Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “remote training”

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
NASA NTRS2010

High Resolution ECG for Evaluation of Heart Function During Exposure to Subacute Hypobaric Hypoxia

High altitude climbing presents a wide spectrum of health risks, including exposure to hypobaric hypoxia. Risks are also typically exacerbated by the difficulty in appropriately monitoring for early signs of organ dysfunction in remote areas. We investigated whether high resolution advanced ECG analysis might be helpful as a non-invasive and easy-to-use tool (e.g., instead of Doppler echocardiography) for evaluating early signs of heart overload in hypobaric hypoxia. Nine non-acclimatized healthy trained alpine rescuers (age 43.7 plus or minus 7.3 years) climbed in four days to the altitude of 4,200 m on Mount Ararat. Five-minute high-resolution 12-lead electrocardiograms (ECGs) were recorded (Cardiosoft) in each subject at rest in the supine position on different days but at the same time of day at four different altitudes: 400 m (reference altitude), 1,700 m, 3,200 m and 4,200 m. Changes in conventional and advanced resting ECG parameters, including in beat-to-beat QT and RR variability, waveform complexity, signal-averaged, high-frequency and spatial/spatiotemporal ECG was estimated by calculation of the regression coefficients in independent linear regression models. A p-value of less than 0.05 was adopted as statistically significant. As expected, the RR interval and its variability both decreased with increasing altitude, with trends k = -96 ms/1000 m with p = 0.000 and k = -9 ms/1000 m with p = 0.001, respectively. Significant changes were found in P-wave amplitude, which nearly doubled from the lowest to the highest altitude (k = 41.6 microvolt/1000 m with p = 0.000), and nearly significant changes in P-wave duration (k = 2.9 ms/1000 m with p = 0.059). Changes were less significant or non-significant in other studied parameters including those of waveform complexity, signal-averaged, high-frequency and spatial/spatiotemporal ECG. High resolution ECG analysis, particularly of the P wave, shows promise as a tool for monitoring early changes in heart function due to exposure to high altitude.

Zupet, Petra↗
DOE OSTI2025

Interactive Rotated Object Detection for Novel Class Detection in Remotely Sensed Imagery

In this paper we propose IRTR-DETR an Interactive and Real-Time Rotated DEtection TRansformer that extends IRTDETR to predict rotated bounding boxes. IRTR-DETR maintains the Human-In-The-Loop (HIL) workflow of IRTDETR but introduces rotation-aware heads for improved detection of objects with arbitrary orientations. Similarly to IRTDETR IRTR-DETR can be trained with a small labeled sample set in an interactive setting but we show that it can also be pretrained on related but not identical data--such as a building damage dataset--before being applied to tasks like identifying buildings under construction. We demonstrate the efficacy of our approach on the publicly available Tiny-DOTA and xBD dataset as well as two study-cases on proprietary datasets of greenhouses and houses under construction ("waffle homes"). Detecting greenhouses is highly relevant in the context of damage assessment while "waffle homes" aid understanding typical floorplans and building codes in different areas both thereby supporting population modeling emergency response and policy planning. Our method outperforms the state of the art in interactive rotated object detection on the Tiny-DOTA dataset by 5.7 percent and improves upon the non interactive RTDETR by 7.85 to 19.39 percent (depending on the number of provided samples) while maintaining its real-time efficiency.

Burges, Marvin [ORNL] (ORCID:0000000312690769)↗
DOE OSTI2025

SIDDA: SInkhorn Dynamic Domain Adaptation for image classification with equivariant neural networks

