Charles Bachmann | Rochester Institute of Technology (original) (raw)

Papers by Charles Bachmann

Research paper thumbnail of Wavelength dependence of the bidirectional reflectance distribution function (BRDF) of beach sands

Applied Optics, 2015

The wavelength dependence of the dominant directional reflective properties of beach sands was de... more The wavelength dependence of the dominant directional reflective properties of beach sands was demonstrated using principal component analysis and the related correlation matrix. In general, we found that the hyperspectral bidirectional reflectance distribution function (BRDF) of beach sands has weak wavelength dependence. Its BRDF varies slightly in three broad wavelength regions. The variations are more evident in surfaces of greater visual roughness than in smooth surfaces. The weak wavelength dependence of the BRDF of beach sand can be captured using three broad wavelength regions instead of hundreds of individual wavelengths.

Research paper thumbnail of How many spectral bands are necessary to describe the directional reflectance of beach sands?

SPIE Proceedings - Algorithms and Technologies for Multispectral, Hyperspectral and Ultraspectral Imagery XXII XXII, 2016

Spectral variability in the visible, near-infrared and shortwave directional reflectance factor of... more Spectral variability in the visible, near-infrared and shortwave directional reflectance factor of beach sands and freshwater sheet flow is examined using principal component and correlation matrix analysis of in situ measurements. In previous work we concluded that the hyperspectral bidirectional reflectance distribution function (BRDF) of beach sands in the absence of sheet flow exhibit weak spectral variability, the majority of which can be described with three broad spectral bands with wavelength ranges of 350-450 nm, 700-1350 nm, and 1450-2400 nm. Observing sheet flow on sand we find that a thin layer of water enhances reflectance in the specular direction at all wavelengths and that spectral variability may be described using four spectral band regions of 350-450 nm, 500-950 nm, 950-1350 nm, and 1450-2400 nm. Spectral variations are more evident in sand surfaces of greater visual roughness than in smooth surfaces, regardless of sheet flow.

Research paper thumbnail of Coastal characterization from hyperspectral imagery: An intercomparison of retrieval properties from three coast types

2010 IEEE International Geoscience and Remote Sensing Symposium, 2010

Over the past three years, the Naval Research Laboratory has conducted three remote sensing campa... more Over the past three years, the Naval Research Laboratory has conducted three remote sensing campaigns in regions which included barrier island, coral, and mangrove coastal characteristics. These investigations were conducted at the Virginia Coast . During each of the studies, hyperspectral imagery (HSI) was acquired over the site in conjunction with both land and water spectral and geotechnical measurements. For coastal classification, such as retrievals of bottom reflectance, bathymetry, beach composition, and vegetation type, subtle variations in spectral properties of physical and biological features are often significant. Beach composition in these coast types differed dramatically due to the origin of the sands. These findings suggest the need to develop models specific to a particular coast type.

Research paper thumbnail of Mariana Islands-Hyperspectral Airborne Remote Environmental Sensing Experiment 2010

Prescribed by ANSI Std. Z39.18

Research paper thumbnail of Coastal Characterization from Hyperspectral Imagery

Imaging and Applied Optics Congress, 2010

... Naval Research Laboratory, Remote Sensing Division, Coastal Science and Interpretation Sectio... more ... Naval Research Laboratory, Remote Sensing Division, Coastal Science and Interpretation Section, Code 7232, 4555 Overlook Ave., SW, Washington, DC 20375; bachmann@nrl.navy.mil ... mangroves, shrubs, coral, algae, SAVs Dominance Type sand, shell hash, silt, clay ...

