HMA-T Phase 2 Call

HMA-T Phase 2 Call

HMA FO Task 1 : EO Metadata Extensions Frdric Houbie, ERDAS Steven Smolders, GIM Andrew Woolf, STFC Dominic Lowe, STFC 6 July 2010, ESA (Frascati) Overview Planning Task description Deliverables EO Product Extensions XML Schemas for EO Products Cataloguing of EO Product

OGC standardisation status 6 July 2010, ESA (Frascati) Slide 2 Task 1 Planning We are here 6 July 2010, ESA (Frascati) Slide 3 Tasks description EO Metadata Analysis of the extensions update of the EO Metadata, update of the cataloguing model OGC Process EO Metadata v2 Cataloguing of EO Metadata v2 using ebRIM model 6 July 2010, ESA (Frascati) Slide

4 MTR deliverables Analysis of the extensions (HMAFOT1-TN-0002-ERDAS-15.doc) http://wiki.services.eoportal.org/tiki-download_wiki_attachment.php? attId=830&download=y O&M Extension for EO Product http://wiki.services.eoportal.org/tiki-download_wiki_attachment.php? attId=831&download=y UML Model & Documentation http://wiki.services.eoportal.org/tiki-download_wiki_attachment.php? attId=833&download=y Should be upload as a web site instead of an archive Currently on http://projects-eu.erdas.com/HMAFO/doc/ ebRIM Cataloguing for EO Products http://wiki.services.eoportal.org/tiki-download_wiki_attachment.php? attId=832&download=y 6 July 2010, ESA (Frascati) Slide 5

EO Product Types Base EO Product Optical Radar Atmospheric Altimetry Limb Looking Synthesis & Systematic Products 6 July 2010, ESA (Frascati) Slide 6 EO GML extension : schemas

Current version based on GML 3.1.1 Schema implementation options straightforward migration to GML 3.2.1 model-driven approach to GML 3.2.1 O&M more extensive refactoring 6 July 2010, ESA (Frascati) Feedback by questionnaire First list of requirements, then questionnaire Purpose : get input from stakeholders End of April, distributed to all HMA people involved in the task (consortium, agencies, MD workshop participants) Set of responses received end of May The question about schemas was solved mid of June during the HMA AWG at DLR

6 July 2010, ESA (Frascati) Slide 8 Model-driven approach Agreed to follow model-driven approach UML model is the main artefact Schemas, documentation, etc. auto-exported Follows UMLGML encoding rules (ISO 19136:2007 Annex E) Implications Must not design at XML level (instead at UML level) E.g. cant add XML attributes to schema The tools are integral UML model is shared and may be updated by any approved committers to svn (via Enterprise Architect) Something of a culture change 6 July 2010, ESA (Frascati) Slide 9 UML Model http://projects-eu.erdas.com/HMAFO/doc/ 6 July 2010, ESA (Frascati) Slide 10 O&M: status, implications (1/3)

Published 4-Jun-2010 as ISO/DIS 19156 Mature now available for purchase! Agreed by stakeholders and AWG to base HMA on OM_Observation rather than gml:Observation Some differences with respect to GML model (next slide), BUT... HMA uses only very small core of O&M, THEREFORE... ISO/DIS 19156 very unlikely to change in a way that impacts HMA 6 July 2010, ESA (Frascati) Slide 11 O&M: status, implications (2/3) The majority of changes occur with adoption of GML 3.2.1: change of namespace on all GML elements addition of gml:id in many places move of EarthObservationMetadata Only six additional changes for O&M: 1. gml:validTime om:PhenomenonTime 2. mandatory om:resultTime

but could provide standardised replacement for eop:ProcessingDate, or set to unknown 3. gml:using om:procedure 4. mandatory om:observedProperty but provides standardised slot for physical parameter info (LIM, ALT, OPT), and is nillable 5. gml:target om:featureOfInterest 6. gml:resultOf om:result .... ....