Modern neural networks (NNs) often do not generalize well in the presence of a ‘covariate shift’; that is, in situations where the training and test data distributions differ, but the conditional distribution of classification labels given the data remains unchanged. In such cases, NN generalization can be reduced to a problem of learning more robust, domain-invariant features. Domain adaptation (DA) methods include a broad range of techniques aimed at achieving this; however, these methods have struggled with the need for extensive hyperparameter tuning, which then incurs significant computational costs. In this work, we introduce SInkhorn Dynamic Domain Adaptation (SIDDA), an out-of-the-box DA training algorithm built upon the Sinkhorn divergence, that can achieve effective domain alignment with minimal hyperparameter tuning and computational overhead. We demonstrate the efficacy of our method on multiple simulated and real datasets of varying complexity, including simple shapes, handwritten digits, real astronomical observations, and remote sensing data. These datasets exhibit covariate shifts due to noise, blurring, differences between telescopes, and variations in imaging wavelengths. SIDDA is compatible with a variety of NN architectures, and it works particularly well in improving classification accuracy and model calibration when paired with symmetry-aware equivariant NNs (ENNs). We find that SIDDA consistently enhances the generalization capabilities of NNs, achieving up to a ${\approx}40\%$ improvement in classification accuracy on unlabeled target data, while also providing a more modest performance gain of $\lesssim 1\%$ on labeled source data. We also study the efficacy of DA on ENNs with respect to the varying group orders of the dihedral group DN, and find that the model performance improves as the degree of equivariance increases. Finally, if SIDDA achieves proper domain alignment, it also enhances model calibration on both source and target data, with the most significant gains in the unlabeled target domain—achieving over an order of magnitude improvement in the expected calibration error and Brier score. SIDDA’s versatility across various NN models and datasets, combined with its automated approach to domain alignment, has the potential to significantly advance multi-dataset studies by enabling the development of highly generalizable models.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1979

Visual observations over oceans

Important factors in locating, identifying, describing, and photographing ocean features from space are presented. On the basis of crew comments and other findings, the following recommendations can be made for Earth observations on Space Shuttle missions: (1) flyover exercises must include observations and photography of both temperate and tropical/subtropical waters; (2) sunglint must be included during some observations of ocean features; (3) imaging remote sensors should be used together with conventional photographic systems to document visual observations; (4) greater consideration must be given to scheduling earth observation targets likely to be obscured by clouds; and (5) an annotated photographic compilation of ocean features can be used as a training aid before the mission and as a reference book during space flight.

Terry, R. D.↗
NASA NTRS2007

ER-2: Flying Laboratory for Earth Science Studies

The National Aeronautics and Space Administration (NASA) Dryden Flight Research Center (DFRC), (Edwards, California, USA) has two Lockheed Martin Corporation (Bethesda, Maryland) Earth Research-2 (ER-2) aircraft that serve as high-altitude and long-range flying laboratories. The ER-2 has been utilized to conduct scientific studies of stratospheric and tropospheric chemistry, land-use mapping, disaster assessment, preliminary testing and calibration and validation of satellite sensors. The ER-2 aircraft provides experimenters with a wide array of payload accommodation areas with suitable environment control with required electrical and mechanical interfaces. Missions may be flown out of DFRC or from remote bases worldwide. The NASA ER-2 is utilized by a variety of customers, including U.S. Government agencies, civilian organizations, universities, and state governments. The combination of the ER-2 s range, endurance, altitude, payload power, payload volume and payload weight capabilities complemented by a trained maintenance and operations team provides an excellent and unique platform system to the science community.

Navarro, Robert↗
NASA NTRS2021

Modernizing Mechatronics Course With Quantum Engineering

Mechatronics is the synergistic application of mechanics, electronics, control engineering, and computer science in the development of electromechanical products and systems, through integrated design. This paper proposes to extend the mechatronics course beyond traditional engineering topics, and to modernize the mechatronics instructions with complementary quantum engineering topics. With the recent rapid advances in quantum technologies such as quantum communications, sensing, computers, and algorithms, it is imperative to train the next generation of engineers and prepare them for their future careers in the ever-changing industry in such areas. Furthermore, due to such progress and advances in the fields associated with quantum mechanics, the integration of quantum technologies with classical mechanical systems will be inevitable both in terms of educational and technological standpoints in future. To address the educational needs of the future engineers in such areas of significant importance, quantum entanglement and quantum cryptography experiments, as two fundamental topics in quantum mechanics, are brought into the mechatronics course in an initiative that is reported in this paper. The integrated quantum and mechatronics topics also provides opportunities for open discussions on exploring the interface of quantum technologies and classical engineering systems, which can potentially push the engineering boundaries beyond classical possibilities by accessing the quantum advantages. An innovative online remote demonstration of such quantum experiments are developed and presented to the students. This course has been offered to undergraduate students once with successful results. The students were able to remotely access the experiments, perform the experiments and collect data. The successful result of such quantum experiments is also reflected in a course survey, presented in this paper, even though the quantum mechanics topics offered in this course are unfamiliar to engineering students and hence more challenging. The paper reports, and aims to promote, the integration of selected quantum technology topics with the mechatronics course for training engineering students in this rapidly growing area.