Research paper thumbnail of Mixtures of projection pursuit models: an automated approach to land cover classification in Landsat Thematic Mapper imagery

Geoscience and Remote Sensing IEEE International Symposium, 1999

Unsupervised projection pursuit methods and principal component analysis are compared for extract... more Unsupervised projection pursuit methods and principal component analysis are compared for extraction of features from Landsat Thematic Mapper imagery. On sequestered test data, PP projections improved separation of individual categories from all other categories; improvement ranged from a few percent to as much as ≈22%. End-to-end classification of land-cover, combining these features with supervised classifiers, is also described. For sequestered

Research paper thumbnail of Modeling and Estimation of Sediment Density from Hyperspectral BRDF

The angular dependence of sand and sediment reflectance, as characterized by the bidirectional re... more The angular dependence of sand and sediment reflectance, as characterized by the bidirectional reflectance distribution function (BRDF), provides a mechanism for observing and modeling the effects of density in reflectance data. Density effects in granular materials have been described in radiative transfer models and are known, for example, to influence the size of the opposition effect . In some instances, such as in the description of the coherent backscatter opposition effect, explicit models for wavelength dependence have also been derived .

Research paper thumbnail of Geodatabase development to support hyperspectral imagery exploitation

2010 IEEE International Geoscience and Remote Sensing Symposium, 2010

Geodatabase development for coastal studies conducted by the Naval Research Laboratory (NRL) is e... more Geodatabase development for coastal studies conducted by the Naval Research Laboratory (NRL) is essential to support the exploitation of hyperspectral imagery (HSI). NRL has found that the remote sensing and mapping science community benefits from coastal classifications that group coastal types based on similar features. Selected features in project geodatabases relate to significant biological and physical forces that shape the coast. The project geodatabases help researchers understand factors that are necessary for imagery post processing, especially those features having a high degree of temporal and spatial variability. NRL project geodatabases include a hierarchy of environmental factors that extend from shallow water bottom types and beach composition to inland soil and vegetation characteristics. These geodatabases developed by NRL allow researchers to compare features among coast types. The project geodatabases may also be used to enhance littoral data archives that are sparse. This paper highlights geodatabase development for recent remote sensing experiments in barrier island, coral, and mangrove coast types.

Research paper thumbnail of Very Shallow Water Bathymetry Retrieval from Hyperspectral Imagery at the Virginia Coast Reserve (VCR'07) Multi-Sensor Campaign

IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium, 2008

A number of institutions, including the Naval Research Laboratory (NRL), have developed look up t... more A number of institutions, including the Naval Research Laboratory (NRL), have developed look up tables for remote retrieval of bathymetry and in-water optical properties from hyperspectral imagery (HSI) . For bathymetry retrieval, the lower limit is the very shallow water case (here defined as < 2m), a depth zone which is not well resolved by many existing bathymetric LIDAR sensors, such as SHOALS . The ability to rapidly model these shallow water depths from HSI directly has potential benefits for combined HSI/LIDAR systems such as the Compact Hydrographic Airborne Rapid Total Survey (CHARTS) . In this study, we focused on the validation of a near infra-red feature, corresponding to a local minimum in absorption (and therefore a local peak in reflectance), which can be correlated directly to bathymetry with a high degree of confidence. Compared to other VNIR wavelengths, this particular near-IR feature corresponds to a peak in the correlation with depth in this very shallow water regime, and this is a spectral range where reflectance depends primarily on water depth (water absorption) and bottom type, with suspended constituents playing a secondary role.

Research paper thumbnail of Optical Remote Sensing of the Environment (ORS)

Research paper thumbnail of Retrieval of Substrate Bearing Strength from Hyperspectral Imagery during the Virginia Coast Reserve (VCR’07) Multi-Sensor Campaign

Marine Geodesy, 2010

Hyperspectral imagery (HSI) derived from remote sensing can delineate surface prop-erties of subs... more Hyperspectral imagery (HSI) derived from remote sensing can delineate surface prop-erties of substrates such as type, moisture, and grain size. These are critical parameters that determine the substrate bearing strength. Although HSI only sees the surface layer, statistics can be ...

