John Bolton's Initiatives - Appendix A

APPENDIX A: DDF Proposals



Note: The Director's Discretionary Fund Program was discontinued in 2005


1997 Director’s Discretionary Fund Proposal
Principal Investigator: John Bolton/730

Proposal Title: A Commercial Off The Shelf (COTS) Image Correlation Tracker (for further information see The Hexapod Project)


Background
The Image Correlation Tracker (ICT) is a device that will follow (track) a moving video image. The device works by acquiring sequential video images and comparing the displacements of those images. The displacement of the sequential images is presented in terms of X and Y displacements and rotation.

The typical application of the ICT is image stabilization. Solar physicists have used an ICT to control the motion of their telescopes to follow features on the Sun. The military has used ICTs to stabilize weapons systems. These systems have been built with dedicated image processors resulting in, particularly in the case of the military systems, very expensive and complex systems.

Objectives
The objective of this DDF proposal is to build an ICT system with commercial off the shelf (COTS) components. Recent developments in image processor boards for the personal computer (PC) should make this feasible. Several suppliers make complete image processing systems for the PC, including user friendly development tools. A wide variety of video cameras are also available. This approach is innovative in the sense that the people usually associated with the COTS image processors and with traditional ICT systems do not interact. I have used this approach bringing together typically unrelated disciplines successfully before, in a different context, in the development of an airborne imaging spectrometer. This approach, using COTS products, will be very useful for many Goddard projects. It will be particularly useful for the newer, younger GSFC employees to use this approach as it does not require the level of experience of the senior staff.

R&D Plan
The R&D plan is straightforward. As the scope of the project is clearly defined, the key steps are identification of the appropriate hardware and software, and then the assembly and programming of the system. Testing and evaluation are also straightforward as the applications and systems with which the ICT will operate have already been identified.

Month Activity
1 Market study and background research
2 Order system components
3 Plan system and test fixtures
4 Receive components
5 Assemble components and test development tools
6 Evaluate initial results and plan final system
7 Assemble and program final system
8 Test and debug final system
9 Evaluate performance of final system
10 Improve system performance
11 Field test system
12 Write final report

Context
The ICT has many potential applications. My particular interest is to use the system to stabilize an imaging spectrometer in airborne applications. The system could be used, similarly, to stabilize any sort of airborne imaging instrument. Accurate image stabilization will provide geometrically corrected raw data, eliminating the need for tedious post-processing. These airborne imaging instruments are used extensively in Goddard's ongoing campaigns, and will be used even more when similar instruments are used for underflight calibration and validation of the EOS instruments.

Stabilization of imaging instruments is typically done with inertially stabilized systems. The advantage of the ICT is that it uses the image itself as the stabilization reference. This provides for a much more accurate frame of reference. It also provides for more control flexibility. For the imaging spectrometer for example, exact image registration can be provided by programming the ICT to shift the sequential images by precisely the amount required by the spectrometer which operates in the "pushbroom" mode.

Though this proposal has no co-investigators, a large number of people and organizations, both Governmental and Academic, and even some in the public and private sectors, who will benefit from this development have been identified. It is expected that a lot of informal assistance will be obtained, particularly late in the project when field testing is required and people are eager to see how this system will enhance the performance of their imagers.

Budget

Manpower
PI 0.25 time

Line Items
PC system $ 3500
PC system software $ 2000
Imaging system hardware $ 15000
Imaging system software $ 2000
Development Tools $ 1500
Video Camera $ 1500
Video camera controller $ 500
Data Storage hardware $ 2000
Test fixtures $ 1500
Total $ 29500


1999 Director’s Discretionary Fund Proposal
Principal Investigator: John Bolton/730

Proposal Title: A Digital Terrain Mapping System Feasibility Study

Background:

Terrain mapping is a very important component of remote sensing. Typical remotely sensed imagery comes out “flat”. Without the additional elevation information of a terrain map, remotely obtained imagery is not of much use for many applications. There are very powerful methods for combining imagery with terrain maps. While there are a lot of terrain maps available, notably from the USGS, there are many applications that require custom maps, either for areas that do not have high enough spatial resolution or for areas in which changes have occurred. Producing a traditional terrain map is a tedious process at best, typically involving aerial photography and a lot of manual operations. Alternative means, such as radar and lidar can be used but do not yield the detail available through the photogrammetric method.

The system proposed here would demonstrate a principle to produce “on-the-fly” digital terrain maps. Instead of using stereo area images, this system would acquire forward- and aft-looking digital line scan images in a single camera. These images, with their fixed look angles, would be combined to produce elevation information without tedious processing. The feasibility of this scheme has been demonstrated in theory, but as yet no hardware has been built to enable an end-to-end test of the concept. This proposal would verify the feasibility of the fundamental, and most risky part, of the digital terrain mapping system. Completion of the project within one year will not be a problem.

As there is no stereo, line-scan, photogrammetric quality, imaging system available to do this project, a relatively simple test camera will have to be built. Fortunately, all of the components are available either as commercial, off-the-shelf (COTS) hardware and software, or from the partners mentioned below, or from the Goddard/Wallops optical facility. This proposal is very cost effective as many of the expensive remote sensing capabilities required for this project (precision instrument pointing system, flight time for data collection, photogrammetric verification, photogrammetric data processing systems, etc.) are available to the PI for little or no cost.