Ghazinejad, Maziar↗
NASA NTRS2021

A Semi-Empirical Scheme for Bathymetric Mapping in Shallow Water by ICESat-2 and Sentinel-2: A Case Study in the South China Sea

To derive shallow water bathymetry for coastal areas, a common approach is to deploy a scanning airborne bathymetric light detection and ranging (LiDAR) system or a shipborne echosounder for ground surveys. However, recent advancements in satellite remote sensing, including the Ice, Cloud and land Elevation Satellite-2 (ICESat-2) offer new tools for generating satellite derived bathymetry (SDB). The key payload onboard ICESat-2 is the Advanced Topographic Laser Altimeter System (ATLAS), a micro-pulse, photon-counting LiDAR system, simultaneously emitting six separate 532 nm beams at 10 kHz pulse rate. However, despite its high resolution, the major limitation for bathymetry is that ICESat-2 only provides along-track height profiles, leaving observation gaps between the parallel ground tracks. Merging ICESat-2 observations with optical multispectral imagery, as demonstrated herein, provides an effective solution for deriving a full scene of water depth in light of the spectral attenuation behavior. This study aims to combine ICESat-2 and Sentinel-2 optical data to derive shallow water bathymetry (depth <20 m) at six islands and reefs in the South China Sea. ICESat-2 ATL03 point clouds of georeferenced photons are first filtered to determine the seafloor elevation along the ground track. Results indicate a root-mean-square error (RMSE) of 0.26–0.61 m as compared with independent observations from an airborne LiDAR campaign. Next, three semi-empirical functions, namely the Modified Linear/Polynomial/Exponential Ratio Models with its kernel formed by the log ratio between Sentinel-2′s green and blue bands, are used to fit the spectral data with ICESat-2 height profiles. After water depth mapping using the trained model, independent ICESat-2 point clouds are used to validate the Sentinel-2 derived bathymetry. The RMSE values of the three models using the weighted average of multiple images for these six islands are within 0.50–0.90 m in 0–15 m deep. We also demonstrate that a synthesis of satellite laser altimetry and optical remote sensing can produce SDB results that potentially meet the requirement of category C in Zones of Confidence (ZOC) of the Electronic Navigational Chart (ENC) in 0–8 m deep. It is foreseen that ICESat-2 will be a helpful tool for mapping coastal and shallow waters around the world especially where bathymetric data are unavailable.

Coastal Bathymetry↗
NASA NTRSDate not supplied

In Situ Water Quality Data for the Chesapeake Bay

This paper examines in situ water quality datameasured during2020-2021in the Chesapeake Bay for comparison with optical satellite data. Thiscollection was performed as part of a NASA project aiming to develop new methods for water quality monitoring from satellite remote sensingusing artificial intelligence. Our objective is to use insitu data as ground-truth to provide water quality classifications, or labels,to their overlapping (in time and location)satellite imagery. Having such labeled data, can help us achieve our project’s longer-termgoal:to train artificial intelligencemodelsto recognize features in spectral informationfor monitoringwater qualityfrom satellites. Because routine monitoring by state agencies is conducted at discrete locations, we obtained a flow-through system operated from small boats to measure waterquality parameters along transects for comparison with two-dimensional maps collected from space, with an initial focus on low oxygenevents, due to their large spatial extent and regular occurrence each summer.We also evaluated similar in situ data collected during 1984-2021by the Chesapeake Program.

Nargess Memarsadeghi↗
DOE OSTI2023

Data, scripts, and figures associated with a manuscript studying impact of climate and topography on post-fire vegetation recovery.

