Brent D. Bartlett

Geospatial Disaster Response during the Haiti Earthquake: A Case Study Spanning Airborne Deployment, Data Collection, Transfer, Processing, and Dissemination

Immediately following the 12 January 2010 earthquake in Haiti, a disaster response team from Rochester Institute of Technology, ImageCat Inc., and Kucera International, funded by the Global Facility for Disaster Reduction and Recovery group of the World Bank, collected 0.15 m airborne imagery and two points/m2 lidar data for 650 km2 over a period of seven days. Data were transferred to Rochester, New York for processing at rates that approached 400 Mb/s using Internet2, ortho-rectified with a 24-hour turnaround, and distributed to response agencies through file or disk transfer. A unique response effort, dubbed the Global Earth Observation - Catastrophe Assessment Network (GEO-CAN) and headed by ImageCat, utilized over 600 experts from 23 different countries to generate rapid turnaround damage assessment products. This paper highlights the airborne data collection, transfer, processing, and product development effort, which arguably has raised the bar in terms of response to large-scale disasters.

The SHARE 2012 data campaign

A multi-modal (hyperspectral, multispectral, and LIDAR) imaging data collection campaign was conducted just south of Rochester New York in Avon, NY on September 20, 2012 by the Rochester Institute of Technology (RIT) in conjunction with SpecTIR, LLC, the Air Force Research Lab (AFRL), the Naval Research Lab (NRL), United Technologies Aerospace Systems (UTAS) and MITRE. The campaign was a follow on from the SpecTIR Hyperspectral Airborne Rochester Experiment (SHARE) from 2010. Data was collected in support of the eleven simultaneous experiments described here. The airborne imagery was collected over four different sites with hyperspectral, multispectral, and LIDAR sensors. The sites for data collection included Avon, NY, Conesus Lake, Hemlock Lake and forest, and a nearby quarry. Experiments included topics such as target unmixing, subpixel detection, material identification, impacts of illumination on materials, forest health, and in-water target detection. An extensive ground truthing effort was conducted in addition to collection of the airborne imagery. The ultimate goal of the data collection campaign is to provide the remote sensing community with a shareable resource to support future research. This paper details the experiments conducted and the data that was collected during this campaign.

Spectro-polarimetric bidirectional reflectance distribution function determination of in-scene materials and its use in target detection applications

For sensing systems that characterize the spectro-polarimetric radiance reaching the camera, the origin of the sensed phenomenology is a complex mixture of sources. While some of these sources do not contribute to the polarimetric signature, many do such as the polarization state of the downwelled sky radiance, the target and background p-BRDF(polarimetric bidirectional reflectance distribution function), the polarization state of the upwelled path radiance, and the sensor Mueller matrix transfer function. In this paper we derive portions of the p-BRDF in terms of both the spectral diffuse and polarimetric specular components of the reflectance using an in-scene calibration technique. This process is applied to simulated data, laboratory data, and data from a field collection. Spectra of a car panel for clean and contaminated states derived using laboratory data are injected into a hyperspectral image cube. It is shown how this target can be identified using a target specific tracking vector derived from its polarimetric signature as it moves between spatial locations within a scene.

Atmospheric compensation in the presence of clouds: an adaptive empirical line method (AELM) approach

Many algorithms exist to invert airborne imagery from units of either radiance or sensor specific digital counts to units of reflectance. These compensation algorithms remove unwanted atmospheric variability allowing objects on the ground to be analyzed. Low error levels in homogenous atmospheric conditions have been demonstrated. In many cases however, clouds are present in the atmosphere which introduce error into the inversion at unacceptable levels. For example, the relationship that is defined between sensor reaching radiance and ground reflectance in a cloud free scene will not be the same as in an adjacent region with clouds in the surround. A novel method has been developed which utilizes ground based measurements to modify the empirical line method (ELM) approach on a per-pixel basis. A physics based model of the atmosphere is used to generate a spatial correction for the ELM. Creation of this model is accomplished by analyzing whole-sky imagery to produce a cloud mask which drives input parameters to the radiative transfer (RT) code MODTRAN. The RT code is run for several different azimuth and zenith orientations to create a three-dimensional representation of the hemisphere. The model is then used to achieve a per-pixel correction by adjusting the ELM slope spatially. This method is applied to real data acquired over the atmospheric radiation measurement (ARM) site in Lamount, OK. Performance of the method is evaluated with the Hyperspectral Digital Imagery Collection Experiment (HYDICE) instrument. The sensitivity to spectral sampling is also assessed by down-sampling the HYDICE data to the spectral response of the multi-spectral system Wildfire Airborne Sensor Program LITE (WASP Lite). Finally a method to utilize this approach when additional sensors (like a sky camera) are not available is suggested.

Anomaly detection of man-made objects using spectropolarimetric imagery

In the task of automated anomaly detection, it is desirable to find regions within imagery that contain man-made structures or objects. The task of separating these signatures from the scene background and other naturally occurring anomalies can be challenging. This task is even more difficult when the spectral signatures of the man-made objects are designed to closely match the surrounding background. As new sensors emerge that can image both spectrally and polarimetrically, it is possible to utilize the polarimetric signature to discriminate between many types of man-made and natural anomalies. One type of passive imaging system that allows for spetro-polarimetric data to be collected is the pairing of a liquid crystal tunable filter (LCTF) with a CCD camera thus creating a spectro-polarimetic imager (SPI). In this paper, an anomaly detection scheme is implemented which makes use of the spectral Stokes imagery collected by this sensing system. The ability for the anomaly detector to find man-made objects is assessed as a function of the number of spectral bands available and it is shown that low false alarm rates can be achieved with relatively few spectral bands.

