Social and Behavioral Sciences Commons^™

Composite Style Pixel And Point Convolution-Based Deep Fusion Neural Network Architecture For The Semantic Segmentation Of Hyperspectral And Lidar Data, Kevin T. Decker, Brett J. Borghetti

Faculty Publications

Multimodal hyperspectral and lidar data sets provide complementary spectral and structural data. Joint processing and exploitation to produce semantically labeled pixel maps through semantic segmentation has proven useful for a variety of decision tasks. In this work, we identify two areas of improvement over previous approaches and present a proof of concept network implementing these improvements. First, rather than using a late fusion style architecture as in prior work, our approach implements a composite style fusion architecture to allow for the simultaneous generation of multimodal features and the learning of fused features during encoding. Second, our approach processes the higher …

Global Gnss-Ro Electron Density In The Lower Ionosphere, Dong L. Wu, Daniel J. Emmons Ii, Nimalan Swarnalingam

Faculty Publications

Lack of instrument sensitivity to low electron density (N_e) concentration makes it difficult to measure sharp N_e vertical gradients (four orders of magnitude over 30 km) in the D/E-region. A robust algorithm is developed to retrieve global D/E-region N_e from the high-rate GNSS radio occultation (RO) data, to improve spatiotemporal coverage using recent SmallSat/CubeSat constellations. The new algorithm removes F-region contributions in the RO excess phase profile by fitting a linear function to the data below the D-region. The new GNSS-RO observations reveal many interesting features in the diurnal, seasonal, solar-cycle, and magnetic-field-dependent variations in the …

A Comparison Of Sporadic-E Occurrence Rates Using Gps Radio Occultation And Ionosonde Measurements, Rodney Carmona, Omar A. Nava, Eugene V. Dao, Daniel J. Emmons

Faculty Publications

Sporadic-E (E_s) occurrence rates from Global Position Satellite radio occultation (GPS-RO) measurements have shown to vary by a factor of five between studies, motivating the need for a comparison with ground-based measurements. In an attempt to find accurate GPS-RO techniques for detecting E_s formation, occurrence rates derived using five previously developed GPS-RO techniques are compared to ionosonde measurements over an eight-year period from 2010–2017. GPS-RO measurements within 170 km of a ionosonde site are used to calculate E_s occurrence rates and compared to the ground-truth ionosonde measurements. The techniques are compared individually for each ionosonde site …

Machine Learning Land Cover And Land Use Classification Of 4-Band Satellite Imagery, Lorelei Turner [*], Torrey J. Wagner, Paul Auclair, Brent T. Langhals

Faculty Publications

Land-cover and land-use classification generates categories of terrestrial features, such as water or trees, which can be used to track how land is used. This work applies classical, ensemble and neural network machine learning algorithms to a multispectral remote sensing dataset containing 405,000 28x28 pixel image patches in 4 electromagnetic frequency bands. For each algorithm, model metrics and prediction execution time were evaluated, resulting in two families of models; fast and precise. The prediction time for an 81,000-patch group of predictions wasmodels, and >5s for the precise models, and there was not a significant change in prediction time when a …

Regional High-Resolution Benthic Habitat Data From Planet Dove Imagery For Conservation Decision-Making And Marine Planning, Steven R. Schill, Valerie Pietsch Mcnulty, F. Joseph Pollock, Fritjof Lüthje, Jiwei Li, David E. Knapp, Joe D. Kington, Trevor Mcdonald, George T. Raber, Ximena Escovar-Fadul, Gregory P. Asner

Faculty Publications

High-resolution benthic habitat data fill an important knowledge gap for many areas of the world and are essential for strategic marine conservation planning and implementing effective resource management. Many countries lack the resources and capacity to create these products, which has hindered the development of accurate ecological baselines for assessing protection needs for coastal and marine habitats and monitoring change to guide adaptive management actions. The PlanetScope (PS) Dove Classic SmallSat constellation delivers high-resolution imagery (4 m) and near-daily global coverage that facilitates the compilation of a cloud-free and optimal water column image composite of the Caribbean’s nearshore environment. These …

An Operational Overview Of The Export Processes In The Ocean From Remote Sensing (Exports) Northeast Pacific Field Deployment, David A. Siegel, Ivona Cetinić, Jason R. Graff, Craig M. Lee, Norman Nelson, Mary Jane Perry, Inia Soto Ramos, Deborah K. Steinberg, Ken Buesseler, Roberta Hamme, Andrea J. Fassbender, David Nicholson, Melissa M. Omand, Marie Robert, Andrew Thompson, Vinicius Amaral, Michael Behrenfeld, Claudia Benitez-Nelson, Kelsey Bisson, Emmanuel Boss, Philip W. Boyd, Mark Brzezinski, Kristen Buck