Research paper thumbnail of Bathymetry Retrieval from Hyperspectral Imagery in the Very Shallow Water Limit: A Case Study from the 2007 Virginia Coast Reserve (VCR'07) Multi-Sensor Campaign

Marine Geodesy, 2010

... Alan Weidemann c , Wesley Goode c , C. Reid Nichols d , Patrick Woodward d , Kevin McIlhany e... more ... Alan Weidemann c , Wesley Goode c , C. Reid Nichols d , Patrick Woodward d , Kevin McIlhany e , Victoria Hill f , Richard Zimmerman f , Daniel ... com) InGaAs SWIR HSI sensor operating in the 0.9–1.7 μm spectral region; and a single-channel Indigo Systems Merlin midwave IR ...

Research paper thumbnail of Improving the performance of classifiers in high-dimensional remote sensing applications: an adaptive resampling strategy for error-prone exemplars (ARESEPE)

IEEE Transactions on Geoscience and Remote Sensing, 2003

... Index Terms—Active learning, active sampling, barrier islands, hyperspectral, land-cover clas... more ... Index Terms—Active learning, active sampling, barrier islands, hyperspectral, land-cover classification, Virginia Coast Reserve. ... A number of the previous approaches to active learning are also slow, especially for high-dimensional applications such as hyperspectral imagery. ...

Research paper thumbnail of Field Expedient Water Level Measurements to Support the Exploitation of Hyperspectral Imagery

Research paper thumbnail of The effects of atmospheric correction schemes on the hyperspectral imaging of littoral environments

We investigate the effects of several atmospheric correction schemes (ACS) on spectral determinat... more We investigate the effects of several atmospheric correction schemes (ACS) on spectral determination, biophysical parameter estimation, and unsupervised classification from hyperspectral data collected over a coastal New Jersey tidal marsh. The ACS examined include: two modes from ACORN 4.0 (modes 1.0 and 1.5), ATREM, FLAASH, Tafkaa-6S, and Tafkaa-tabular. Results from the comparative analysis of derived spectra reveal a high degree of similarity for all methods for terrestrial derived spectra but considerably less so for spectra obtained from aquatic environments. Likewise, the similarities in the terrestrial spectra translate to significant correlations among the ACS for two of four computed biophysical reflectance indices (NDVI and Red Edge Wavelength). Of the remaining two indices, PRI and SIPI, computed values from both ACORN modes were not correlated with other ACS. Results from ISODATA classification for land and water environments, show a consistently high level of agreement...

Research paper thumbnail of NRL Hyperspectral Imagery Trafficability Tool (HITT): Software and Spectral-Geotechnical Look-up Tables for Estimation and Mapping of Soil Bearing Strength from Hyperspectral Imagery

This report provides details of a new hyperspectral exploitation tool for developing trafficabili... more This report provides details of a new hyperspectral exploitation tool for developing trafficability special maps from hyperspectral imagery. This new tool, known as the NRL Hyperspectral Imagery Trafficability Tool (HITT), is packaged as a plug-in to the commercial software package ENVI/IDL. HITT is one of several tools being developed and validated by NRL to support coastal characterization from spectral region imagery. This report provides a step-by-step description of the operation of HITT, working through specific examples that begin with a hyperspectral data cube and end with a special map of estimated trafficability in a coastal region. The basis of this new product is a set of spectral-geotechnical libraries and models developed during remote sensing and calibration/validation campaigns conducted by NRL and collaborating institutions in four different coastal types.

Research paper thumbnail of Automated Estimation of Spectral Neighborhood Size in Manifold Coordinate Representations of Hyperspectral Imagery: Implications for Anomaly Finding, Bathymetry Retrieval, and Land Applications

IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium, 2008

Research paper thumbnail of Manifold coordinate representations of hyperspectral imagery: Improvements in algorithm performance and computational efficiency

2010 IEEE International Geoscience and Remote Sensing Symposium, 2010

Several years ago, NRL first demonstrated a computationally efficient framework for optimizing a ... more Several years ago, NRL first demonstrated a computationally efficient framework for optimizing a set of intrinsic manifold coordinates for high-dimensional data such as hyperspectral imagery (HIS). Since that time, NRL has continued to improve the fidelity of the representation to describe the details of the nonlinear structure in even greater detail. In addition, working closely with CelesTech, several improvements have been made that increase the computational efficiency, accelerating processing speeds from hours to minutes.

