Mapping the Path to Digital Sensor Calibration USGS

Mapping the Path to Digital Sensor Calibration USGS Land Remote Sensing Program ASPRS Camera Calibration Session March 9, 2005 George Lee USGS WRG gylee@usgs. gov Greg Stensaas USGS NC EROS stensaas@usgs. gov Phil Rufe USGS MCMC prufe@usgs. gov U. S. Department of the Interior U. S. Geological Survey National Center for EROS Remote Sensing Technologies Group 1 Jon Christopherson USGS NC EROS/SAIC jonchris@usgs. gov

History USGS has been calibrating aerial film cameras since 1973 USGS actively involved with establishing requirements and standards for photo imagery for many years Film camera standards since 1950’s with the National Bureau of Standards National Map Accuracy standards since the 1940’s In 1999 ASPRS panel asked USGS to develop standards, policies and guidelines for the digital aerial mapping community USGS Land Remote Sensing Program has a cross center project team actively working to answer ASPRS recommendations U. S. Department of the Interior U. S. Geological Survey 2

ASPRS Recommendations 1. 2. 3. 4. 5. 6. 7. The USGS Optical Science Laboratory (OSL) should continue to calibrate film mapping cameras using the present calibrator and the Simultaneous Multiframe Analytical Calibration (SMAC) program. Initiate the design, development, and implementation of a digital camera calibration capability at the USGS (est. required investment - $4 M). Conduct research efforts in order to support a reliable and cost effective transition to digital acquisition systems (est. required investment - $1 M) Initiate the design, development and implementation of an in situ (flight) calibration process. A calibration/verification process must be established for satellite imagery. Develop a U. S. Standard for camera and sensor calibration. Adequate funding should be sought in order to ensure the continued operation of the Optical Science Laboratory (OSL), as well as to provide for the improvements and extensions described in the preceding recommendations. U. S. Department of the Interior U. S. Geological Survey 3

The USGS Responds…. . U. S. Department of the Interior U. S. Geological Survey 4

One: Continue Operating OSL 1. “The USGS Optical Science Laboratory (OSL) should continue to calibrate film mapping cameras using the present calibrator and the Simultaneous Multiframe Analytical Calibration (SMAC) program. ” Ongoing today USGS has invested in upgrades and continues to do so Hardware upgrades, electronics, software USGS in year-long search for glass plates Finally identifying plates that will work Invested in multi-year glass plate supply Continuing to investigate film and electronic alternatives The USGS is committed to continuing OSL operations foreseeable future U. S. Department of the Interior U. S. Geological Survey 5

Two & Three: Research/Build Digital Capabilities 2. 3. “Initiate the design, development, and implementation of a digital camera calibration capability at the USGS (estimated required investment - $4 M). ” “Conduct research efforts in order to support a reliable and cost effective transition to digital acquisition systems (estimated required investment $1 M)” USGS has already invested in the following research capabilities Pictometry for Small/Medium-Format calibration cage OSU for In-Situ methodology for digital (and potentially film? ) SDSU-developed MTF tools; lab & in-situ Requesting funding for additional capabilities Further develop & validate in-situ Build & expand USGS sensor laboratory at EDC Further work on MTF tool (see more in Recommended Plans) U. S. Department of the Interior U. S. Geological Survey 6

Four: In-Situ Calibration 4. “Initiate the design, development and implementation of an in situ (flight) calibration process. ” USGS working with OSU, SDSU, industry, manufacturers to test and validate several different methods Developed in-situ ranges for independent validation Incorporates EROS instrumentation capabilities USGS actively involved with industry and NASA SSC in Digital Airborne Product Verification (see Phil Rufe’s presentation) Digital Sensor manufacturers working this also Very promising! U. S. Department of the Interior U. S. Geological Survey 7

Five: Satellite Data 5. “A calibration/verification process must be established for satellite imagery. ” USGS, partnered with NASA and NGA, leading the Joint Agency Commercial Imagery Evaluation (JACIE) team. On-going for four years now Hosted at USGS Headquarters in Reston JACIE evolving to address airborne and international datasets USGS leading CRSSP implementation USGS also involved in other Satellite Cal/Val Landsat(s) 5 & 7, EO-1 ALI and Hyperion, LDCM Active in CEOS WGCV, Euro. SDR, ASPRS/ISPRS RObotic Lunar Observatory (ROLO) U. S. Department of the Interior U. S. Geological Survey 8

Six: Develop Standards 6. Develop a U. S. Standard for camera and sensor calibration. USGS has had standards for other data for years Existing National Map Accuracy Standards served paper maps for decades DOQ standards served initial digitized aerial imagery The limited capabilities of the sensors and processes originally used in producing digital orthoimagery led to simpler standards (primarily geometry) The USGS will lead an effort to define new Digital Remote Sensing Data Standards Desire ASPRS and industry participation Possibility for international standards w/ ISPRS Data Standards Activity can begin as soon as possible! U. S. Department of the Interior U. S. Geological Survey 9

Seven: USGS Funding 7. Adequate funding should be sought in order to ensure the continued operation of the Optical Science Laboratory (OSL), as well as to provide for the improvements and extensions described in the preceding recommendations. OSL calibrations are paid by fees charged to camera owners Fees likely to increase in FY 06 This will be first fee increase since 1999 USGS Funding has been limited Has limited work on digital and other areas Project is established and ready for growth l U. S. Department of the Interior U. S. Geological Survey Optimistic for future 10