Objectives:

· Build a simple stereo line scan camera and data acquisition system to acquire raw, digital elevation image data
· Acquire sample data, and test data processing methods
· Refine data acquisition and processing methods
· Report on the accuracy and efficiency of “on-the-fly” results compared to traditional results

Research Development Plan:

· Finalize system design and acquire components (4 months)
· Assemble camera and data acquisition hardware (2 months)
· Run laboratory tests of system (1 month)
· Run preliminary flight tests of system (1 month)
· Run simultaneous flight tests with a traditional data acquisition system (2 months)
· Compare traditional results to results obtained with new system (1 month)

Context:

As terrain mapping is a very important component of remote sensing, it fits very closely with Goddard’s mission. As the proposed system to obtain digital terrain maps is new, innovative, and somewhat risky, it fits well with the objectives of the DDF. This project will build on previous development work in hyperspectral remote sensing, precision instrument pointing, and image processing. Extensive discussions with experts in the terrain mapping business have led to the development of this proposal. We can expect cooperation with these experts at no cost to the project. These partners include; The Ohio State University Center for Mapping (a NASA Commercial Space Center), Code 730.1 (Systems, Analysis, Process & Tools Office (IMDC)), Code 571 (Guidance, Navigation and Control Center), 3D Imaging (a commercial GIS and photogrammetry company), Richard Crouse & Assoc. (a provider of aerial photography services). The method of developing and testing instrumentation on an airborne platform is an important new process to be established for the IMDC (Code 730.1).

This proof-of-concept system will not have all of the features required for a fully capable digital terrain mapping system. The successful completion of this DDF will allow us to seek funding to build the full up system. It is very likely that we could work with the Goddard Commercial Technology Office (CTO) to find a partner for the advanced development phases.

Budget:

· Design and build Digital, Line Scan, Stereo Camera $16,500
(line scan cameras, controllers, power supplies, cables, and camera interface)
· Interface Camera with Data Acquisition system $14,000
(camera interface, personal computer and mass data storage system and software)
· Acquire and Evaluate Test Data $8,500
(image processing hardware and software)
· Report and Demonstration $2,000
· Total $41,000

PI Time 0.25 man-years


2000 Director’s Discretionary Fund Proposal

Principal Investigator: John Bolton/730

Proposal Title: A Hyperspectral Imaging Radiometer Transfer Standard (HIRTS)

Background:

Calibration and characterization of Earth remote sensing instruments is a complex process. Traditional calibration involves a highly reflecting integrating sphere and standard lamps. While this procedure has been well developed and is widely accepted, it suffers from several problems. Among these are source stability, source spectral distribution, and portability.

Efforts have been made to develop detector-based standards to supplement or replace source-based standards. The ultimate standard is a source. It cannot, however, be used in any practical situation so its values must be transferred to other sources via a detector-based transfer standard. The detector-based standards that have been developed are either broad band devices, or broad band devices with a filter of some type.

A hyperspectral imaging radiometer transfer standard (HIRTS) provides multiple features that provide solutions to the problems described above. The primary features of the system are:

· Wide spectral coverage (extendable with multiple spectrometers)
· High spectral resolution (dependent on detector array selected)
· Multiple scan, statistical enhancement capability
· Imaging or multi-spatial capability
· Portability, and potential for field use

Recent developments in hyperspectral imaging and related technologies have made it possible to build a very sophisticated system with commercial off-the-shelf (COTS) parts. This work is an innovative application, based on previous work in the development of airborne hyperspectral systems, and the calibration and characterization of those systems. Components will be chosen to optimize the features required for calibration. The primary operational difference between the proposed system and an airborne hyperspectral imager is the scan rate and the dynamic range. Great care will be taken to thoroughly characterize the instrument to understand the spectral and spatial purity characteristics that are so important for calibration.

Several NASA organizations and projects would participate in this work. In-house efforts in Code 920 to develop calibration systems would be augmented. This new approach would supplement the previous work on detector-based standards done at NIST and the British NPL. Field testing, using the instrument as a traditional imaging spectrometer, would prove its capability to provide ground truth measurements and to calibrate remote sensing instruments.

Objectives:

· Build a stable hyperspectral imager as a HIRTS test bed
· Test radiometric stability in the laboratory
· Acquire data from various standard sources, and in the field
· Refine data acquisition and processing methods
· Report on the accuracy and efficiency of hyperspectral results compared to traditional results

Research and Development Plan:

· Finalize system design and acquire components (4 months)
· Assemble camera and data acquisition hardware (2 months)
· Run laboratory stability and performance tests of system (1 month)
· Run preliminary functional stability and performance tests of system (1 month)
· Run simultaneous stability tests with a traditional data acquisition system (2 months)
· Compare traditional results to results obtained with new system (1 month)

Context:

As instrument calibration is a very important component of remote sensing, it fits very closely with Goddard’s mission. As the proposed system to provide a calibration reference is new, innovative, and somewhat risky, it fits well with the objectives of the DDF. This project will build on previous development work in hyperspectral remote sensing, and instrument calibration and characterization. Extensive discussions with experts in calibration and characterization of instruments have led to the development of this proposal. We can expect cooperation with these experts at no cost to the project. These partners include: Code 920, NIST, NPL, and The University of Arizona. Flight Projects with an interest in calibration and in hyperspectral instrumentation will also participate.

