Enrichment and Search

Any asset registered in the IIIF Cloud Services platform can be accompanied by any number of supporting files, known as adjuncts.^{33p (opens in a new tab)}

Adjuncts are added via the portal or via the API. Any file can be an adjunct; the platform doesn't need to know what the file is. At the simplest level, the platform can be used to store and publish related files, for whatever purpose you need. Examples are transcriptions, or data generated from machine learning pipelines that are using the platform for storage.

If marked for publication, the adjunct will appear in the seeAlso, rendering or annotations property of any IIIF Canvases in Manifests generated from named queries, or from Manifests explicitly created. Again, this is convenient for storing and publishing any files related to individual assets, for any purpose.

The list of adjuncts also appears in the Single Asset Manifest - this is a IIIF Manifest available for every individual asset that uses IIIF to present all the files and services the platform can provide for an individual asset - image services, access control, adjuncts.^{488 (opens in a new tab)}

• You can supply the metadata and the content of an adjunct file directly via the API.
• You can register an adjunct using an origin, similar to an image asset, and the platform will fetch the file from the origin and store it.^??

Adjuncts providing text for search

If an adjunct is in a text format understood by the platform, it can be used to generate IIIF Search services. Formats include plain text, METS-ALTO, hOCR and W3C TextualBody annotations.

There are two types of Search Service provided by the platform.

• Manifest-level Search Within is available on any managed IIIF Manifest (not manifests generated from named queries). This is suitable for finding particular words and phrases within one book or document, and is supported by viewers like UV and Mirador.^{63p (opens in a new tab)}
• Fuzzy^name search is available on Collections, including virtual collections from queries^{61pc (opens in a new tab)} (e.g., search within a facet). It supports typical Lucene-derived search features. You would typically consume this in your own custom web applications rather than from a viewer.^{83p (opens in a new tab)}

Both types of search are provided as IIIF Search API endpoints.

Platform-generated adjuncts^{13p (opens in a new tab)}

Sometimes you may already have adjuncts, and can use the platform to store them (and sometimes generate search services from them). These may come from a digitisation workflow (e.g., OCR files), or they may come from ad hoc enrichment (e.g., a project generating datasets for each image, or AI-driven descriptions of each image).

But you can also get the IIIF Cloud Services Platform to generate adjuncts using a number of enrichment services such as:

• OCR
• Handwriting recognition
• Image analysis and description
• ...more to come

You can do this in the Portal or via the API.

Putting it together

This means that in the portal, or via the API, you can:

• Create a manifest by registering a set of images using any of the methods described in IIIF Images (these methods apply to any asset type not just images).
• Specify you want the platform to generate OCR data from the images.
• Specify you want the platform to attach a IIIF Search API service to the Manifest.

These latter two behaviours can be set as defaults, so you could simply go to the portal, select some images by browsing your own Dropbox or Google Drive, and immediately create a published, searchable IIIF Manifest that works in any viewer.

touched 2025-09-23T12:04:54