Faculty Publications

The goal of the EXport Processes in the Ocean from RemoTe Sensing (EXPORTS) field campaign is to develop a predictive understanding of the export, fate, and carbon cycle impacts of global ocean net primary production. To accomplish this goal, observations of export flux pathways, plankton community composition, food web processes, and optical, physical, and biogeochemical (BGC) properties are needed over a range of ecosystem states. Here we introduce the first EXPORTS field deployment to Ocean Station Papa in the Northeast Pacific Ocean during summer of 2018, providing context for other papers in this special collection. The experiment was conducted with …

Analyzing Satellite Ocean Color Match-Up Protocols Using The Satellite Validation Navy Tool (Savant) At Moby And Two Aeronet-Oc Sites, Adam Lawson, Jennifer Bowers, Sherwin Ladner, Richard Crout, Christopher Wood, Robert Arnone, Paul Martinolich, David Lewis

Faculty Publications

The satellite validation navy tool (SAVANT) was developed by the Naval Research Laboratory to help facilitate the assessment of the stability and accuracy of ocean color satellites, using numerous ground truth (in situ) platforms around the globe and support methods for match-up protocols. The effects of varying spatial constraints with permissive and strict protocols on match-up uncertainty are evaluated, in an attempt to establish an optimal satellite ocean color calibration and validation (cal/val) match-up protocol. This allows users to evaluate the accuracy of ocean color sensors compared to specific ground truth sites that provide continuous data. Various match-up constraints may …

Per-Pixel Cloud Cover Classification Of Multispectral Landsat-8 Data, Salome E. Carrasco [*], Torrey J. Wagner, Brent T. Langhals

Faculty Publications

Random forest and neural network algorithms are applied to identify cloud cover using 10 of the wavelength bands available in Landsat 8 imagery. The methods classify each pixel into 4 different classes: clear, cloud shadow, light cloud, or cloud. The first method is based on a fully connected neural network with ten input neurons, two hidden layers of 8 and 10 neurons respectively, and a single-neuron output for each class. This type of model is considered with and without L2 regularization applied to the kernel weighting. The final model type is a random forest classifier created from an ensemble of …

Assessment Of Normalized Water-Leaving Radiance Derived From Goci Using Aeronet-Oc Data, Mingjun He, Shuangyan He, Xiaodong Zhang, Feng Zhou, Peiliang Li

Faculty Publications

The geostationary ocean color imager (GOCI), as the world’s first operational geostationary ocean color sensor, is aiming at monitoring short-term and small-scale changes of waters over the northwestern Pacific Ocean. Before assessing its capability of detecting subdiurnal changes of seawater properties, a fundamental understanding of the uncertainties of normalized water-leaving radiance (nLw) products introduced by atmospheric correction algorithms is necessarily required. This paper presents the uncertainties by accessing GOCI-derived nLw products generated by two commonly used operational atmospheric algorithms, the Korea Ocean Satellite Center (KOSC) standard atmospheric algorithm adopted in GOCI Data Processing System (GDPS) and the NASA standard atmospheric …

Learning Set Representations For Lwir In-Scene Atmospheric Compensation, Nicholas M. Westing [*], Kevin C. Gross, Brett J. Borghetti, Jacob A. Martin, Joseph Meola

Faculty Publications

Atmospheric compensation of long-wave infrared (LWIR) hyperspectral imagery is investigated in this article using set representations learned by a neural network. This approach relies on synthetic at-sensor radiance data derived from collected radiosondes and a diverse database of measured emissivity spectra sampled at a range of surface temperatures. The network loss function relies on LWIR radiative transfer equations to update model parameters. Atmospheric predictions are made on a set of diverse pixels extracted from the scene, without knowledge of blackbody pixels or pixel temperatures. The network architecture utilizes permutation-invariant layers to predict a set representation, similar to the work performed …

Methods For Real-Time Prediction Of The Mode Of Travel Using Smartphone-Based Gps And Accelerometer Data, Bryan D. Martin, Vittorio Addona, Julian Wolfson, Gediminas Adomavicius, Yingling Fan

Faculty Publications

We propose and compare combinations of several methods for classifying transportation activity data from smartphone GPS and accelerometer sensors. We have two main objectives. First, we aim to classify our data as accurately as possible. Second, we aim to reduce the dimensionality of the data as much as possible in order to reduce the computational burden of the classification. We combine dimension reduction and classification algorithms and compare them with a metric that balances accuracy and dimensionality. In doing so, we develop a classification algorithm that accurately classifies five different modes of transportation (i.e., walking, biking, car, bus and rail) …