This proof-of-concept system will have all of the features required for a fully capable HIRTS test bed. The successful completion of this DDF will allow us to seek funding to build the full up system. We will work with the Goddard Commercial Technology Office (CTO) to find a partner for commercialization of the system.

Budget:

· Design and build hyperspectral imager $26,500
(Spectrometer, CCD array and camera, optics, controllers, power supplies)
· Interface camera with data acquisition system $4,000
(Camera interface, personal computer and mass data storage system and software)
· Acquire and evaluate test data $18,500
(Acquire test source and compare with standard laboratory sources)
· Report and demonstration setup $2,000
(Prepare working system for demonstration purposes)
· Total $52,000

PI Time 0.25 man-years


2001 Director’s Discretionary Fund Proposal
Principal Investigator: John Bolton/730

Co-Investigators: Joel Gambino/573, Patrick Coronado/935

Proposal Title: A Geolocation System for Remote Sensing Instruments

Background:

One of the primary problems in processing remote sensing data is geolocation, associating a location on the surface of the Earth with each data point. Typically, remote sensing data must be geolocated by referring to ground control points. This is a labor-intensive, costly, and often haphazard business. The task is particularly difficult for non-imaging instruments.

Some efforts have been made to develop instrument systems incorporating a geolocation capability. These typically consist of an off-the-shelf Global Positioning System (GPS) and an Inertial Navigation System (INS) system bolted on to the instrument, and supplied with some data collection software. Most of these systems do not make any provision for integration of the geolocation and instrument data, and none of them make any provision for instrument attitude correction. Laborious integration and post-processing of the data is still required.

The proposed geolocation system would provide multiple features that would solve the problems described above. The primary features of the system are:

· Continuous recording of attitude and position information
· Automatic boresighting of instrument and platform
· Integration of attitude and position information into the instrument data stream
· Control capability for an attitude correction system

Recent developments in GPS and INS systems and related technologies have made it possible to build a very sophisticated system with commercial off-the-shelf (COTS) parts. This work is an innovative application, based on previous work in the development of geolocation and positioning systems, remote sensing instruments, and the calibration and characterization of those systems. Components will be chosen to optimize the features required for geolocation.

Several NASA organizations and projects would participate in this work. The GN&C Center would provide expertise in GPS and INS Systems. The Applied Information Sciences Branch would provide instrument integration and data processing expertise. Instrument developers from the Earth Sciences would determine performance requirements and test instrumentation. This capability will be particularly useful for laser-ranging instruments. An existing partnership with Dr. Charles Nguyen and the Department of Engineering at Catholic University will be utilized for the attitude correction system part of this project.

Objectives:

· Assemble an integrated GPS/INS system
· Test geolocation and attitude determination capability in the laboratory
· Acquire sample data from various instruments
· Integrate instrument data with geolocation and attitude data
· Use geolocation information to control attitude correction system

Research and Development Plan:

· Finalize system design and acquire components (4 months)
· Assemble and test geolocation system and acquire instrument data (4 months)
· Run laboratory system performance tests (2 months)
· Run preliminary attitude correction system tests (2 months)

Context:

As geolocation is an integral component of Earth remote sensing, it fits very closely with Goddard’s mission. The successful completion of this DDF would provide a very powerful tool for facilitating and simplifying remote sensing data collection and processing. This capability is, for example, of immediate interest to investigators at the Department of Natural Resources (DNR) of the State of Maryland. It could be immediately applied to remote sensing applications for the Chesapeake Bay watershed, which are well supported by the Chesapeake Bay Foundation. Once the system is integrated with remote sensing instruments, it will have a significant education and outreach potential, as it will make data processing and collection accessible to researchers with few resources and little remote sensing experience.

As the proposed system to provide a new method for geolocation is innovative, and somewhat risky, it fits well with the objectives of the DDF. This project will build on previous development work in airborne remote sensing, attitude control systems, and data processing. Interest in this capability has been expressed by many instrument development groups, both here at Goddard and elsewhere. The principles developed in this project will be applicable to orbital systems as well as airborne systems.

This proof-of-concept geolocation system will have all of the features required for a fully operational system. The successful completion of this DDF will allow us to seek funding to build a full-up system for a specific instrument or instrument suite. We will work with the Goddard Commercial Technology Office (CTO) to find a partner for commercialization of the system.

