Retrieving Lost Community Stories -- Linking Regional Archival Photo Collections using Advanced Visual Technologies — Digital Humanities Australasia 2018

Retrieving Lost Community Stories -- Linking Regional Archival Photo Collections using Advanced Visual Technologies (125)

Lei Wang 1 , Hilary Powell 2
  1. university of wollongong, Wollongong, NSW, Australia
  2. Wollongong City Libraries, Wollongong, NSW, Australia

This is a collaborative project between the University of Wollongong (UOW), UOW Library, Illawarra Museum and Historical Society, and Wollongong City Libraries supported by the UOW Community Engagement Grants Scheme. It is a scalable, sustainable, and long-term project that can be continuously improved and expanded to link larger regional, state and national collections.

The aim of the project is to use advanced visual technologies to improve access to images, often with poor or limited descriptive metadata, held in three different archival photo collections. The result is the development of an interactive website – ImageLink - to match photos across the three collections based on how they look, not the words that have been used to describe them, enabling the community to easily explore and engage with local social, cultural, environmental and economic history through images.

Computer algorithms are used to directly match photos held in the three collections. Community members can explore the collections by clicking query images to find similar images across the collections. Alternatively, users may search these linked collections by uploading unidentified images to find similar images and information. In the near future, the project is going to further enhance the website by allowing the community to annotate unidentified photos, cluster similar photos, and determine unique ones.

This project aligns well with the aim of DHA 2018 “to advance and critically assess the uses of digital technologies in humanities research.” It evaluates the state-of-the-art computer algorithms for archival photo search and develops new algorithms to further improve the search performance and experience. This project fits the theme of DHA 2018 (i.e., making connections) in that it connects the images from different archival collections; the knowledge from different disciplines; and the people from different organisations. In addition, this project demonstrates the synergy between computer science and digital humanities by working with local and regional communities, which is again a focus emphasised in DHA 2018.

This project is also supported by the Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) and undertaken with the assistance of resources from the National Computational Infrastructure (NCI), which is supported by the Australian Government.

Explore ImageLink - https://ssl.informatics.uow.edu.au/naair/CEGS_Home.php

This project was reported by WINNEWS at https://youtu.be/G2dTXn0w5O8 

#DHA2018