Accuracy Assessment Of Pictometry® Height Measurements Stratified By Cardinal Direction And Image Magnification Factor, Daniel Unger, David Kulhavy, I-Kuai Hung, Yanli Zhang

Faculty Publications

The aim of this project was to ascertain if Pictometry^® estimated height could be used in lieu of field-based height estimation. Height of a light pole measured with a telescopic height pole was compared to Pictometry^® hyperspatial 4-inch (10.2 centimeters) multispectral imagery estimated light pole height on the campus of Stephen F. Austin State University, Nacogdoches, Texas. Average percent agreement between light pole height and Pictometry^® estimated light pole height summarized by Pictometry^® image magnification factors at 100%, 125%, 150%, 200%, and 300% magnification were within 98% of light pole height with percent disagreement ranging from …

Incorporating Applied Undergraduate Research In Senior To Graduate Level Remote Sensing Courses, Richard Henley, Daniel Unger, David Kulhavy, I-Kuai Hung

Faculty Publications

An Arthur Temple College of Forestry and Agriculture (ATCOFA) senior spatial science undergraduate student engaged in a multi-course undergraduate research project to expand his expertise in remote sensing and assess the applied instruction methodology employed within ATCOFA. The project consisted of performing a change detection land-use/land-cover classification for Nacogdoches and Angelina counties in Texas using satellite imagery. The dates for the imagery were spaced approximately ten years apart and consisted of four different acquisitions between 1984 and 2013. The classification procedure followed and expanded upon a series of concrete theoretical remote sensing principles, transforming the four remotely sensed raster images …

Evaluating Tree Height Using Pictometry® Hyperspatial Imagery, Daniel Unger, David Kulhavy, Matthew A. Wade

Faculty Publications

This study evaluated the use of Pictometry® hyperspatial 4-inch (10.2 centimeters) multispectral imagery to estimate height of baldcypress trees on the campus of Stephen F. Austin State University (SFASU), Nacogdoches, Texas. Actual tree heights of 60 baldcypress trees measured with a telescopic height pole were compared to Pictometry® estimated tree height. Linear correlation coefficients (r) and coefficient of determinations (R2) between actual tree height and Pictometry® estimated tree height for all 60 tress, and the shortest 30 and tallest 30 trees, were calculated. A paired t-test (alpha = 0.05) was calculated for all 60 tress, and the shortest 30 and …

Sub-Pixel Classification Of Forest Cover Types In East Texas, Joey Westbrook, I-Kuai Hung, Daniel Unger, Yanli Zhang

Faculty Publications

Sub-pixel classification is the extraction of information about the proportion of individual materials of interest within a pixel. Landcover classification at the sub-pixel scale provides more discrimination than traditional per-pixel multispectral classifiers for pixels where the material of interest is mixed with other materials. It allows for the un-mixing of pixels to show the proportion of each material of interest. The materials of interest for this study are pine, hardwood, mixed forest and non-forest. The goal of this project was to perform a sub-pixel classification, which allows a pixel to have multiple labels, and compare the result to a traditional …

Identifying Well Pads In The Haynesville Shale Region, Louisiana And Texas, With Digital Imagery, Darinda Dans, Daniel Unger, Kenneth W. Farrish, I-Kuai Hung

Faculty Publications

The Haynesville Shale is an underlying rock formation in northwest Louisiana and northeast Texas that contains vast quantities of natural gas. With new technology has come the ability to extract more natural gas from one of the largest gas deposits in the United States. With increased production, increased change in the local ecosystem will occur. It is necessary to examine oil and gas exploration effects on the local ecosystem due to changes in land cover, such as habitat loss and increased soil erosion. Remotely sensed imagery were utilized to ascertain the use of various digital image processing techniques to determine …

Design Of A Comprehensive Geographic Information System For The Administration Of El Camino Real De Los Tejas National Historic Trail, Jeffrey M. Williams

Faculty Publications

Stephen F. Austin State University’s Arthur Temple College of Forestry and Agriculture’s (ATCOFA) Geographic Information Systems (GIS) Laboratory were engaged by the National Park Service (NPS) National Trails System-Intermountain Region to provide GIS services supporting the NPS’s development of a Comprehensive Management Plan for El Camino Real de los Tejas National Historic Trail (ELTE). The scope of work was completed under an agreement with the Gulf Coast Cooperative Ecosystem Studies Unit sponsored by the Texas AgriLife Research Program at Texas A&M University. ATCOFA assisted the NPS in the coordination of local landowner and other local stakeholder contacts, conducted archival research …