This data package is associated with the publication “Impact of Topography and Climate on Post-fire Vegetation Recovery Across Different Burn Severity and Land Cover Types through Machine Learning” submitted to Remote Sensing of Environment (Zahura et al. 2023). In this research, a machine learning algorithm, random forest (RF), was utilized to examine the impact of climate and topography on post-fire vegetation recovery. We used enhanced vegetation index (EVI) to examine varying burn severity and land cover types. The data package includes the input files for RF model training, outputs from model predictions and analysis, and python scripts to run the model, analyze the results to understand model performance and interpretability, and plot manuscript figures. This data package contains three folders (Data, Scripts, and Figures), a file-level metadata (FLMD) csv, and a data dictionary (dd) csv. Please see Postfire_recovery_flmd.csv for a list of all files contained in this data package and descriptions for each. The data dictionary (Postfire_recovery_dd.csv) describes the csv column headers. The “Data” folder provides all the inputs and outputs to train the RF model, evaluate performance, and interpret predictions. The “Scripts” folder contains python scripts and jupyter notebooks for model training and result analysis. The “Figures” folder includes the figures used in the manuscript in “.png” and “.jpg” format.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1985

Display comparison for six-degree-of-freedom force/torque control

The experimental design and the results of a study of a force and torque sensing and display (FTSD) system, developed at the Jet Propulsion Laboratory, are presented. The FTSD system was incorporated into the one-g version of the Space Shuttle Remote Manipulator System (RMS). The tests consisted of evaluating alternative graphics formats for presenting force and torque information at the RMS work station, and of testing the utility of the sensor and the display system as an aid to specific RMS tasks. The participating subjects were all trained RMS operators with experience on the one-g version of the system, three of them with previous experience using the Space Shuttle RMS on orbit. The results support the use of the FTSD system as an aid to RMS operations: the system was found to facilitate the control of forces when the RMS is under partially or wholly constrained conditions. The graphics format employed in the present version of the FTSD system was found to be superior to earlier formats.

Baggen, A.↗
NASA NTRS2002

STS-111 Flight Day 6 Highlights

The sixth day of STS-111 flight begins with a live shot of the Space Shuttle Endeavor. Dan Bursch Flight Engineer is shown inside the U.S. Lab Destiny. Astronaut Peggy Whitson and Flight Engineer Carl Walz installed the Mobile Remote Servicer Base System (MBS). The MBS latch is attached to the bar of the Mobile Transporter of the International Space Station. A transfer of supplies is shown inside the U.S. Lab Destiny. Mission Specialists Philippe Perrin and Commander Ken Cockrel talk to the media about EVA training.

Source record↗
NASA NTRS2006

Are You Talking to Me? Dialogue Systems Supporting Mixed Teams of Humans and Robots

This position paper describes an approach to building spoken dialogue systems for environments containing multiple human speakers and hearers, and multiple robotic speakers and hearers. We address the issue, for robotic hearers, of whether the speech they hear is intended for them, or more likely to be intended for some other hearer. We will describe data collected during a series of experiments involving teams of multiple human and robots (and other software participants), and some preliminary results for distinguishing robot-directed speech from human-directed speech. The domain of these experiments is Mars-analogue planetary exploration. These Mars-analogue field studies involve two subjects in simulated planetary space suits doing geological exploration with the help of 1-2 robots, supporting software agents, a habitat communicator and links to a remote science team. The two subjects are performing a task (geological exploration) which requires them to speak with each other while also speaking with their assistants. The technique used here is to use a probabilistic context-free grammar language model in the speech recognizer that is trained on prior robot-directed speech. Intuitively, the recognizer will give higher confidence to an utterance if it is similar to utterances that have been directed to the robot in the past.

Dowding, John↗
NASA NTRS2008

Absorption in Extended Inhomogeneous Clouds

The launch of several different sensors, including CloudSat, into the A-train constellation of satellites allows us for the first time to compute absorption that can occur in realistic vertically inhomogeneous clouds including multiple cloud decks. CloudSat data show that these situations are common. Therefore, understanding vertically inhomogeneous clouds is important from both climate and satellite atmospheric composition remote sensing perspectives. Satellite passive sensors that operate from the near IR to the UV often rely on radiative cloud pressures derived from absorption in oxygen bands (A, B, gamma, or O2-O2 bands) or from rotational-Raman scattering in order to retrieve information about atmospheric trace gases. The radiative cloud pressure is distinct from the physical cloud top derived from thermal infrared measurements. Therefore, the combination of information from different passive sensors yields some information about the cloud vertical profile. When either or both the clouds or atmospheric absorbers (trace gases and aerosols) are vertically inhomogeneous, the use of an effective cloud pressure derived from these approaches may lead to errors. Here, we focus on several scenarios (deep convective clouds and distinct two layer clouds) based on realistic cloud optical depth vertical profiles derived from the CloudSatfMODIS combination. We focus on implications for trace-gas column amount retrievals (specifically ozone and NO2) and derived surface UV irradiance from the Ozone Monitoring Instrument (OMI) on the Atrain Aura platform.