Research paper thumbnail of A New Data-Driven Approach to Modeling Coastal Bathymetry from Hyperspectral Imagery Using Manifold Coordinates

Proceedings of OCEANS 2005 MTS/IEEE, 2005

Research paper thumbnail of Modeling Coastal Waters from Hyperspectral Imagery using Manifold Coordinates

2006 IEEE International Symposium on Geoscience and Remote Sensing, 2006

[2], we introduced a direct data driven method of modeling nonlinear structure in hyperspectral i... more [2], we introduced a direct data driven method of modeling nonlinear structure in hyperspectral imagery based on Isometric Mapping . More recently, we have further improved the scaling of the approach [2], making it a practical method for large-scale hyperspectral scenes. The new method extracts a set of data manifold coordinates that directly parameterize nonlinearities present in hyperspectral imagery, both on land and in the water column. In the water column, this is particularly important because of the nonlinear, attenuating properties of the medium. In this paper, we model hyperspectral imagery acquired by the NRL PHILLS [5] at the Indian River Lagoon, Florida in July 2004. In our previous efforts [3] using a small subset of data derived from the surf zone outside of the lagoon, dominant manifold coordinates were shown to parameterize bathymetry directly with a high degree of correlation to a radiative transfer look-up table (LUT) approach. In the present work, we construct a full scene manifold coordinate representation and use this as the basis of a LUT for samples with known depths as determined by the SHOALS LIDAR. Sequestered test data presented to the manifold based LUT yield a mean estimated depth which differs from the LIDAR retrieved depth by less than 0.44m for depths between 0-10m with a standard deviation less than 1.2m.

Research paper thumbnail of Wavelength dependence of the bidirectional reflectance distribution function (BRDF) of beach sands

Applied Optics, 2015

The wavelength dependence of the dominant directional reflective properties of beach sands was de... more The wavelength dependence of the dominant directional reflective properties of beach sands was demonstrated using principal component analysis and the related correlation matrix. In general, we found that the hyperspectral bidirectional reflectance distribution function (BRDF) of beach sands has weak wavelength dependence. Its BRDF varies slightly in three broad wavelength regions. The variations are more evident in surfaces of greater visual roughness than in smooth surfaces. The weak wavelength dependence of the BRDF of beach sand can be captured using three broad wavelength regions instead of hundreds of individual wavelengths.

Research paper thumbnail of How many spectral bands are necessary to describe the directional reflectance of beach sands?

SPIE Proceedings - Algorithms and Technologies for Multispectral, Hyperspectral and Ultraspectral Imagery XXII XXII, 2016

Spectral variability in the visible, near-infrared and shortwave directional reflectance factor of... more Spectral variability in the visible, near-infrared and shortwave directional reflectance factor of beach sands and freshwater sheet flow is examined using principal component and correlation matrix analysis of in situ measurements. In previous work we concluded that the hyperspectral bidirectional reflectance distribution function (BRDF) of beach sands in the absence of sheet flow exhibit weak spectral variability, the majority of which can be described with three broad spectral bands with wavelength ranges of 350-450 nm, 700-1350 nm, and 1450-2400 nm. Observing sheet flow on sand we find that a thin layer of water enhances reflectance in the specular direction at all wavelengths and that spectral variability may be described using four spectral band regions of 350-450 nm, 500-950 nm, 950-1350 nm, and 1450-2400 nm. Spectral variations are more evident in sand surfaces of greater visual roughness than in smooth surfaces, regardless of sheet flow.

