Search NASA⌕ Search

Engineering topics

Yang, Lexie

Publications and source records attributed to Yang, Lexie.

At least 19 records

Multivariate Testing of Sampling Techniques to Address Class Imbalance in Building Use Type Classification

This study addresses the challenges inherent in building use type classification, particularly focusing on the issue of class imbalance in the training datasets for machine learning classifiers. We comprehensively analyze the efficacy of various class-balancing sampling techniques. Employing Monte Carlo simulations and Bayesian optimization, we evaluated the performance of multiple sampling methods, including Random Oversampling, Random Undersampling, SMOTE, Borderline-SMOTE, and ADASYN, across a dataset encompassing nine southeastern coastal states of the United States. Our findings reveal that simple random over- and undersampling techniques outperform more sophisticated methods. Additionally, we show inherent value in creating an imbalance in training data to effectively train a machine learning classifier for distinguishing between residential and nonresidential buildings. This study provides valuable guidance for future research on building use type classification research and lays essential groundwork for developing attribute-rich building stock datasets.

Adams, Daniel↗

OReole-FM: successes and challenges toward billion-parameter foundation models for high-resolution satellite imagery

While the pretraining of Foundation Models (FMs) for remote sensing (RS) imagery is on the rise, models remain restricted to a few hundred million parameters. Scaling models to billions of parameters has been shown to yield unprecedented benefits including emergent abilities, but requires data scaling and computing resources typically not available outside industry R&D labs. In this work, we pair high-performance computing resources including Frontier supercomputer, America's first exascale system, and high-resolution optical RS data to pretrain billion-scale FMs. Our study assesses performance of different pretrained variants of vision Transformers across image classification, semantic segmentation and object detection benchmarks, which highlight the importance of data scaling for effective model scaling. Moreover, we discuss construction of a novel TIU pretraining dataset, model initialization, with data and pretrained models intended for public release. By discussing technical challenges and details often lacking in the related literature, this work is intended to offer best practices to the geospatial community toward efficient training and benchmarking of larger FMs.

Ambrozio Dias, Philipe↗

Efficient Extraction Of Building Elevation Attributes For Flood Risk Management Using Airborne LiDAR Data

In this paper, we address the need for extracting two key building elevation attributes—Lowest Adjacent Grade (LAG) and Highest Adjacent Grade (HAG)—which are crucial for effective flood risk management. Conventional methods, involving onsite surveying or the use of optical imagery-derived building footprints combined with Digital Elevation Models (DEMs), often face misalignment and time discrepancy issues due to varied remote sensing sources. We introduce a new, scalable method that exclusively relies on airborne LiDAR data to overcome these challenges. Our approach employs an object-based ground filtering technique, and the results were evaluated using two different DEMs and building footprint sets. The findings demonstrate that our single-source method, utilizing only airborne LiDAR data, significantly improves the accuracy of LAG and HAG calculations compared to traditional methods that use hand-digitized building footprints. The proposed approach offers a solution for comprehensive flood risk management endeavors.

Song, Hunsoo↗

Towards Geospatial Knowledge Graph Infused Neuro-Symbolic AI for Remote Sensing Scene Understanding

Deep learning has proven its effectiveness in numerous tasks for remote sensing scene understanding. However there is an increasing interest to explore fusion of domain-specific background information to the deep neural network to further improve its performance. Remote sensing researchers are also working towards developing models that generalize and adapt to multiple applications. Generalization challenges coupled with the scarcity of large corpora of high-quality noise-free labelled data, have together fueled an interest for leveraging background information. Knowledge graphs serve as excellent choice to represent domain-specific information in a structured, standardized and extensible manner. Integrating symbolic knowledge representations in the form of Knowledge Graph Embedding (KGE) to perform neuro-symbolic reasoning is an emerging research direction promising significant impacts. This vision paper seeks to position ideas and provoke early thoughts toward advancing neuro-symbolic artificial intelligence in the context of geospatial challenges. Specifically, it conceptualizes and elaborates on an architecture for infusing geospatial knowledge from knowledge graph in a deep neural network pipeline. As guiding case studies - land-use land-cover classification, object detection and instance segmentation can benefit from infusing spatio-contextual information with remote sensing imagery. The discussion further reflects on and articulates the challenges and explainable AI opportunities anticipated when scaling and maintaining large-scale geospatial knowledge graphs.