.... .... .... 1.2.1 Slide 12 DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234 NOMINAL TBD ACQUIRED TLS

TLS 2001-08-22T11:02:47.999 2001-08-22T11:02:47.000 2001-08-22T11:02:47.999 PHR 1A PHR OPTICAL PX

0.7 12 12 ASCENDING 12 -14.0 -13.9 0 0 0 2.1025 43.516667 2.861667 43.381667 2.65 42.862778 1.896944 42.996389 2.1025 43.516667

2.374167 43.190833 QUICKLOOK epsg:4326 http://eop.cnes.fr/catalog/DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234.jpg CLOUD VECTOR http://eop.cnes.fr/catalog/DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234.gml example eop instance (06-080r4) xsi:schemaLocation="http://earth.esa.int/eop ../../../OGC-06-080%20(GML%20EO%20App%20Schema)/reverse-engineered_06-080r4/xsd_gml3.2/eop.xsd" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:gml="http://www.opengis.net/gml/3.2" xmlns:eop="http://earth.esa.int/eop" gml:id="DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234"> 2001-08-22T11:02:47.000

2001-08-22T11:02:47.999 PHR 1A PHR OPTICAL PX 0.7 12 12 ASCENDING 12 -14.0 -13.9 0 0 0

2.1025 43.516667 2.861667 43.381667 2.65 42.862778 1.896944 42.996389 2.1025 43.516667 2.374167 43.190833 QUICKLOOK epsg:4326 http://eop.cnes.fr/catalog/DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234.jpg

CLOUD VECTOR http://eop.cnes.fr/catalog/DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234.gml DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234 NOMINAL TBD ACQUIRED TLS TLS 2001-08-22T11:02:47.999 1.2.1 GML3.2 changes namespace changes (30)

mandatory gml:id (7) move EarthObservationMetadata gml:id="DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234"> 2001-08-22T11:02:47.000 2001-08-22T11:02:47.999 PHR 1A PHR OPTICAL PX 0.7

12 12 ASCENDING 12 -14.0 -13.9 0 0 0 2.1025 43.516667 2.861667 43.381667 2.65 42.862778 1.896944 42.996389 2.1025 43.516667 2.374167 43.190833

QUICKLOOK epsg:4326 http://eop.cnes.fr/catalog/DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234.jpg CLOUD VECTOR http://eop.cnes.fr/catalog/DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234.gml DS_PHR1A_20010822110247_TLS_PX_E123N45_0101_01234 NOMINAL TBD ACQUIRED TLS TLS

2001-08-22T11:02:47.999 1.2.1 extra O&M changes om:PhenomenonTime om:resultTime om:procedure om:observedProperty om:featureOfInterest om:result O&M XML (1/2) XML schemas proposed in OGC document 10-025 Observations and Measurements XML Implementation Most of the document irrelevant for HMA Only one of the 18 requirements/conformance classes is relevant: Requirements class: Generic observation data: http://www.opengis.net/spec/OMXML/2.0/req/observation Target type Data instance Dependency http://www.opengis.net/doc/IS/GML/3.2/clause/2.4 Dependency urn:iso:ts:iso:19139:clause:6

Requirement http://www.opengis.net/spec/OMXML/2.0/req/observation/observationvalid Must be schema valid for OM_Observation Any XML element in the substitution group of om:OM_Observation SHALL be well-formed and valid Requirement http://www.opengis.net/spec/OMXML/2.0/req/observation/result-typeconsistent The content model of any om:result element SHALL be consistent with the value of the xlink:href attribute of the om:type element if one is present as a sub-element of the parent om:OM_Observation, according to the mapping given in Table 5. 6 July 2010, ESA (Frascati) Must have correct result type for O&M specialised observations (irrelevant for HMA) Slide 14 O&M XML (2/2) Changes incorporated into O&M XML for HMA: Observation procedure (EarthObservationEquipment) may be inline Previously, was by-reference only

observedProperty is nillable Thus, for the two new mandatory elements, may use (if necessary): unknown resultTime nill observedProperty indeterminatePosition="unknown"/> 6 July 2010, ESA (Frascati) Slide 15 Tools support (1/2) Model-driven approach means automated schema generation from UML model

Two open-source tools available: FullMoon Java, based on Xquerys over XML (stored in eXist database) Available from: http://projects.arcs.org.au/trac/fullmoon/ O&M XML now supported ShapeChange Java, based on EA Automation API Available from: http://www.interactive-instruments.de/index.php?id=28&L=1 Significant ongoing development in context of INSPIRE used for Annex I will eventually support O&M XML since it will be used for Annex II/III 6 July 2010, ESA (Frascati) Slide 16 Tools support (2/2) Tool differences Both tools used during HMA-FO Currently using FullMoon Slight differences in model stereotypes (e.g. FeatureType instead of featureType) and tagged values Maybe 1-2 days effort required to migrate a model from one to the other No substantive differences in schemas produced