Research paper thumbnail of Coastal characterization from hyperspectral imagery: An intercomparison of retrieval properties from three coast types

2010 IEEE International Geoscience and Remote Sensing Symposium, 2010

Over the past three years, the Naval Research Laboratory has conducted three remote sensing campa... more Over the past three years, the Naval Research Laboratory has conducted three remote sensing campaigns in regions which included barrier island, coral, and mangrove coastal characteristics. These investigations were conducted at the Virginia Coast . During each of the studies, hyperspectral imagery (HSI) was acquired over the site in conjunction with both land and water spectral and geotechnical measurements. For coastal classification, such as retrievals of bottom reflectance, bathymetry, beach composition, and vegetation type, subtle variations in spectral properties of physical and biological features are often significant. Beach composition in these coast types differed dramatically due to the origin of the sands. These findings suggest the need to develop models specific to a particular coast type.

Research paper thumbnail of Mariana Islands-Hyperspectral Airborne Remote Environmental Sensing Experiment 2010

Prescribed by ANSI Std. Z39.18

Research paper thumbnail of Coastal Characterization from Hyperspectral Imagery

Imaging and Applied Optics Congress, 2010

... Naval Research Laboratory, Remote Sensing Division, Coastal Science and Interpretation Sectio... more ... Naval Research Laboratory, Remote Sensing Division, Coastal Science and Interpretation Section, Code 7232, 4555 Overlook Ave., SW, Washington, DC 20375; bachmann@nrl.navy.mil ... mangroves, shrubs, coral, algae, SAVs Dominance Type sand, shell hash, silt, clay ...

Research paper thumbnail of Mixtures of projection pursuit models: an automated approach to land cover classification in Landsat Thematic Mapper imagery

Geoscience and Remote Sensing IEEE International Symposium, 1999

Unsupervised projection pursuit methods and principal component analysis are compared for extract... more Unsupervised projection pursuit methods and principal component analysis are compared for extraction of features from Landsat Thematic Mapper imagery. On sequestered test data, PP projections improved separation of individual categories from all other categories; improvement ranged from a few percent to as much as ≈22%. End-to-end classification of land-cover, combining these features with supervised classifiers, is also described. For sequestered

Research paper thumbnail of Modeling and Estimation of Sediment Density from Hyperspectral BRDF

The angular dependence of sand and sediment reflectance, as characterized by the bidirectional re... more The angular dependence of sand and sediment reflectance, as characterized by the bidirectional reflectance distribution function (BRDF), provides a mechanism for observing and modeling the effects of density in reflectance data. Density effects in granular materials have been described in radiative transfer models and are known, for example, to influence the size of the opposition effect . In some instances, such as in the description of the coherent backscatter opposition effect, explicit models for wavelength dependence have also been derived .

Research paper thumbnail of Geodatabase development to support hyperspectral imagery exploitation

2010 IEEE International Geoscience and Remote Sensing Symposium, 2010

Geodatabase development for coastal studies conducted by the Naval Research Laboratory (NRL) is e... more Geodatabase development for coastal studies conducted by the Naval Research Laboratory (NRL) is essential to support the exploitation of hyperspectral imagery (HSI). NRL has found that the remote sensing and mapping science community benefits from coastal classifications that group coastal types based on similar features. Selected features in project geodatabases relate to significant biological and physical forces that shape the coast. The project geodatabases help researchers understand factors that are necessary for imagery post processing, especially those features having a high degree of temporal and spatial variability. NRL project geodatabases include a hierarchy of environmental factors that extend from shallow water bottom types and beach composition to inland soil and vegetation characteristics. These geodatabases developed by NRL allow researchers to compare features among coast types. The project geodatabases may also be used to enhance littoral data archives that are sparse. This paper highlights geodatabase development for recent remote sensing experiments in barrier island, coral, and mangrove coast types.