Assessing The Efficacy Of Modis Satellite-Derived Start Of Growing Season For Jurisdictional Determination Of East Texas Bottomland Hardwood Wetlands, Karen Malone, Hans Michael Williams, I-Kuai Hung, Daniel Unger

Faculty Publications

Introduction: Crucial to the determination of a jurisdictional wetland is the definition of “growing season”. Satellite imagery is being utilized in other ecological applications, but is lagging in wetland growing season determination. Both cost and temporal limitations historically have restrained use of satellite imagery in assessing the start up of the growing season. Multiple commercial satellites are available that provide high resolution imagery, but the cost are prohibitive for most studies. The National Aeronautics and Space Administration (NASA) and the U.S. Geological Survey (USGS) jointly manage the Landsat and the Moderate-resolution Imaging Spectroradiometer (MODIS) satellite programs. Landsat Enhanced Thematic Mapper …

Gis Aided Archaeological Research Of El Camino Real De Los Tejas With Focus On The Landscape And River Crossings Along El Camino Carretera., Jeffrey M. Williams

Faculty Publications

Many generations of indigenous pathways through the forests of eastern Texas have their origins obscured in antiquity. Utilized by early European explorers, these pathways became modified through heavy use and the expansions and improvements needed to accommodate easy passage of European horses and carts and finally the heavy wagons of Anglo-American settlers. The first road through Texas, El Camino Real de Los Tejas, utilized portions of these early trails.

El Camino Carretera (known as the cart road) is an early segment of El Camino Real de los Tejas that crossed the Sabine River at the boundary between Texas and Louisiana. …

Merging Gps Data With High Spatial Resolution Multispectral Imagery: An Urban Recreation Case Study, Daniel Unger, David Kulhavy, Jerome E. Benson Ii

Faculty Publications

In 1992 a disc golf course was created by Alpha Phi Omega, Nu Sigma Chapter of Stephen F. Austin State University within Pecan Park in the c ity of Nacogdoches, Texas. Using constructs from Landscape Ecology, in cluding structure, function and change within a land mosaic provided the basis for establishment of the course. The addition of the disc golf cour se modified the use of the park promoting cultural cohesion among the disc golf enthusiasts. To aid the recreational enjoyment of golf participants, vector m aps of each fairway were created when the disc course was developed and loca …

Accuracy Assessment Of Land Cover Maps Derived From Multiple Data Sources, Daniel Unger, Hillary Tribby, Hans Michael Williams, I-Kuai Hung

Faculty Publications

Maximum Likelihood (ML) and Artificial Neural Network (ANN) supervised classification methods were used to demarcate land cover types within IKONOS and Landsat ETM+ imagery. Three additional data sources were integrated into the classification process: Canopy Height Model (CHM), Digital Terrain Model (DTM) and Thermal data. Both the CHM and DTM were derived from multiple return small footprint LIDAR. Forty maps were created and assessed for overall map accuracy, user's accuracy, producer's accuracy, kappa statistic and Z statistic using classification schemes from U.S.G.S. 1976 levels 1 and 2 and T.G.l.C. 1999 levels 2 and 4. Results for overall accuracy of land …

Spatial Analysis Of Historic Cemeteries: Using High Spatial Resolution Imagery As A Visual Aid, Richard E. Brooks, Daniel Unger, I-Kuai Hung

Faculty Publications

Oak Grove Cemetery, located within the City of Nacogdoches in Nacogdoches County Texas, is one of the earliest cemeteries in the county dating to the early 1800’s. Several historic Texans are interred within this cemetery including Thomas J. Rusk and Charles S. Taylor who was the great-great-grandfather of current United States Senator Kay Bailey Hutchison. Due to a fire circa 1910 many of the records for the original section of the cemetery were lost. In the summer of 2006, the GPS coordinates of each grave marker within the cemetery were plotted on a backdrop of 6 inch spatial resolution multispectral …

Remotely Sensed Data To Map Forest Age Class By Cover Type In East Texas, Daniel Unger, I-Kuai Hung, Jeffrey M. Williams, James Kroll, Dean W. Coble, Jason Grogan

Faculty Publications

• Remote sensing in conjunction with ground truthing, can accurately quantify forest composition and age distributions in East Texas.
• Method uses standardized and readily available data available to the general public.
• Method was shown to be effective in terms of time and cost.

Search And Recovery Of The Space Shuttle Columbia: A Geospatial 1st Responder Perspective, Jeffrey M. Williams

Faculty Publications

A first person account of the Texas geospatial volunteers and their efforts to recover the remains of the Space Shuttle Columbia and her crew lost over eastern Texas and western Louisiana on February 1st, 2003.