Digital Aerial Camera Observations There are: Many classes of cameras/systems Many potential users & uses of data Growth of Digital Aerial Sensors has skyrocketed Many attempting to exploit this new technology Digital aerial is waiting to “take off” Some digital systems equal the geometric performance of traditional film cameras Tend to be the more expensive systems; cheaper systems may be capable in future l These capabilities were not even available in 1999! Some manufacturers have good calibration in place already Digital can offer benefits not found in film Many more benefits possible; need applications development U. S. Department of the Interior U. S. Geological Survey 11

The Problem with Digital Cameras Digital offers amazing possibilities to aerial mapping – very promising Digital technology is evolving at incredible pace Technologies are very diverse: pushbroom, staring array, multispectral, multi-sensor, etc. Many Different Technologies all feed into digital imaging Sensor, Inertial Measurement Unit, GPS, processing tools The USGS believes it is impossible to cover all of these technologies and capabilities with one single “umbrella” camera calibration standard So where do we go? U. S. Department of the Interior U. S. Geological Survey 12

The Future: The Proposed USGS Plan for Digital Data Quality Assurance U. S. Department of the Interior U. S. Geological Survey 13

The USGS Proposed Plan Focused on Sensors Focused on Processes Focused on the End Products U. S. Department of the Interior U. S. Geological Survey 14

Focus on Sensors - “Sensor Systems” Without sensors there are no processes or products Increasing capabilities and complexities will bring challenges The USGS will offer “Type Certification” of digital systems Certification includes not only sensors and associated systems, but also the entire “chain” USGS will work with manufacturers & developers to certify chain(s) e. g. Leica ADS-40 with ADS 40 -specific calibration technique(s), ADS 40 processing software, etc. U. S. Department of the Interior U. S. Geological Survey 15

Focus on Processes All products are the result of processes The quality of aerial mapping imagery is at least as dependent on the processes as on sensors Process control at least as important to customers as sensors Processes make/break the quality chain Quality demands that processes are controlled – and documented U. S. Department of the Interior U. S. Geological Survey 16

Focus on Processes (cont. ) Processes include: Which sensors/systems – configuration control Maintenance of sensor/systems Calibration procedures, history Data handling, particularly steps where data is manipulated Processing software versions, etc. Any other external inputs All of this information must be documented, ideally in the metadata associated with the final product U. S. Department of the Interior U. S. Geological Survey 17

Focus on the End Products The ultimate goal is to assure the quality of the final product Previous work with Film Cameras was only with sensors: Camera Calibration National Map Accuracy Standards focused on end products Work in between governed only by professionalism and dedication And ground control Digital world has many different processes! U. S. Department of the Interior U. S. Geological Survey 18

“Classes” of Digital Data Industry needs “Classes” of Digital Products Applies to those generating, processing, selling, buying and working with these data Can apply to aerial, satellite, even lead to standards for nonimaging products, i. e. LIDAR. Will develop different quality “classes” of digital mapping data Based on resolution & accuracy Initial Classes for geometric and spatial quality Followed soon (? ) by radiometric quality USGS offers to lead standards development, with ASPRS/ISPRS U. S. Department of the Interior U. S. Geological Survey 19

Examples of Data Classes The following are notional examples to illustrate what data classes could be: Class 0: No geometric/geodetic accuracy implied, “Pretty pictures” Class A: Ground Sample Distance (GSD) > 1. 0 meter, geometric accuracies <0. 5 pixel size Class B: Ground Sample Distance (GSD) < 1. 0 meter, geometric accuracies <0. 4 m Class C: GSD < 0. 5 m, geometric/geodetic <0. 2 m Class D: GSD < than 0. 15 m, geometric/geodetic <0. 1 m Class E: (Reserved for future capabilities) U. S. Department of the Interior U. S. Geological Survey 20

What This All Means USGS to work with sensor/system manufacturers Proving sensor capabilities and “calibrate-ability” now falls on manufacturers Hardware/Software Certification burden is transferred from many individual flyers to (relatively) few manufacturers Allows more in-depth understanding of each system, working directly with technologists Helps keep burden of hardware/software performance on those who developed it USGS to work with ASPRS to promote education of new standards for use by consumers U. S. Department of the Interior U. S. Geological Survey 21

What This All Means (cont. ) Consumers benefit by clearer, defined choices for data Easier procurement/contracting Burden on flyers/producers now shifted to processes Insists that defined processes be carried out and documented Aids in troubleshooting, maintaining high standards Flyers are/should be doing this anyway USGS Certification means more: A Greater Assurance of Data Quality! U. S. Department of the Interior U. S. Geological Survey 22

Work Needed to be Done USGS to establish contacts points for manufacturers, begin interaction Funding – the longer we wait, the more it costs USGS partially funded in FY 05 Establish IADIWG funding mechanisms required for FY 05 USGS and IADIWG need to define funding mechanisms for FY 06 and FY 07 Define infrastructure needed In-situ ranges Hardware and Software requirements Laboratories – radiometry especially Establish guidelines, policy, standards, and boiler plate specifications IADIWG involved in standards development & approval IADIWG to begin specifying data “classes” in their work, requesting USGS standards, educating users & industry U. S. Department of the Interior U. S. Geological Survey 23