Designing and Creating Content

​Images, including high definition and ultra high-definition images, may be uploaded by the LINKED CANVAS image management tool. Images are stored using the new International Image Interoperability Framework (IIIF). Legacy image metadata can be ingested and combined with new metadata as part of the image cataloging process. IIIF enables images and points of interest to be accessed and annotated via HTTP. Leveraging IIIF's open data APIs, it is even possible to annotate and index externally-curated images (stored on third-party IIIF image servers) alongside internally-curated images.

Step 1: Image Management

​Images, including high definition and ultra high-definition images, may be uploaded by the LINKED CANVAS image management tool. Images are stored using the new International Image Interoperability Framework (IIIF). Legacy image metadata can be ingested and combined with new metadata as part of the image cataloging process. IIIF enables images and points of interest to be accessed and annotated via HTTP. Leveraging IIIF's open data APIs, it is even possible to annotate and index externally-curated images (stored on third-party IIIF image servers) alongside internally-curated images.
Ontologies and controlled vocabularies underpin knowledge discovery and contextual analysis. LINKED CANVAS is a semantic web application that includes an integral suite of ontology and concept scheme management tools. These enable external Linked Open Data resources, such as Getty Vocabularies, Iconclass, Europeana, GeoNames, DBpedia and WIKIDATA, to be used alongside internally-developed ontologies and knowledge organization systems. Adopting and reusing Linked Open Data authorities not only speeds up development timelines and reduces project costs, it also creates a framework for further content enrichment and collaboration on a global scale.

Step 2: Ontology Management

Ontologies and controlled vocabularies underpin knowledge discovery and contextual analysis. LINKED CANVAS is a semantic web application that includes an integral suite of ontology and concept scheme management tools. These enable external Linked Open Data resources, such as Getty Vocabularies, Iconclass, Europeana, GeoNames, DBpedia and WIKIDATA, to be used alongside internally-developed ontologies and knowledge organization systems. Adopting and reusing Linked Open Data authorities not only speeds up development timelines and reduces project costs, it also creates a framework for further content enrichment and collaboration on a global scale.
Image annotation tools allow Points of Interest (POIs) to be highlighted, labeled and described. Each point of interest is assigned its own unique HTTP-URI, thereby supporting sub-image indexing and direct retrieval. The image annotation toolset supports several annotation methods. Rapid annotation techniques include using drop-pins to identify single coordinate points, or rectangles and circles to identify simple bound areas. Fine-tuned annotation techniques allow individual figurative details to be outlined and highlighted. All POIs allow for on-canvas captions and roll-over descriptions, as well as an innovative method called monochrome masking, which highlights each POI by masking the surrounding area with a monochrome filter.

Step 3: Image Annotation

Image annotation tools allow Points of Interest (POIs) to be highlighted, labeled and described. Each point of interest is assigned its own unique HTTP-URI, thereby supporting sub-image indexing and direct retrieval. The image annotation toolset supports several annotation methods. Rapid annotation techniques include using drop-pins to identify single coordinate points, or rectangles and circles to identify simple bound areas.  Fine-tuned annotation techniques allow individual figurative details to be outlined and highlighted. All POIs allow for on-canvas captions and roll-over descriptions, as well as an innovative method called monochrome masking, which highlights each POI by masking the surrounding area with a monochrome filter.
While annotations add value to images, conceptually indexing those annotations using ontologies facilitates a quantum leap forward in knowledge discovery and contextual analysis. Images and points of interest within images can be indexed using predicates and controlled vocabulary concepts from an extensible set of ontologies and knowledge organization systems. These ontologies build connections between artworks, artists and associated cultural resources.

Step 4: Semantic Indexing

While annotations add value to images, conceptually indexing those annotations using ontologies facilitates a quantum leap forward in knowledge discovery and contextual analysis. Images and points of interest within images can be indexed using predicates and controlled vocabulary concepts from an extensible set of ontologies and knowledge organization systems. These ontologies build connections between artworks, artists and associated cultural resources.
All data created inside Linked Canvas is stored as IIIF images or RDF triples. Using APIs, content can be published to existing websites and/or new applications. Content may optionally be published via a SPARQL endpoint. Alternatively, the tool can be used as a completely self-contained web authoring and web publication tool. This step includes components for: faceted search across an entire image collection; browsable image galleries of featured images; search inside individual images; on-canvas tooltips to describe specific visual details; arrangement of POIs spatially or narratively using POI hierarchies; query expansion and refinement using concept explorer and concept tag clouds; linking to related images based on ontological connections; instant pan and zoom from search results to highlighted visual features; pinch and swipe around images and watch the discovery panel dynamically update to describe the visual features and concepts currently in view; explore time-sequenced guided-tours including audio and video narration.

Step 5: Publication & Discovery

   
All data created inside Linked Canvas is stored as IIIF images or RDF triples. Using APIs, content can be published to existing websites and/or new applications. Content may optionally be published via a SPARQL endpoint. Alternatively, the tool can be used as a completely self-contained web authoring and web publication tool. This step includes components for:
  • faceted search across an entire image collection;
  • browsable image galleries of featured images;
  • search inside individual images;
  • on-canvas tooltips to describe specific visual details;
  • arrangement of POIs spatially or narratively using POI hierarchies;
  • query expansion and refinement using concept explorer and concept tag clouds;
  • linking to related images based on ontological connections;
  • pan and zoom from search results to specific visual features;
  • pinch and swipe images and watch the discovery panel dynamically update to describe the concepts associated with what is currently in view;
  • explore time-sequenced guided-tours including audio and video narration.