Budget:

· Design and build GPS/INS system $31,500
(Acquire GPS – INS systems, antennas, and control and data acquisition software)
· Acquire and evaluate test data $2,500
(Acquire instrument data and integrate with GPS/INS data)
· Control attitude correction system with GPS/INS $9,000
(Upgrade attitude correction system controllers and integrate with GPS/INS)
· Total $43,000

PI Time 0.25 man-years

Co-I Time 0.10 man-years (each)


2002 Director’s Discretionary Fund Education Outreach Proposal
Principal Investigator: John Bolton/740

Co-Investigators: Michael Hubenthal/130

Proposal Title: Development of a Remote Sensing Education and Outreach Laboratory

1) Relationship of the Proposal to Goal 3 of the Goddard Strategic Plan:

The proposed Remote Sensing Education and Outreach Laboratory (RSEOL) has relevance to all of the Objectives of all three Strategies of Goal 3 of the Goddard Strategic Plan. The goal of the RSEOL is to make remotely sensed data and remote sensing technology available to people who are not remote sensing specialists. In this process of accomplishing this goal, the RSEOL will:

Strategy 1, Objective 1: make users aware of Goddard’s mission activities and results, specifically those of the EOS Project, by providing detailed mission information and access to mission data. All features of the RSEOL will be available on the Internet, providing universal access.

Strategy 1, Objective 2: provide a mechanism for Goddard employees from several Codes, to make users aware of their work by linking to their data products and technological developments.

Strategy 1, Objective 3: will include educational and tutorial materials as an integral part of the program.

Strategy 1, Objective 4: will form partnerships with educational institutions and research organizations to provide wider access to remotely sensed data and a better understanding of remote sensing technology.

Strategy 2, Objective 1: will serve as a test bed and development facility for use by educators to develop programs related to remote sensing.

Strategy 2, Objective 2: will provide educators the ability to use Goddard’s resources to supplement scientific and technical education in general.

Strategy 3, Objective 1: as a primary goal, provide easy access to Goddard mission information.

Strategy 3, Objective 2: will serve as a model remote sensing lab and information source that others may use as they see fit.

2) Clarification of the Relationship to a GSFC Line of Business:

The RSEOL is directly related to the Earth Science, Education, and the Technology Development business lines of Goddard. The RSEOL will both provide information about the Goddard Earth Science projects, and will provide access to the data and data products from those projects. This is an improvement over existing outreach projects in that it will allow the users to actually “do” remote sensing with real, relevant remote sensing data. As the RSEOL is designed to serve users over the entire range of remote sensing capabilities it will serve to match the requirements for training with the knowledge and expertise of various users. RSEOL will also support education through research on and development of products and services that facilitate the application of remote sensing technology to enhance the knowledge and skills of students in the areas of science, math, technology, engineering, and geography. School systems will be able to communicate their curriculum needs to the RSEOL remote-sensing experts for support and development. We will also form partnerships with local universities and organizations in the remote sensing business such as Towson University and the Maryland Space Grant Consortium.

3) Innovative nature of the Proposal and Success Criteria:

This proposal is innovative in that it will make remote sensing data and technology accessible to people with no previous experience. The RSEOL will provide a site, both in-house and on-line, where innovative methods for disseminating remotely sensed data and technology may be developed. There is currently no such facility available at Goddard. While there are many remote sensing laboratories, and many web sites with information about remote sensing, there is no single source for extensive information about remote sensing, nor is there one that highlights the activities at Goddard. There is no on-line site that provides for the kind of extensive user feedback and interaction with Goddard staff that is planned for the RSEOL. The project will be successful if it can provide useful remote sensing information to users with no formal remote sensing experience.

4) Customer:

The customer of the RSEOL will be anybody with an interest in learning about or developing user-friendly methods for remote sensing, particularly, school systems and students that are working to address national and state level technology content standards for all grade levels. The American Association for the Advancement of Science (AAAS) concludes that students “must use different tools to do different things in science and to solve practical problems. Through design and technology projects students can engage in problem-solving related to a wide range of real-world contexts.” (Project 2061). Early possible customers could be local Maryland districts that have already established partnerships with the GSFC Education Programs Office. This proposal applies to anyone from an elementary school student who might have some experience with the GLOBE Program (for example) to a wetlands researcher who needs some specific information about the Chesapeake Bay (for example). As the RSEOL develops and as more teachers and researchers participate, a “shopping list” of remote sensing capabilities and educational programs will be made available.

5) Scale-Up Potential and Deployment Strategy:

The RSEOL proposed here is intended to be a prototype or development platform for remote sensing laboratories. It is expected that users can either come in to work at the Laboratory, or access the Laboratory via an RSEOL web page on the Internet. Progress and updates will be shared with other NASA missions/projects at the monthly Code 900 education meetings and with regular updates to the Education Programs Office throughout the development period. The investigation will follow the NASA Earth Science Education (ESE) Product Review guidelines and will be submitted for review. Upon receiving a favorable recommendation, the web site will be referenced and linked through the Educator Resource Center (ERC) and ESE resources. Awareness training for GSFC Aerospace Education Services Program (AESP) personnel and GSFC Education Programs (Code 130.3) staff will be scheduled. As new data is acquired and new technologies are developed, these will be shared with the participants and made available via the Internet. Partnerships developed with local universities and organizations will be further exploited.