Joiner, Joanna↗
NASA NTRS2016

Automated Segmentation and Classification of Coral using Fluid Lensing from Unmanned Airborne Platforms

In recent years, there has been a growing interest among biologists in monitoring the short and long term health of the world's coral reefs. The environmental impact of climate change poses a growing threat to these biologically diverse and fragile ecosystems, prompting scientists to use remote sensing platforms and computer vision algorithms to analyze shallow marine systems. In this study, we present a novel method for performing coral segmentation and classification from aerial data collected from small unmanned aerial vehicles (sUAV). Our method uses Fluid Lensing algorithms to remove and exploit strong optical distortions created along the air-fluid boundary to produce cm-scale resolution imagery of the ocean floor at depths up to 5 meters. A 3D model of the reef is reconstructed using structure from motion (SFM) algorithms, and the associated depth information is combined with multidimensional maximum a posteriori (MAP) estimation to separate organic from inorganic material and classify coral morphologies in the Fluid-Lensed transects. In this study, MAP estimation is performed using a set of manually classified 100 x 100 pixel training images to determine the most probable coral classification within an interrogated region of interest. Aerial footage of a coral reef was captured off the coast of American Samoa and used to test our proposed method. 90 x 20 meter transects of the Samoan coastline undergo automated classification and are manually segmented by a marine biologist for comparison, leading to success rates as high as 85%. This method has broad applications for coastal remote sensing, and will provide marine biologists access to large swaths of high resolution, segmented coral imagery.

algorithms segments classifications↗
NASA NTRS2015

Automated Segmentation and Classification of Coral Using Fluid Lensing from Unmanned Airborne Platforms

In recent years, there has been a growing interest among biologists in monitoring the short and long term health of the worlds coral reefs. The environmental impact of climate change poses a growing threat to these biologically diverse and fragile ecosystems, prompting scientists to use remote sensing platforms and computer vision algorithms to analyze shallow marine systems. In this study, we present a novel method for performing coral segmentation and classification from aerial data collected from small unmanned aerial vehicles (sUAV). Our method uses Fluid Lensing algorithms to remove and exploit strong optical distortions created along the air-fluid boundary to produce cm-scale resolution imagery of the ocean floor at depths up to 5 meters. A 3D model of the reef is reconstructed using structure from motion (SFM) algorithms, and the associated depth information is combined with multidimensional maximum a posteriori (MAP) estimation to separate organic from inorganic material and classify coral morphologies in the Fluid-Lensed transects. In this study, MAP estimation is performed using a set of manually classified 100 x 100 pixel training images to determine the most probable coral classification within an interrogated region of interest. Aerial footage of a coral reef was captured off the coast of American Samoa and used to test our proposed method. 90 x 20 meter transects of the Samoan coastline undergo automated classification and are manually segmented by a marine biologist for comparison, leading to success rates as high as 85. This method has broad applications for coastal remote sensing, and will provide marine biologists access to large swaths of high resolution, segmented coral imagery.