6 July 2010, ESA (Frascati) Slide 17 Schema issues (1/4) Circular geometry Use gml:CircleByCenterPoint {instead of | in addition to} existing Footprint geometry Overall quality flag Use EOMetadata.status and extend StatusValue (ACQUIRED, ARCHIVED, CANCELLED, FAILED, PLANNED, POTENTIAL, REJECTED, QUALITYNOTACCEPTABLE)? Also, O&M provides an EarthObservation.resultQuality element ATM error in DataLayer Propose to correct English: specy species Then, include new attribute speciesError:Measure 6 July 2010, ESA (Frascati) Slide 18 Schema issues (2/4) Parameter information Not quite finalised Use SWE Common v1.0 Phenomenon model

gml:ReferenceType for referring to external parameter definitions (e.g. CF Standard names) class Figure: phenomenon +base 1 Type PropertyType +base +base 1 0..1 Type ConstrainedPropertyType + + OWS Service Reference otherConstraint: CharacterString [0..*] singleConstraint: Any [0..*] T ype PropertyTypeSeries 1..* +component

T ype CompoundPropertyType + T ype CompositePropertyType +/ constraintList: Sequence [1..*] + otherConstraint: CharacterString [0..*] No standardised UML available Implementing direct mapping in FullMoon tool 6 July 2010, ESA (Frascati) Composition dimension: Integer Slide 19 Schema issues (3/4) Schematrons being finalised Need to add ALT, LMB, SSP Schematron support now pretty good (e.g. within editor), TeamEngine should have support for schematron SSP Footprint uses geographicalName attribute Probably should use gml:locationName

Class names distinguished by namespace Nevertheless, is there a case to use product prefixes, e.g. ATM_EarthObservationResult? 6 July 2010, ESA (Frascati) Slide 20 ATS/ETS Linked to requirements in the documents Requirements for Schemas can be as easy as validation against XML Schemas & Schematron rules (J. Herring, S. Cox) Extract from OMXML Requirement Class http://www.opengis.net/spec/OMXML/2.0/req/observation Target type Requirement Data instance

Requirement http://www.opengis.net/spec/OMXML/2.0/req/observation/result-type-consistent http://www.opengis.net/spec/OMXML/2.0/req/observation/observation-valid Any XML element in the substitution group of om:OM_Observation SHALL be well-formed and valid The content model of any om:result element SHALL be consistent with the value of the xlink:href attribute of the om:type element if one is present as a sub-element of the parent om:OM_Observation, according to the mapping given in Table 5. 6 July 2010, ESA (Frascati) Slide 22 ATS/ETS (2) Conformance Class 6 July 2010, ESA (Frascati) Slide 23 EO Product Cataloguing Current version is OGC Standard 1.0 EO Product Extension Package of ebRIM 3.0

Used with CS-W 2.0.2 & ebRIM AP 1.0 6 July 2010, ESA (Frascati) Slide 25 Current Model (EOP) 6 July 2010, ESA (Frascati) Slide 26 Issues of the current model Complexity of the model structure Maintainability and extensibility of the EO EP Full mapping = EO GML duplication Simple change in cardinality impacts the model Adding new metadata element even if they are not queryable (ex : add several product locations with different formats) impacts the model 6 July 2010, ESA (Frascati) Slide 27 Retrieving Metadata Number of requests to display 100 results:

1 GetRecords + 100 GetRepositoryItem = 101 requests 6 July 2010, ESA (Frascati) Slide 28 Improvements Simplify model Improve GetRecords operation 6 July 2010, ESA (Frascati) Slide 29 Simplify model <> <> EOProductTypes <> EOP <> SAR <> OPT <> ATM

<> LIM <> <> <> EOProduct +objectType: urn:...:EOProduct +parentIdentifier: String +beginPosition: Date +endPosition: Date +imageQualityDegradation: Double +acquisitionType: String +acquisitionSubType: String +orbitDirection: String +status: String +orbitNumber: int +productType: String +acquisitionStation: String +lastOrbitNumber: int +archivingCenter: String +multiExtentOf: Geometry +cloudCoverPercentage: int +snowCoverPercentage: int <> ALT Slots to add ? - startWavelength:GMLMeasure - endWavelength: GMLMeasure - spectralRange: [VISIBLE, INFRARED, UV, NEAR-INFRARED]

<> ExternalIdentifier Content GML O&M Metadata File File <> EOAcquisitionPlatform <> AcquiredBy source <> SSP +associationType: urn:...:ArchivedIn target The archivingCenter have been added to the EOProduct since the EOArchivingInformation has only one queryable slot. +objectType: urn:...:EOAcquisitionPlatform +name: String +instrumentShortName: String +platformSerialIdentifier: String +sensorType: String