6) Project Plan and Budget:

The basic infrastructure for the RSEOL already exists in Building 16, Room S004. In order to get it up and running properly, hardware and software enhancements are required. This would cost about $2,500 for the essentials. To add innovative capabilities such as peer-to-peer networking and collaborative computing, which are extremely useful for handling large remote sensing data sets and for processing that data, another $5,000, for a total of $7.500, would be required.

A most useful addition to the laboratory would be a full-time resident teacher or remote sensing specialist. This could probably be arranged either through the Education Office, the Code 420 Special Projects Initiative office, or through Code 900.


2002 Director’s Discretionary Fund Proposal
Principal Investigator: John Bolton/740

Co-Investigators: Michael Comberiate/730/Special Projects Initiative Office/422
Beau Legeer/Research Systems Incorporated

Proposal Title: Remote Sensing On-Line

Background:

Because of the difficulty in obtaining and processing remotely sensed data, access to this data has traditionally been limited to a relatively small community of remote sensing experts. Recent gains in the popularity of Geographic Information Systems (GIS) and the development of new sources of remotely sensed data have highlighted the need for easy access to this data.

Two problems with remotely sensed data are large data volumes and difficulty in processing the data to produce a useful product (scientific information). Several software packages are available at no cost for processing remotely sensed data, but these still require that the data be located and downloaded, and they still require some degree of expertise in handling the data. Remote Sensing On-Line (RSOL) would make data acquisition and processing capabilities available to users via a web server. By simply using a web browser the user could select the required data and create the desired data product. Only the data that is actually needed would be transmitted to the user via the Internet. Every operation would include links to instructions and a remote sensing tutorial, so that the user can, if needed, obtain instant help and enlightenment.

Extensive research has shown that there is no capability like this currently available. There are many tools that support image processing, but none of them support the entire range of remote sensing imaging processing and most are only useful to experts. There are a few systems, some of them available on-line, that provide image processing and manipulation, but there are none that provide full capabilities for remote sensing data processing. This idea has received enthusiastic support from both users of remote sensing data and from providers of data and image processing software. Several Goddard organizations, including Codes 130 (Education Office), 420 (EOS Project), 580 (Science Data Systems and Advanced Data Management), and 900 (Applied Information Sciences and Education) are planning to participate in the RSOL Project. The project is somewhat risky in that it will be challenging to program the web server to provide all of the required functions and still be a user-friendly interface.

Objectives:

· Develop a user-friendly data selection and location system (spatial and temporal)
· Provide access to full (all spectral bands), remotely sensed, data sets (from multiple sources)
· Provide both image manipulation capability and access to scientific data products
· Allow downloading of the data products

Research and Development Plan:

· Evaluate software for RSOL system (2 months)
· Develop a “strawman” user-friendly web interface to the RSOL system (3 months)
· Incorporate the image acquisition and data processing software (2 months)
· Link functions to quick instructions and tutorials (3 months)
· Make RSOL available to the remote sensing community for evaluation and comments (2 months and continuing)

Context:

The RSOL project fits in very well with Goddard’s mission. By providing remotely sensed data and technological information to the non-expert it will help to “enhance the Nation’s technological and scientific literacy by sharing the information and knowledge that result from the performance of Goddard’s mission” (Goddard Strategic Plan, Goal 3). This Project will build on the basic framework for a remote sensing on-line system that has been developed in partnership with Research Systems Incorporated, and supported by the EOS Project Office. As mentioned above, several groups within Goddard are planning to cooperate on this project. We can also expect the participation of several organizations outside Goddard, including Towson University, The Chesapeake Bay Program, and StormCenter.Com. The RSOL will also provide a user-friendly interface for newly developed applications of EOS data such as the Rapid Response Project that has recently made available Terra data to help fight forest fires. This project will provide some of the new hires in the Information Technology (IT) discipline here at Goddard with the chance to develop their skills. Teachers and summer student interns will be available to work on this Project through the Special Projects Initiatives Office.

Budget:

· RSOL Software $10,500 (Acquire data acquisition, manipulation, and processing software)
· Web Server Hardware and Software $5,500
· Web Design $2,500 (Develop user-friendly interface)
· Help and Tutorial Development $3,000 (Write and/or incorporate on-line files to provide directions)
· Total $21,500

PI Time 0.25 man-years
Co-I Time 0.10 man-years (each)


2003 Director’s Discretionary Fund Education and Outreach Proposal
Principal Investigator: John Bolton/420

Co-Investigators: Michael Hubenthal/130, Kathy Bender (Resource Specialist)

Proposal Title: Development of a “Mini-Grid” for a Student Remote Sensing Laboratory

1) Relationship of the Proposal to Goal 3 of the Goddard Strategic Plan:

The proposed “Mini-Grid” for a student Remote Sensing Laboratory has relevance to all of the Objectives of all three Strategies of Goal 3 of the Goddard Strategic Plan. The goal of the Mini-Grid is to apply the principles of Grid Computing to make remotely sensed data and remote sensing technology more readily available to small laboratories that do not have all the latest computer hardware and technology:

Strategy 1, Objective 1: make users aware of Goddard’s mission activities and results, specifically those of the EOS Project, by facilitating access to detailed EOS mission information and mission data. All features of the Mini-Grid will be available on the Internet, providing universal access.

