Deep learning has recently accelerated our ability to study clouds using satellite remote sensing, and machine learning classifiers have enabled detailed studies of cloud morphology. A major limitation of deep learning approaches to this problem, however, is the large number of hand-labeled samples that are required for training. This work applies a recently developed self-supervised learning scheme to train a deep convolutional neural network (CNN) to map marine cloud imagery to vector embeddings that capture information about mesoscale cloud morphology and can be used for satellite image classification. The model is evaluated against existing cloud classification datasets and several use cases are demonstrated, including training cloud classifiers with very few labeled samples, interrogation of the CNN’s learned internal feature representations, cross-instrument application, and resilience against sensor calibration drift and changing scene brightness. The self-supervised approach learns meaningful internal representations of cloud structures and achieves comparable classification accuracy to supervised deep learning methods without the expense of creating large hand-annotated training datasets. Significance Statement Marine clouds heavily influence Earth’s weather and climate, and improved understanding of marine clouds is required to improve our atmospheric modeling capabilities and physical understanding of the atmosphere. Recently, deep learning has emerged as a powerful research tool that can be used to identify and study specific marine cloud types in the vast number of images collected by Earth-observing satellites. While powerful, these approaches require hand-labeling of training data, which is prohibitively time intensive. This study evaluates a recently developed self-supervised deep learning method that does not require human-labeled training data for processing images of clouds. We show that the trained algorithm performs competitively with algorithms trained on hand-labeled data for image classification tasks. We also discuss potential downstream uses and demonstrate some exciting features of the approach including application to multiple satellite instruments, resilience against changing image brightness, and its learned internal representations of cloud types. The self-supervised technique removes one of the major hurdles for applying deep learning to very large atmospheric datasets.

54 ENVIRONMENTAL SCIENCES↗

PlantSegNet: 3D point cloud instance segmentation of nearby plant organs with identical semantics

In this study, we introduce PlantSegNet, a novel neural network model for instance segmentation of nearby objects with similar geometric structures. Our work addresses the challenges of instance segmentation of plant point clouds, including the difficulty of annotating and labeling point clouds, the loss of local structural information in neural network components, and the generation of large numbers of incorrect small clusters due to poor choices of the loss function. One of the key contributions of our approach is a digital twin of sorghum, i.e., a procedural sorghum model, which was used to generate point clouds of sorghum fields. This allowed us to create a large-scale, annotated, synthetic dataset of sorghum plants that we used to train our PlantSegNet model. We demonstrated the effectiveness of our method in segmenting instances of sorghum leaves grown in outdoor field settings. To the best of our knowledge, this is the first study to address this specific instance segmentation problem for plants grown in such a setting. We compared our proposed method with other state-of-the-art methods for indoor settings, including SGPN and TreePartNet, on both synthetic and real data. Furthermore, our results show that PlantSegNet outperforms these methods regarding accuracy, robustness, and efficiency.

97 MATHEMATICS AND COMPUTING↗

METAREP: JCVI metagenomics reports—an open source tool for high-performance comparative metagenomics

JCVI Metagenomics Reports (METAREP) is a Web 2.0 application designed to help scientists analyze and compare annotated metagenomics datasets. It utilizes Solr/Lucene, a high performance scalable search engine, to quickly query large data collections. Furthermore, users can use its SQL-like query syntax to filter and refine datasets. METAREP provides graphical summaries for top taxonomic and functional classifications as well as a GO, NCBI Taxonomy and KEGG Pathway Browser. Users can compare absolute and relative counts of multiple datasets at various functional and taxonomic levels. Advanced comparative features comprise statistical tests as well as multidimensional scaling, heatmap and hierarchical clustering plots. Summaries can be exported as tab-delimited files, publication quality plots in PDF format. A data management layer allows collaborative data analysis and result sharing.

59 BASIC BIOLOGICAL SCIENCES↗

Identifying Disinformation Using Rhetorical Devices in Natural Language Models

Foreign disinformation campaigns are strategically organized, extended efforts using disinformation – false or misleading information deliberately placed by an adversary – to achieve some goal. Disinformation campaigns pose severe threats to our nation’s security by misinforming decision makers and negatively influencing their actions when they are operating on limited amounts of evidence. Current efforts rely on subject matter experts to manually identify disinformation, or on computers and traditional natural language processing algorithms to identify patterns in data to calculate the probability that something is disinformation or not. While both have their merits and successes, subject matter experts are unable to keep up with the high volumes of global information and traditional natural language algorithms do not do well in identifying “why” something is disinformation or not. Our hypothesis is that we can identify disinformation by looking at the way someone speaks, in the rhetorical devices they use. We have curated and annotated a dataset designed for multiple natural language processing tasks, but specifically useful for disinformation detection algorithms.

97 MATHEMATICS AND COMPUTING↗

2025 TEM Workshop

The TEM Data Management Workshop will take place on August 26 from 9 a.m. to 12 p.m. MT, and will be held virtually on TEAMS. The primary goal of this workshop is to engage NSUF users and stakeholders in discussions about the data needs for the utilization of AI and ML in the analysis of TEM data. Key topics to be covered include data storage, data sharing, data tagging, metadata inclusion, standardized data formats, data augmentation, and annotated training datasets. Additionally, the workshop will provide valuable insights into resources such as the Nuclear Research Data System (NRDS) for data storage and sharing, as well as open-source codes for data analysis.

Bachhav, Mukesh↗