Potnis, Abhishek↗

Scaling Automatic Vector Data Alignment to Satellite Imagery

Given the tremendous volume of accessible Earth Observation (EO) data, there is a need to develop scalable Geospatial Artificial Intelligence (GeoAI) solutions for time-sensitive applications. Scalability in this context refers to rapidly processing large-scale EO data using high performance computing resources. Accurate mapping of the built environment from remote sensing (RS) imagery has been one of the crucial components in GeoAI workflows for a wide spectrum of humanitarian applications. Derived vector data of built environment is often leveraged for disaster preparedness and response activities. However, factors such as differences in ortho-rectification, atmospheric conditions and human error, results in spatial misalignment between vector data and the timely available RS imagery. Model training for downstream tasks such as object detection, change analysis, etc., is negatively impacted due to such spatial misalignment. Although there has been progress towards automatic alignment of vector data, the lack of scalability remains an open research challenge. This paper proposes to leverage parallel computing to optimize an automatic vector data alignment workflow. It further employs CPU-level multi-core parallelism for improving the performance of the workflow for scalable built environment mapping. We report observations and discuss findings from the preliminary experiments performed on the Summit Supercomputer.

Potnis, Abhishek↗

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↗

USA Structures Phase 2 Technical Report

Spatially accurate data of critical infrastructures are essential to effective disaster preparedness, response, and recovery. Precise location and building outlines provide the most accurate data for characterizing impacts of hazards and effectively serve response, recovery, and mitigation efforts, as well as the people affected by the disaster. Since 2017, Oak Ridge National Laboratory (ORNL) has partnered with the Federal Emergency Management Agency (FEMA) to establish a comprehensive and open source national database of building footprints called USA Structures. Several key attributions have been added to the dataset to support rapid disaster response. In the most recent update to the dataset, ORNL developed two additional attributions to the structures, leveraging several authoritative data sources.

42 ENGINEERING↗

Providing Geospatial Intelligence through a Scalable Imagery Pipeline

This chapter describes ORNL’s (Oak Ridge National Laboratory’s) contributions to imagery preprocessing for geospatial intelligence research and development (R&D) in four sections. First, we discuss challenges involved in building an effective imagery preprocessing workflow and the world-class high-performance computing (HPC) resources at ORNL available to process petabytes of imagery data. Second, we highlight how we developed imagery preprocessing tools over three decades while paving the way for our current cutting-edge machine learning and computer vision algorithms that are impacting humanitarian and disaster response efforts. Third, we discuss how PIPE modules work together to turn raw images into analysis-ready datasets. Fourth, we look toward the future and discuss planned advancements to PIPE and computing trends that will affect geospatial intelligence R&D.

Reith, Andrew↗

UAS Edge Computing of Energy Infrastructure Damage Assessment

Energy infrastructure assessments are needed within 72 hours of natural disasters, and previous data collection methods have proven too slow. We demonstrate a scalable end-to-end solution using a prototype unmanned aerial system that performs on-the-edge detection, classification (i.e., damaged or undamaged), and geo-location of utility poles. The prototype is suitable for disaster response because it requires no local communication infrastructure and is capable of autonomous missions. Collections before, during, and after Hurricane Ida in 2021 were used to test the system. The system delivered an F1 score of 0.65 operating with a 2.7 s/frame processing speed with the YOLOv5 large model and an F1 score of 0.55 with a 0.48 s/frame with the YOLOv5 small model. Geo-location uncertainty in the bottom half of the frame was ~8 m, mostly driven by error in camera pointing measurement. With additional training data to improve performance and detect additional types of features, a fleet of similar drones could autonomously collect actionable post-disaster data.