Research paper thumbnail of Very Shallow Water Bathymetry Retrieval from Hyperspectral Imagery at the Virginia Coast Reserve (VCR'07) Multi-Sensor Campaign

IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium, 2008

A number of institutions, including the Naval Research Laboratory (NRL), have developed look up t... more A number of institutions, including the Naval Research Laboratory (NRL), have developed look up tables for remote retrieval of bathymetry and in-water optical properties from hyperspectral imagery (HSI) . For bathymetry retrieval, the lower limit is the very shallow water case (here defined as < 2m), a depth zone which is not well resolved by many existing bathymetric LIDAR sensors, such as SHOALS . The ability to rapidly model these shallow water depths from HSI directly has potential benefits for combined HSI/LIDAR systems such as the Compact Hydrographic Airborne Rapid Total Survey (CHARTS) . In this study, we focused on the validation of a near infra-red feature, corresponding to a local minimum in absorption (and therefore a local peak in reflectance), which can be correlated directly to bathymetry with a high degree of confidence. Compared to other VNIR wavelengths, this particular near-IR feature corresponds to a peak in the correlation with depth in this very shallow water regime, and this is a spectral range where reflectance depends primarily on water depth (water absorption) and bottom type, with suspended constituents playing a secondary role.

Research paper thumbnail of Optical Remote Sensing of the Environment (ORS)

Research paper thumbnail of Retrieval of Substrate Bearing Strength from Hyperspectral Imagery during the Virginia Coast Reserve (VCR’07) Multi-Sensor Campaign

Marine Geodesy, 2010

Hyperspectral imagery (HSI) derived from remote sensing can delineate surface prop-erties of subs... more Hyperspectral imagery (HSI) derived from remote sensing can delineate surface prop-erties of substrates such as type, moisture, and grain size. These are critical parameters that determine the substrate bearing strength. Although HSI only sees the surface layer, statistics can be ...

Research paper thumbnail of Bathymetry Retrieval from Hyperspectral Imagery in the Very Shallow Water Limit: A Case Study from the 2007 Virginia Coast Reserve (VCR'07) Multi-Sensor Campaign

Marine Geodesy, 2010

... Alan Weidemann c , Wesley Goode c , C. Reid Nichols d , Patrick Woodward d , Kevin McIlhany e... more ... Alan Weidemann c , Wesley Goode c , C. Reid Nichols d , Patrick Woodward d , Kevin McIlhany e , Victoria Hill f , Richard Zimmerman f , Daniel ... com) InGaAs SWIR HSI sensor operating in the 0.9–1.7 μm spectral region; and a single-channel Indigo Systems Merlin midwave IR ...

Research paper thumbnail of Improving the performance of classifiers in high-dimensional remote sensing applications: an adaptive resampling strategy for error-prone exemplars (ARESEPE)

IEEE Transactions on Geoscience and Remote Sensing, 2003

... Index Terms—Active learning, active sampling, barrier islands, hyperspectral, land-cover clas... more ... Index Terms—Active learning, active sampling, barrier islands, hyperspectral, land-cover classification, Virginia Coast Reserve. ... A number of the previous approaches to active learning are also slow, especially for high-dimensional applications such as hyperspectral imagery. ...

Research paper thumbnail of Field Expedient Water Level Measurements to Support the Exploitation of Hyperspectral Imagery

Research paper thumbnail of The effects of atmospheric correction schemes on the hyperspectral imaging of littoral environments

We investigate the effects of several atmospheric correction schemes (ACS) on spectral determinat... more We investigate the effects of several atmospheric correction schemes (ACS) on spectral determination, biophysical parameter estimation, and unsupervised classification from hyperspectral data collected over a coastal New Jersey tidal marsh. The ACS examined include: two modes from ACORN 4.0 (modes 1.0 and 1.5), ATREM, FLAASH, Tafkaa-6S, and Tafkaa-tabular. Results from the comparative analysis of derived spectra reveal a high degree of similarity for all methods for terrestrial derived spectra but considerably less so for spectra obtained from aquatic environments. Likewise, the similarities in the terrestrial spectra translate to significant correlations among the ACS for two of four computed biophysical reflectance indices (NDVI and Red Edge Wavelength). Of the remaining two indices, PRI and SIPI, computed values from both ACORN modes were not correlated with other ACS. Results from ISODATA classification for land and water environments, show a consistently high level of agreement...