Characterization of material reflectance variation through measurement and simulation

The characterization of material reflectance properties is important in the analysis of hyperspectral and polarization imagery as well as accurate simulation of such images. This paper merges the results of empirical reflectance property (spectral pBRDF) measurements with detailed model based simulations. The empirical data are collected with a laboratory spectroradiometer as well as an RIT-developed spectro-polarimetric imaging goniometer. The modeling uses an adaptation of RITs Digital Imaging and Remote Sensing Image Generation (DIRSIG) model to capture the radiative transfer in rough surfaces with micron-scale features. Measurements and model results for several man-made materials under various conditions are presented.

Improved temperature retrieval methods for the validation of a hydrodynamic simulation of a partially frozen power plant cooling lake

The ALGE code is a hydrodynamic model developed by Savannah River National Laboratory (SRNL) to derive the power output levels of an electric generation facility from observing the associated cooling pond with an aerial imaging platform. Over the past two years work has been completed to extend the capabilities of the model to incorporate snow and ice as possible phenomena in the modeled environment. In order to validate the extension of the model, intensive ground truth data as well as high-resolution aerial infrared imagery were collected during the winters of 2008-2009 and 2009-2010, for a combined eight months of data collection. Due to the harsh and extreme environmental conditions automatic data collection instruments were designed and deployed. Based on experience gained during the first collection season and equipment design failures, overhauls in the design and operation of the automated data collection buoys were performed. In addition, a more thorough and robust twofold calibration technique was implemented within the aerial imaging chain to assess the accuracy of the retrieved surface temperatures. By design, the calibration method employed in this application uses ground collected, geolocated water surface temperatures and in-flight blackbody imagery to produce accurate temperature maps of the pond in interest. A sensitivity analysis was implemented within the data reduction technique to produce accurate sensor reaching temperature values using designed equipment and methods for temperature retrieval at the waters surface.

Use of remote sensing data to enhance the performance of a hydrodynamic simulation of a partially frozen power plant cooling lake

The effectiveness of a power generation sites cooling pond has a significant impact on the overall efficiency of a power plant. The ability to monitor a cooling pond using thermal remote sensing, coupled with hydrodynamic models, is a valuable tool for determining the driving characteristics of a cooling system. However, the thermodynamic analysis of a cooling lake can become significantly more complex when a power generation site is located in a northern climate. The heated effluent from a power plant entering a cooling lake is often not enough to keep a lake from freezing during winter months. Once the lake is partially or fully frozen, the predictive capabilities of the hydrodynamic model are weakened due to an insulating surface layer of ice and snow. Thermal imagery of a cooling pond was collected over a period of approximately 16 weeks in tandem with high-density thermal measurements both in open water and embedded in ice, meteorological data, and snow layer characterization data. The proposed research presents a method to employ thermal imagery to improve the performance of a 3-D hydrodynamic model of a power plant cooling pond in the presence of ice and snow.

Anomaly detection with varied ground sample distance utilizing spectropolarimetric imagery collected using a liquid crystal tunable filter

Liquid crystal tunable filters (LCTFs) are a technology that can act as both a spectral and linear polarization filter for an imaging device. Paired with the appropriate hardware, a LCTF can be configured to collect hyperspectral Stokes imagery which contains both spectral as well as polarimetric information on a per-pixel level basis. This data is used to investigate the utility of spectro-polarimetric data with standard spectral analysis algorithms, in this case anomaly detection. A method to simulate different ground sample distances (GSDs) is used to illustrate the effect on algorithm performance. In this paper, a spectro-polarimetric imager is presented that can collect spectro-polarimetric image cubes in units of calibrated sensor reaching radiance. The system is used to collect imagery of two scenes, each containing die-cast scale vehicles and different background types. An anomaly detector is applied to the intensity and polarized image cubes to find those pixels that are different from the background spectrally and/or polarimetrically. The effect of changing the apparent GSD on the anomaly detection performance is explored. This shows that applying anomaly detection to spectro-polarimetric data can improve the false alarm rate over standard spectral data for finding certain types of man-made objects in complex backgrounds.

Spectro-polarimetric BRDF determination of objects using in-scene calibration materials for polarimetric imagers

For sensing systems that characterize the spectro-polarimetric radiance reaching the camera, the origin of the sensed phenomenology is a complex mixture of sources. While some of these sources do not contribute to the polarimetric signature, many do such as the downwelled sky polarization, the target and background p- BRDF(polarimetric bi-directional reflectance distribution function), the upwelled sky polarization, and the camera Mueller matrix transfer function. In this paper we investigate candidate in-scene calibration materials potentially allowing for portions of the p-BRDF to be derived for material surfaces throughout the scene. Extraction of target p-BRDF from the sensed spectro-polarimetric energy may result in improved target detection performance in the future. Results using both synthetic and real data are presented.