algorithms segments classifications↗
NASA NTRS2021

A Chlorophyll-a Algorithm for Landsat-8 Based on Mixture Density Networks

Retrieval of aquatic biogeochemical variables, such as the near-surface concentration of chlorophyll-a (Chla) in inland and coastal waters via remote observations, has long been regarded as a challenging task. This manuscript applies Mixture Density Networks (MDN) that use the visible spectral bands available by the Operational Land Imager (OLI) aboard Landsat-8 to estimate Chla. We utilize a database of co-located in situ radiometric and Chla measurements (N = 4,354), referred to as Type A data, to train and test an MDN model (MDN(A)). This algorithm’s performance, having been proven for other satellite missions, is further evaluated against other widely used machine learning models (e.g., support vector machines), as well as other domain-specific solutions (OC3), and shown to offer significant advancements in the field. Our performance assessment using a held-out test data set suggests that a 49% (median) accuracy with near-zero bias can be achieved via the MDN(A) model, offering improvements of 20 to 100% in retrievals with respect to other models. The sensitivity of the MDN(A) model and benchmarking methods to uncertainties from atmospheric correction (AC) methods, is further quantified through a semi-global matchup dataset (N = 3,337), referred to as Type B data. To tackle the increased uncertainties, alternative MDN models (MDN(B)) are developed through various features of the Type B data (e.g., Rayleigh-corrected reflectance spectra ρ(s)). Using held-out data, along with spatial and temporal analyses, we demonstrate that these alternative models show promise in enhancing the retrieval accuracy adversely influenced by the AC process. Results lend support for the adoption of MDN(B) models for regional and potentially global processing of OLI imagery, until a more robust AC method is developed. Index Terms—Chlorophyll-a, coastal water, inland water, Landsat-8, machine learning, ocean color, aquatic remote sensing.

Brandon Smith↗
NASA NTRS1997

Applying Space Technology to Enhance Control of an Artificial Arm

At the present time, myoelectric prostheses perform only one function of the hand: open and close with the thumb, index and middle finger coming together to grasp various shaped objects. To better understand the limitations of the current single-function prostheses and the needs of the individuals who use them, The Institute for Rehabilitation and Research (TIRR), sponsored by the National Institutes of Health (August 1992 - November 1994), surveyed approximately 2500 individuals with upper limb loss. When asked to identify specific features of their current electric prosthesis that needed improvement, the survey respondents overwhelmingly identified the lack of wrist and finger movement as well as poor control capability. Simply building a mechanism with individual finger and wrist motion is not enough. Individuals with upper limb loss tend to reject prostheses that require continuous visual monitoring and concentration to control. Robotics researchers at NASA's Johnson Space Center (JSC) and Rice University have made substantial progress in myoelectric teleoperation. A myoelectric teleoperation system translates signals generated by an able-bodied robot operator's muscles during hand motions into commands that drive a robot's hand through identical motions. Farry's early work in myoelectric teleoperation used variations over time in the myoelectric spectrum as inputs to neural networks to discriminate grasp types and thumb motions. The resulting schemes yielded up to 93% correct classification on thumb motions. More recently, Fernandez achieved 100% correct non-realtime classification of thumb abduction, extension, and flexion on the same myoelectric data. Fernandez used genetic programming to develop functions that discriminate between thumb motions using myoelectric signal parameters. Genetic programming (GP) is an evolutionary programming method where the computer can modify the discriminating functions' form to improve its performance, not just adjust numerical coefficients or weights. Although the function development may require much computational time and many training cases, the resulting discrimination functions can run in realtime on modest computers. These results suggest that myoelectric signals might be a feasible teleoperation medium, allowing an operator to use his or her own hand and arm as a master to intuitively control an anthropomorphic robot in a remote location such as outer space.

Atkins, Diane↗
NASA NTRS2011

Fast Image Texture Classification Using Decision Trees

Texture analysis would permit improved autonomous, onboard science data interpretation for adaptive navigation, sampling, and downlink decisions. These analyses would assist with terrain analysis and instrument placement in both macroscopic and microscopic image data products. Unfortunately, most state-of-the-art texture analysis demands computationally expensive convolutions of filters involving many floating-point operations. This makes them infeasible for radiation- hardened computers and spaceflight hardware. A new method approximates traditional texture classification of each image pixel with a fast decision-tree classifier. The classifier uses image features derived from simple filtering operations involving integer arithmetic. The texture analysis method is therefore amenable to implementation on FPGA (field-programmable gate array) hardware. Image features based on the "integral image" transform produce descriptive and efficient texture descriptors. Training the decision tree on a set of training data yields a classification scheme that produces reasonable approximations of optimal "texton" analysis at a fraction of the computational cost. A decision-tree learning algorithm employing the traditional k-means criterion of inter-cluster variance is used to learn tree structure from training data. The result is an efficient and accurate summary of surface morphology in images. This work is an evolutionary advance that unites several previous algorithms (k-means clustering, integral images, decision trees) and applies them to a new problem domain (morphology analysis for autonomous science during remote exploration). Advantages include order-of-magnitude improvements in runtime, feasibility for FPGA hardware, and significant improvements in texture classification accuracy.

Thompson, David R.↗