make.opendata.ch wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
===== KLIO Picture Labeler =====

{{ :project:kpl-head.jpg?500 |}}

Historical archives often have to deal with a huge amount of photographs and other historical images. Once they are digitalized, they need to get categorized: What is pictured on them?

This application is an experimental approach to use automated image-recognition for this task.

This first version runs with [[https://cloud.google.com/vision/|Google Vision]]. A further goal is to integrate other picture-recognition-APIs as well to get better results (e.g. [[https://azure.microsoft.com/en-us/services/cognitive-services/computer-vision/|Microsoft Computer Vision]], [[https://www.ibm.com/watson/developercloud/visual-recognition.html|IBM Visual Recognition]], etc.).

===== Data =====

  * Source-Code: [[https://github.com/zack17/klio-picture-labeler]]
  * Demo-Page: [[https://zack17.github.io/klio-picture-labeler/]]

=== Foto Sources ===

The used demo photos were taken from:
  * Indien-Bilder der Handelsfirma Gebrüder Volkart, Stadtarchiv Winterthur, [[https://data.stadt-zuerich.ch/dataset/winterthur_sar_volkart|Link]]
  * Bildarchiv des Lebensmittelverein Zürich (LVZ), Sozialarchiv Zürich, [[https://data.stadt-zuerich.ch/dataset/sozialarchiv-lvz|Link]]
  * Fotosammlung: Wandervogel - Schweizerischer Bund für alkoholfreie Jugendwanderungen, Sozialarchiv Zürich, [[https://data.stadt-zuerich.ch/dataset/sozialarchiv-wandervogel|Link]]

All licensed under [[https://creativecommons.org/licenses/by/4.0/|CC-by]]


===== Team =====

  * Micha Reiser | [[https://github.com/MichaReiser|GitHub]]  
  * Dominic Studer ([[user:doemee]]) 
  
{{tag>status:concept needs:dev needs:design needs:data needs:expert}}