Strategy 1, Objective 2: provide a mechanism for Goddard employees from several Codes, to make users aware of their work by linking to their data products and technological developments.

Strategy 1, Objective 3: will include educational and tutorial materials as an integral part of the program.

Strategy 1, Objective 4: will form partnerships with educational institutions and research organizations to provide wider access to remotely sensed data and a better understanding of remote sensing technology.

Strategy 2, Objective 1: will serve as a part of a test bed and development facility for use by educators to develop programs related to remote sensing.

Strategy 2, Objective 2: will help provide educators the ability to use Goddard’s resources to supplement scientific and technical education in general.

Strategy 3, Objective 1: will help to provide easy access to Goddard mission information.

Strategy 3, Objective 2: will serve as a model for remote sensing technology that others may use as they see fit.

2) Clarification of the Relationship to a GSFC Line of Business:

The Mini-Grid is directly related to the Earth Science, Education, and the Technology Development business lines of Goddard. The Mini-Grid will assist in providing both provide information about the Goddard Earth Science projects, and access to the data and data products from those projects. This is an improvement over existing outreach projects in that it will provide better tools to allow the users to actually “do” remote sensing with real, relevant remote sensing data. As the Mini-Grid is designed to serve users over the entire range of remote sensing capabilities it will serve to match the requirements for training with the knowledge and expertise of various users. We will also form partnerships to promote the Mini-Grid concept with local universities and organizations in the remote sensing business such as Towson University and the Maryland Space Grant Consortium.

3) Innovative nature of the Proposal and Success Criteria:

This proposal is innovative in that it will help to make remote sensing data and technology accessible to organizations that do not have the best resources. The Mini-Grid will provide the model for a system that may be used in any small remote sensing laboratory. There is currently no such facility available at Goddard. While there are many remote sensing laboratories, and many web sites with information about remote sensing technology, there is no currently available system tailored to this type of small-scale application. The project will be successful if it can enhance the capabilities of small laboratories that do not have access to state-of-the-art remote sensing capabilities.

4) Customer:

The customer of the Mini-Grid will be anybody with an interest in learning about or developing methods for remote sensing, particularly, school systems and students that are working to address national and state level technology content standards for all grade levels. The American Association for the Advancement of Science (AAAS) concludes that students “must use different tools to do different things in science and to solve practical problem. Through design and technology projects students can engage in problem-solving related to a wide range of real-world contexts.” (Project 2061). Early possible customers could be local Maryland districts that have already established partnerships with the GSFC Education Programs Office. This proposal applies to anyone from an elementary school student who might have some experience with the GLOBE Program (for example) to a wetlands researcher who needs some specific information about the Chesapeake Bay (for example). As the Mini-Grid develops and as more teachers and researchers participate, a “shopping list” of remote sensing capabilities and educational programs compatible with the Mini-Grid will be made available.

5) Scale-Up Potential and Deployment Strategy:

The Mini-Grid proposed here is intended to be a prototype or development platform for student or limited resource remote sensing laboratories. The system will be developed using the existing facilities of the Remote Sensing Education and Outreach Laboratory (RSEOL). Development of the Mini-Grid will lead directly to technologies for peer-to-peer networking and collaborative and distributed computing. It is expected that users can either come in to work at the Laboratory, or access information about the Mini-Grid via the RSEOL web pages on the Internet. Progress and updates will be shared with other NASA missions/projects at the monthly Code 900 education meetings and with regular updates to the Education Programs Office throughout the development period. The investigation will follow the NASA Earth Science Education (ESE) Product Review guidelines and will be submitted for review. Upon receiving a favorable recommendation, the web site will be referenced and linked through the Educator Resource Center (ERC) and ESE resources. Awareness training for GSFC Aerospace Education Services Program (AESP) personnel and GSFC Education Programs (Code 130.3) staff will be scheduled. As new data is acquired and new technologies are developed, these will be shared with the participants and made available via the Internet. Partnerships developed with local universities and organizations will be further exploited.

6) Project Plan and Budget:

The basic infrastructure for the Mini-Grid already exists in Building 16, Room S004. In order to get it up and running properly, hardware and software enhancements are required. This would cost about $2,500 for the essentials. Approximately $7500 would be needed for Information Technology (IT) contractor support. To add innovative capabilities such as peer-to-peer networking and collaborative computing, which are extremely useful for handling large remote sensing data sets and for processing that data, another $5,000, for a total of $15,000, would be required.

A most useful addition to the laboratory would be a full-time resident teacher or remote sensing specialist. This could probably be arranged either through the Education Office, the Code 420 Special Projects Initiative office, or through Code 900.


2003 Director’s Discretionary Fund Proposal
Principal Investigator: John Bolton/420

Co-Investigators: John Schott/RIT, Mike Richardson/RIT, Tom Miewald/CVI, Ed Howard/NOAA, Bill Campbell/935

Proposal Title: A Full-Spectral Imaging Feasibility Study

Background:

Current optical remote sensing instrument technology allows the acquisition and digitization of all of the reflected energy (light) across the full-spectral range of interest. The current method for acquiring, transmitting, and processing this data is still based on the “multi-band” approach that has been used for the past thirty years.

