Use Of Browse Or Preview Images In Support Of Data Access PV-2004 5-7 October 2004 Frascati, Italy S. W. Doescher, U. S. Geological Survey R. H. Sunne,

Use Of Browse Or Preview Use Of Browse Or Preview Images In Support Of Data Images In Support Of Data

AccessAccess

PV-20045-7 October 2004

Frascati, Italy

S. W. Doescher, U. S. Geological Survey R. H. Sunne, Science Applications International Corporation

M. Neiers, Science Applications International Corporation

AgendaAgenda

• Introduction• Strategy/philosophy• Early definitions • Radiometric adjustment• Resolution reduction• Opportunities for browse refinement • Current uses of browse• CEOS community characterization of browse

usage • Observations

IntroductionIntroduction

• For more than 30 years, the U. S. Geological Survey (USGS) Earth Resources Observation Systems (EROS) Data Center (EDC), Sioux Falls, SD, has provided access to multiple terabytes of remotely sensed imagery of the earth's surface.

• In 1990s, the advent of the Internet and online systems created a need for browse (or preview imagery) to assist with locational identification and quality review of the archival holding of remotely sensed data.

Strategy/philosophyStrategy/philosophy

• Purpose of Browse– To provide user assistance in determination of

usability of an unique inventory item

• Reasons for browse– Determining extent and location of clouds– Observing quality problems such as

speckling, line drops, and sun glint– Confirming geographical location and area of

coverage

Strategy/philosophyStrategy/philosophy

• General Philosophy – Browse prepared independently of characteristics of

devices used to display – Provide similarity of browse across different inventory– Targeted size of browse imagery is less than .5

megabytes– Produced by reduction in spatial resolution and band

representation– Minimize impact of browse generation – Reference or locational information to be maintained

separately

Early definitionsEarly definitions

• The initial focus of browse – Advanced Very High Resolution Radiometer (AVHRR) – Landsat Multi-Spectral Scanner (MSS)– Landsat Thematic Mapper (TM)

• Early discussions on AVHRR browse options – Single band browse

• Every 4th line and every 5th sample• 10-bit data were reduced to 8-bit • Band 2 was used for daytime data, and band 4 was used for nighttime data • Resultant size 0.51 MB

– Multi-band browse • Same data reduction as the single band above, but for all bands• Provided a “custom" browse option • Potential “custom” examples

– color browse (uses recipe RGB:2,1,1) – normalized difference index or greenness image– Shark Classes (Sea, Sun Glint, Land, Cloud, Snow/Ice)

– Selection of the single band browse was driven by technology considerations

Early definitionsEarly definitions

• Similar discussion were held with Landsat

• For Landsat initially provided multi band– individual bands and – color composite options

• Usage quickly determined that color option was preferred

Radiometric adjustmentRadiometric adjustment

• Typically dynamic range of the image is narrow and stretching adjustments can improve the visibility for the user

• The browse stretch points are determined by using the histogram of the browse image

• The lower stretch point ranged between the points of 2.5 percent and 97.25 percent of the accumulated histogram

Radiometric adjustmentRadiometric adjustment

Without radiometric adjustment

With radiometric adjustment

Resolution reductionResolution reduction

• Early processing reduced resolution by simple sub sampling process

• In the 1995, wavelet processing uses a convolution filter that preserves the high frequencies or edges as the filter is passed over the image. A separate pass of the wavelet process was required for a halving of the resolution. Three passes required for eight-fold reduction

• Recent experimentation with a “pnmscale” produced comparable results to the wavelet process, but requires less processing time than the three passes of the wavelet process

• In all cases, JPEG, with a quality factor of 75, was selected as the compression method and delivery format

Resolution reductionResolution reduction

Sub sample14 sec

Wavelet31 sec

Pnmscale24 sec

Opportunities for browse Opportunities for browse refinementrefinement

• In 1992, the EROS Data Center began operations of TM and MSS Archive Conversion System (TMACS) – Transcribe high-density tape to DCRSi Cassette Tapes (DCT) output. – MSS sub sampled every 6th line and every 6th sample– TM sub sampled at every 16th line and every 16th sample

• Presently, starting operations of the Landsat Archive Conversion System (LACS)– Transcribe DCTs to 9940B tape – MSS via “pnmscale” reduced by factor of 4– TM via “pnmscale” reduced by factor of 8

• Processing applied to three bands for RGB color composite• The reduced resolution image bands are stretched, composited, and

JPEG compressed with a quality factor of 75 to a generation formatted browse

• A JPEG comment field is added with the appropriate metadata information

Current uses of browseCurrent uses of browse

• Traditionally, Earth Observing search systems– User specifies desired characteristics– Metadata inventories are searched– Search results contain a pointer to the browse – Example: Earth Explorer http://earthexplorer.usgs.gov

• Recently, some systems have evolved to “browse first.” – User with an image-based mechanism– browse used as primary mode of guidance– Example: Glovis http://glovis.usgs.gov

Earth ExplorerEarth Explorer

Earth ExplorerEarth Explorer

Earth ExplorerEarth Explorer

GlovisGlovis

GlovisGlovis

GlovisGlovis

GlovisGlovis

GlovisGlovis

Monthly Browse UsageMonthly Browse UsageLandsat Browse Usage

0

50000

100000

150000

200000

250000

300000

350000

Apr/96 Apr/97 Apr/98 Apr/99 Apr/00 Apr/01 Apr/02 Apr/03 Apr/04

Brow

se/M

onth

MSS T M Landsat 7

CEOS community characterization CEOS community characterization of browse usage of browse usage

Committee on Earth Observation Satellites (CEOS) community– Civil agencies involved in earth observation

activities– Provided updates to browse characteristics

available for remotely sensed data– More than 60 sensor instruments

characterized in paper

Observations Observations

• Over the last 10 years of browse usage– Cost of online storage has decreased– End user’s network bandwidth has increased– Resolution and composition of browse has improved

• However– archives have grown from terabytes to petabytes– it is still cost prohibitive to put all data online

• There continues to be a need for browse or quick look products to facilitate the quality and locational review of the archival holding of remotely sensed data