+sensorOperationalMode: String +sensorResolution: Double +swathIdentifier: String cloudCoverPercentage and snowCoverPercentage slots are only relevant for OPT, ATM and SSP products types Slots to add ? - qualityFlag: [OK, NOK] 6 July 2010, ESA (Frascati) Slide 30 Improved GetRecords operation 6 July 2010, ESA (Frascati) Slide 31 OGC Catalog structure 6 July 2010, ESA (Frascati) OGC standardisation targets O&M Extension for EO Products New name : Earth Observation Metadata profile of Observations & Measurements Profile of O&M Keep 06-080 ? No ebRIM Cataloguing for EO Products

Profile of ebRIM No direct link with CS-W, could be used without 6 July 2010, ESA (Frascati) Slide 33 Proposed OGC Fast Track process Ability to have approval of a candidate standard facilitated without formation of a Standards Working Group. The objective is to shorten the approval time but still have TC consensus. For profiles of existing OGC encoding standards (such as GML and SensorML) For widely implemented de-facto standards developed external from the OGC. 6 July 2010, ESA (Frascati) Two forms of Fast Track process Short No public comment period OGC Members can provide comments. Slightly longer 30 day public comment period If comments, comments shall be responded to. Good candidates ?

6 July 2010, ESA (Frascati) Whats next Finalizing model & schemas Testings HMA Skeleton CITE Tests update Team Engine OGC Trunk ERGO branch with SOAP support Copy of ERGO branch in HMA SVN with asynchronous support Local Erdas copy with logging Needs cleaning 6 July 2010, ESA (Frascati) Slide 36

Recently Viewed Presentations

  • @MrJohnBates TRILCon 2017 > Autonomous Vehicles HOW FAR

    @MrJohnBates TRILCon 2017 > Autonomous Vehicles HOW FAR

    work should help to counter the possible disadvantages and negative effects of managing a mixed fleet of autonomous and non-autonomous vehicles. The research on human factors must feed in to the modelling work so that as more is understood about...
  • CHAPTER 2 MATTER - Fox Valley Lutheran High School

    CHAPTER 2 MATTER - Fox Valley Lutheran High School

    PURE SUBSTANCE or MIXTURE? PURE SUBSTANCE fixed _____ definite _____ all the elements (listed on the periodic table) all compounds (all have a formula) Examples: distilled water, salt… MIXTURE a combination of more than one ___ ___ can be separated...
  • Chapter 19: Death and Dying - Pearson Education

    Chapter 19: Death and Dying - Pearson Education

    Chapter 19: Death and Dying Development Across the Lifespan Death is a universal experience, one that we will all eventually face Despite this, the topic has only recently begun to be studied by developmental psychologists Some things that are being...
  • Fuel Poverty in Argyll Rachel McNicol Fuel Poverty

    Fuel Poverty in Argyll Rachel McNicol Fuel Poverty

    ALIenergy's affordable warmth work. Fuel Poverty and what it really means. How fuel poverty affects individuals and their health. Fuel poverty in Argyll and Bute. What can ALIenergy do for people in fuel poverty
  • PowerPoint Sunusu

    PowerPoint Sunusu

    İftitah tekbiri. Hadesten taharet. Kıyam. Kıraat. Setr-i Avret. Rüku. Verilen namazın şartlarının hangileri içindeki şartlar hangileri dışındaki şartlar. Hadesten taharet. Necasetten Taharet. ... Bütün namazların ilk rekatlarında okunan dua .
  • Data Privacy - Simons Institute for the Theory of Computing

    Data Privacy - Simons Institute for the Theory of Computing

    Can we rigorously analyze effect of privacy even when non-private algorithms don't have rigorous analyses? e.g. in hypothesis testing, privacy needs at most ?(1?) blow-up in sample size… but this is suboptimal [Cai-Daskalakis-Kamath `17]. Lower Bounds. Most existing techniques prove...
  • Using Google Forms for Special Education Data

    Using Google Forms for Special Education Data

    Live Form. Easy to use data entry tools. Data can be input by one or many. What are Google Forms? Results form. Information is instantly sent to a pre-built spreadsheet
  • THE ROMAN REPUBLIC - Kyrene School District

    THE ROMAN REPUBLIC - Kyrene School District

    This system is like the veto, which limited the power of Roman consuls. In addition, like Rome, the United States has a written constitution on which its government is based. Citizenship is also an important part of a republican government....