Research paper thumbnail of NRL Hyperspectral Imagery Trafficability Tool (HITT): Software and Spectral-Geotechnical Look-up Tables for Estimation and Mapping of Soil Bearing Strength from Hyperspectral Imagery

This report provides details of a new hyperspectral exploitation tool for developing trafficabili... more This report provides details of a new hyperspectral exploitation tool for developing trafficability special maps from hyperspectral imagery. This new tool, known as the NRL Hyperspectral Imagery Trafficability Tool (HITT), is packaged as a plug-in to the commercial software package ENVI/IDL. HITT is one of several tools being developed and validated by NRL to support coastal characterization from spectral region imagery. This report provides a step-by-step description of the operation of HITT, working through specific examples that begin with a hyperspectral data cube and end with a special map of estimated trafficability in a coastal region. The basis of this new product is a set of spectral-geotechnical libraries and models developed during remote sensing and calibration/validation campaigns conducted by NRL and collaborating institutions in four different coastal types.

Research paper thumbnail of Automated Estimation of Spectral Neighborhood Size in Manifold Coordinate Representations of Hyperspectral Imagery: Implications for Anomaly Finding, Bathymetry Retrieval, and Land Applications

IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium, 2008

Research paper thumbnail of Manifold coordinate representations of hyperspectral imagery: Improvements in algorithm performance and computational efficiency

2010 IEEE International Geoscience and Remote Sensing Symposium, 2010

Several years ago, NRL first demonstrated a computationally efficient framework for optimizing a ... more Several years ago, NRL first demonstrated a computationally efficient framework for optimizing a set of intrinsic manifold coordinates for high-dimensional data such as hyperspectral imagery (HIS). Since that time, NRL has continued to improve the fidelity of the representation to describe the details of the nonlinear structure in even greater detail. In addition, working closely with CelesTech, several improvements have been made that increase the computational efficiency, accelerating processing speeds from hours to minutes.

Research paper thumbnail of A New Data-Driven Approach to Modeling Coastal Bathymetry from Hyperspectral Imagery Using Manifold Coordinates

Proceedings of OCEANS 2005 MTS/IEEE, 2005

Research paper thumbnail of Modeling Coastal Waters from Hyperspectral Imagery using Manifold Coordinates

2006 IEEE International Symposium on Geoscience and Remote Sensing, 2006

[2], we introduced a direct data driven method of modeling nonlinear structure in hyperspectral i... more [2], we introduced a direct data driven method of modeling nonlinear structure in hyperspectral imagery based on Isometric Mapping . More recently, we have further improved the scaling of the approach [2], making it a practical method for large-scale hyperspectral scenes. The new method extracts a set of data manifold coordinates that directly parameterize nonlinearities present in hyperspectral imagery, both on land and in the water column. In the water column, this is particularly important because of the nonlinear, attenuating properties of the medium. In this paper, we model hyperspectral imagery acquired by the NRL PHILLS [5] at the Indian River Lagoon, Florida in July 2004. In our previous efforts [3] using a small subset of data derived from the surf zone outside of the lagoon, dominant manifold coordinates were shown to parameterize bathymetry directly with a high degree of correlation to a radiative transfer look-up table (LUT) approach. In the present work, we construct a full scene manifold coordinate representation and use this as the basis of a LUT for samples with known depths as determined by the SHOALS LIDAR. Sequestered test data presented to the manifold based LUT yield a mean estimated depth which differs from the LIDAR retrieved depth by less than 0.44m for depths between 0-10m with a standard deviation less than 1.2m.