This proposal seeks to investigate the feasibility of using entirely new methods for pre-process­ing, transmitting, and extracting information from full-spectral, remotely sensed data. The goal of the project will be to convert from the current “bytes-per-band” approach to the “spectral curve” approach. This approach has the possibility to greatly simplify instrument characteriza­tion and to significantly reduce data transmission and storage requirements.

The tools currently available for handling and processing full-spectral data will be evaluated. Techniques for handling the data in the form of spectral curves will be proposed and developed. The tools of information theory will be used to minimize the number of bytes required while at the same time losing none of the information content. We will see how the information obtained from an orbiting instrument can be most efficiently transmitted to the ground and utilized to most effectively produce the standard remote sensing data products.

We expect that a basic application of the full spectral imaging principle will reduce data trans­mission and storage requirements by an order of magnitude. Refinement of the principle could produce a reduction of two orders of magnitude. Supplementing full spectral imaging with the principle of spectro-spatial compression could produce another order of magnitude reduction.

Objectives:

· Catalog and benchmark existing band-based capabilities
· Evaluate existing full-spectral data acquisition, transmission, and processing systems
· Investigate currently available alternative data handling and processing systems
· Explore and develop new full-spectral concepts
· Evaluate capabilities of full spectral systems and spectro-spatial compression
· Assess advantages of new and alternative systems

Research and Development Plan:

· Define current multi-band state-of-the-art and data products (1 month)
· Define current full-spectral systems state-of-the-art (1 month)
· Research alternative pre-processing, transmission, and processing systems (5 months)
· Develop “strawman” full-spectral evaluation systems (2 months)
· Compare and evaluate new technology systems (2 months)
· Compile information and prepare report on findings (1 month)

Context:

This proposed project is definitely innovative and also high risk. The risk is not primarily in the availability of technology and applications, but in the acceptance of this approach by the remote sensing community and NASA in particular. This DDF would allow the PI to demonstrate the feasibility and utility of full spectral imaging, and prepare for further investigations and devel­opment of the principle.

The project fits perfectly with Goddard’s Mission as it seeks to develop an innovative technology for Earth science measurements from space that will help “develop and maintain advanced information systems for the display, analysis, archiving and distribution of space and Earth science data”. In addition this research could lead to technologies that could help to “develop National Oceanic and Atmospheric Administration (NOAA) satellite systems that provide environmental data for forecasting and research”.

Budget:

· Analytical Software $6,500
· Supplemental Hardware and associated Software $9,000
· Total $15,500


PI Time 0.25 man-years
Co-I Time 0.10 man-years


2004 Director’s Discretionary Fund Proposal
Principal Investigator: John Bolton/420

Co-Investigators: Kurtis Thome/UAZ

Proposal Title: Vicarious Calibration of Hyperspectral Sensors

Background:

Hyperspectral sensors are capable of acquiring an enormous amount of information. Included in a typical hyperspectral scene are many so-called “pseudo-invariant targets”. These targets have spectral and spatial features that do not change significantly over time. The use of pseudo-invariant targets and of well-characterized ground truth sites to calibrate spaceborne sensors is called “vicarious calibration”.

This proposal seeks to investigate the feasibility of using vicarious calibration to calibrate, verify and characterize a hyperspectral sensor. This approach differs from the traditional approach in that the normal data acquisition process, rather then special “calibration campaigns” would be used to provide the vicarious calibration information. In other words, the instrument calibration process would be integrated into the normal data acquisition process, and it would be continuous.

We expect that the application of vicarious calibration, as described above, to a hyperspectral sensor will eliminate the need for any on-board calibration capability. Not only will the vicarious calibration process provide all the instrument calibration information that is needed, it will also provide information that can be used to develop empirical retrieval algorithms.

Objectives:

· Review the vicarious calibration literature and contact vicarious calibration experts
· Evaluate existing vicarious calibration technologies and capabilities
· Investigate applicability and advantages of current vicarious calibration technologies to hyperspectral sensors
· Explore and develop new statistical methods for vicarious calibration
· Evaluate vicarious calibration versus traditional calibration methods
· Investigate and propose alternative instrument technology to optimize the use of vicarious calibration

Research and Development Plan:

· Review literature and contact calibration experts (3 months)
· Define current vicarious calibration state-of-the-art (1 month)
· Formulate application plan (2 months)
· Develop “strawman” techniques for vicarious calibration (2 months)
· Evaluate vicarious calibration techniques (2 months)
· Compare vicarious calibration to traditional calibration (1 month)
· Compile information and prepare report on findings (1 month)

Context:

This proposed project is innovative and also high risk. It is innovative in that it uses accepted techniques in a new and more extensive fashion. The project risk is not primarily in the availability of technology and applications, but in the acceptance of this approach by the remote sensing community and NASA in particular. This DDF would allow the PI to demonstrate the utility and effectiveness of vicarious calibration for hyperspectral sensors, and prepare for further investigations and development of the principle.