24 POWER TRANSMISSION AND DISTRIBUTION↗

2021 GeoAI Workshop Report: The Trillion Pixel Challenge

The convergence of geospatial big data with advancements from artificial intelligence, cloud infrastructure, and high-performance computing continues to revolutionize mapping and analysis of Earth's surface in unprecedented detail. Rapid innovations in sensing technologies will soon collect geospatial data in even higher resolution and throughput. These developments offer the potential for breakthroughs in science, policy, and national security via end-to-end GeoAI systems that can provide fresh insights into how humans occupy and alter their environment over time. At the 2021 GeoAI Trillion Pixel workshop, international subject matter experts from government, academia, industry, and nonprofit organizations gathered virtually to discuss the Trillion Pixel GeoAI Challenge. The event focused on six major themes currently influencing scientific innovation and breakthroughs. Particular focus was paid to societal impacts. As an additional takeaway message, the gathering identified remaining application gaps and challenges that are in need of stronger community partnerships and collaborations.

58 GEOSCIENCES↗

Few-shot Learning for Post-disaster Structure Damage Assessment

Automating post-disaster damage assessment with remote sensing data is critical for faster surveys of structures impacted by natural disasters. One significant obstacle to training state-of-the-art deep neural networks to support this automation is that large quantities of labelled data are often required. However, obtaining those labels is particularly unrealistic to support post-disaster damage assessment in a timely manner. Few-shot learning methods could help to mitigate this by reducing the amount of labelled data required to successfully train a model while achieving satisfactory results. To this end, we explore a feature reweighting method to the YOLOv3 object detection architecture to achieve few-shot learning of damage assessment models on the xBD dataset. Our results show that the feature reweighting approach yield improved mAP over the baseline with significantly fewer labelled samples. In addition, we use t-SNE to analyze the class-specific reweighting vectors generated by the reweighting module in order to evaluate their inter-class and intra-class similarity. We find that the vectors form clusters based on class, and that these clusters overlap with visually similar classes. Those results show the potential to employ this few-shot learning strategy for rapid damage assessment with post-event remote sensing images.

Bowman, Jordan↗

Entropy and Boundary Based Adversarial Learning for Large Scale Unsupervised Domain Adaptation

Supervised semantic segmentation methods provide state-of-the-art performance, but their performance is limited by the amount of quality labeled data they need for training. Scarcity of labeled data and non-transferablity of models, due to cross-domain discrepancy makes it a bigger challenge for remote sensing imagery analysis. In this work, we approach this problem through adversarial learning, driven by entropy and boundary of region-of-interest for unsupervised domain adaptation. This concept helps with better boundary prediction and encourages target domain entropy maps (probability/uncertainty maps) to be similar to source domains. In particular, we showed that deriving informative entropy through the adversarial learning is essential to enable the adaptation. We used a large scale cross country building extraction dataset to validate the framework. The experimental results show the usefulness of considering boundary and entropy driven adversarial learning for adaptation.

Makkar, Nikhil↗

Mapping Human Dynamics

Researchers at Oak Ridge National Laboratory are mapping the global footprints of human activity with unprecedented spatiotemporal resolution. With a global population now approaching 8 billion people, this herculean effort demands advanced machine learning, artificial intelligence, and one of the world’s fastest supercomputers.

Rose, Amy↗

A Fully Automatic Method for Rapidly Mapping Impacted Area by Natural Disaster

Deep learning based change detection methods have achieved the state-of-the-art performance in several recent studies. However, such methods usually are supervised, and therefore a large number of training samples is often a requisite. Manually preparing those training samples is not only expensive but also time-consuming, which does not fit the need of rapidly mapping the impacted area caused by nature disaster for further rescue mission and damage assessment. In this study, a fully automatic method was proposed to address the issue by automating training sample generation for mapping the impacted area caused by nature disaster. We used the 2011 tornado event in Joplin, Missouri, US, as an example of its application. The generated impacted area map was both visually and quantitatively evaluated against the ground truth data collected by US Federal Emergency Management Agency (FEMA). The results show that the map matches well with the FEMA ground truth data with 86% of major-damaged and destroyed buildings identified by FEMA on the ground also detected by this fully automatic framework using very high resolution (VHR) satellite images.