The project fits perfectly with Goddard’s Mission as it seeks to develop an innovative technology for Earth science measurements from space that will help “develop and maintain advanced information systems for the display, analysis, archiving and distribution of space and Earth science data”. In addition this research could lead to technologies that could help to “develop National Oceanic and Atmospheric Administration (NOAA) satellite systems that provide environmental data for forecasting and research”.

Innovation Summary:

Calibration of spaceborne sensors has always been a difficult and sometimes controversial business. By relying entirely on ground truth, the users of the remotely sensed information will always have an absolute reference on which to base their analyses. This will eliminate the “middleman”, the modeler or the retrieval algorithm developer. Developing the concept of exclusive vicarious calibration for hyperspectral sensors would put Goddard at the forefront of the establishment of an exceptionally powerful remote sensing technology. The principles of vicarious calibration have been known, and the application studied for years. The idea to rely on vicarious calibration exclusively is new. This project will be successful if the remote sensing community sees the value of using vicarious calibration exclusively. The major risk is that the community will not be willing to change, or even give up an approach to remote sensing that it has used for the past 30 years.

Budget:

· Analytical Software $6,500
· Supplemental Hardware and associated Software $9,000
· Travel 3,500
· Total $19,000

 

PI Time 0.25 man-years
Co-I Time 0.10 man-years


2004 Director’s Discretionary Fund Proposal
Principal Investigator: John Bolton/420

Co-Investigators: Mike Richardson/RIT, Stephen Reichenbach/UNL

Proposal Title: A Full-Spectral Imaging Feasibility Study

Background:

Current optical remote sensing instrument technology allows the acquisition and digitization of all of the reflected energy (light) across the full-spectral range of interest. The current method for acquiring, transmitting, and processing this data is still based on the “multi-band” approach that has been used for the past thirty years.

This proposal seeks to investigate the feasibility of using entirely new methods for pre-processing, transmitting, and extracting information from full-spectral, remotely sensed data. The goal of the project will be to convert from the current “bytes-per-band” approach to the “spectral curve” approach. This approach has the possibility to greatly simplify instrument characterization and to significantly reduce data transmission and storage requirements.

The tools currently available for handling and processing full-spectral data will be evaluated. Techniques for handling the data in the form of spectral curves will be proposed and developed. The tools of information theory will be used to minimize the number of bytes required while at the same time losing none of the information content. We will see how the information obtained from an orbiting instrument can be most efficiently transmitted to the ground and utilized to most effectively produce the standard remote sensing data products.

We expect that a basic application of the Full Spectral Imaging (FSI) principle will reduce data transmission and storage requirements by an order of magnitude. Refinement of the principle and supplementing FSI using spectro-spatial compression could produce another order of magnitude reduction.

Objectives:

· Catalog and benchmark existing multi-band and hyperspectral capabilities
· Evaluate existing full-spectral data acquisition, transmission, and processing systems
· Investigate currently available alternative data handling and processing systems
· Explore and develop new full-spectral concepts
· Evaluate capabilities of full spectral systems and spectro-spatial compression
· Assess advantages of new and alternative systems

Research and Development Plan:

· Define current multi-band state-of-the-art and data products (1 month)
· Define current full-spectral systems state-of-the-art (1 month)
· Research alternative pre-processing, transmission, and processing systems (5 months)
· Develop “strawman” full-spectral evaluation systems (2 months)
· Compare and evaluate new technology systems (2 months)
· Compile information and prepare report on findings (1 month)

Context:

This proposed project is definitely innovative and also high risk. The risk is not primarily in the availability of technology and applications, but in the acceptance of this approach by the remote sensing community and NASA in particular. This DDF would allow the PI to demonstrate the feasibility and utility of full spectral imaging, and prepare for further investigations and development of the principle. This research would allow a follow-up to the well-received paper I presented at the 10th international Symposium on Remote Sensing sponsored by the SPIE entitled, “Full Spectral Imaging; A Revisited Approach to Remote Sensing”.

The project fits perfectly with Goddard’s Mission as it seeks to develop an innovative technology for Earth science measurements from space that will help “develop and maintain advanced information systems for the display, analysis, archiving and distribution of space and Earth science data”. In addition this research could lead to technologies that could help to “develop National Oceanic and Atmospheric Administration (NOAA) satellite systems that provide environmental data for forecasting and research”.

Innovation Summary:

Full Spectral Imaging is the next step beyond Hyperspectral Imaging. It can open the doors to major advances in passive optical remote sensing technology, in which significant advances have not been made in more then ten years. Developing FSI would put Goddard at the forefront of the development of the most widely used type of remote sensing. The technology to build a FSI system is currently available “off-the-shelf”. This project will be successful if the remote sensing community sees and accepts the value of FSI. The major risk is that the community will not be willing to give up, much less change, the approach to remote sensing that it has used for the past 30 years.

Budget:

· Analytical Software $6,500
· Supplemental Hardware and associated Software $9,000
· Travel 5,500
· Total $21,000

PI Time 0.25 man-years
Co-I Time 0.10 man-years



This page was last modified on 12 April 2012