Liu, Tao↗

RAPID STRUCTURE DETECTION IN SUPPORT OF DISASTER RESPONSE : A CASE STUDY OF THE 2018 KILAUEA VOLCANO ERUPTION

Disaster response requires timely damage assessment to prioritize rescue and restoration resources. However, providing critical and actionable knowledge after a natural disaster can be challenging due to the scale and the type of damages. This paper describes how remote sensing and machine learning techniques can be used to support rapid structure detection in the wake of a disaster. We use high resolution satellite imagery to identify structures on Hawaii’s Big Island to support the Federal Emergency Management Agency’s response efforts during the 2018 K¯ilauea lava flow incident. This framework specifically showcases the generalizability of CNN models with no need to collect additional training samples to quickly map structures in pre- and post-event imagery and provide timely information to assist government agencies evaluating the extent and potential loss of disaster. With this case study, we further point out future directions to benefit similar larger scale efforts based on the lessons learned.

Laverdiere, Melanie↗

The Trillion Pixel GeoAI Challenge Workshop

Rapid innovations in satellite and airborne remote sensing capabilities holds the promise of collecting high-resolution imagery with daily, even hourly cadence across the entire planet. Availability of such earth observation data streams at varying spatial and temporal scales, coupled with astonishing progress in AI and transformational advances in high performance computing bring into view the possibility of mapping and interpreting the surface of our planet at unprecedented detail. The implications of such capability for science, technology, policy, and security are far reaching. Moreover, even at a modest 5m resolution, 100 trillion pixels will describe the surface of the Earth every day. Consequently, monitoring the pulse of our planet will entail collecting, refining, analyzing, and curating those 100 trillion pixels within a 24-hour period. That is a challenge with unknown solutions as of today’s advances and therefore serves as a key motivation for this workshop. The GEOINT and AI communities have a unique opportunity to enable detailed monitoring the pulse of our planet and obtain new insights into how humans occupy, interact with, and alter the surface of the Earth over time. From a science and technology standpoint, this is a modern-day moonshot that presents numerous challenges particularly related to scaling the analysis to every pixel covering the planet. The challenge of making sense of all these pixels in a timely fashion include image processing at scale, selective analysis of only the pixels that matter, generalizing AI models across heterogeneous locations, computing on hardware that is memory constrained, and deploying automated feature extraction on edge devices. The designing of capable GeoAI workflows and data pipelines will require interdisciplinary efforts and partnerships. The Trillion Pixel Challenge for GeoAI workshop was is promoted as a first community effort to discuss priority research and development directions and to establish the partnerships and collaborations needed to make progress on critical scientific and operational applications from local to regional to planet scale.

97 MATHEMATICS AND COMPUTING↗

Towards Misregistration-Tolerant Change Detection using Deep Learning Techniques with Object-Based Image Analysis

Co-registrating is a common pre-processing step for existing change detection algorithms, but registering bi-temporal images is nontrivial. The use of image patch as input for deep learning techniques provides a natural avenue to apply them in the OBIA framework, and have shown successful performance in the object-based land cover mapping and change detection applications. Even though attempts of applying deep learning techniques for change detection applications have been made with varying success, its application under OBIA framework for change detection have not been conducted and its tolerance for misregistration among temporal images are neither known. This study performed change detection under OBIA framework using deep learning techniques for the first time, and evaluated its performance regarding their tolerance of image misregistration on training and testing dataset. Our results demonstrate the proposed change detection scheme is surprisingly robust to image misregistration on the testing dataset, while classifiers trained with the training dataset containing image misregistration errors suffer from slight decrease of overall accuracy.

Liu, Tao↗