Cuneiform-Sign-Detection-Webapp

This repository contains the web front-end of the web application presented in the article:

Dencker, T., Klinkisch, P., Maul, S. M., and Ommer, B. (2020): Deep Learning of Cuneiform Sign Detection with Weak Supervision using Transliteration Alignment, PLOS ONE, 15:12, pp. 1–21 https://doi.org/10.1371/journal.pone.0243039

The web front-end offers the following functionality:

• create collections of tablet images
• upload tablet images
• apply the sign detector
• visualize sign detections
• annotate cuneiform signs
• annotate lines

The web front-end has been developed using a combination of PHP and JavaScript.

Requirements

• Apache web server (otherwise replace .htaccess files)
• PHP7 (with php-xml, php-curl, php-zip, php-gd packages)

Installation

1) Create a copy of this repository on your machine so that the installed web server makes the web front end available through the browser.

2) Ensure that the cuneiformbrowser/data and cuneiformbrowser/log directory is writable. One of several options is to use the chmod command, e.g. $chmod -R 777 ./cuneiformbrowser/log/

3) Setup your login preferences under cuneiformbrowser/users/users.xml. (WARNING: the user access management is very basic and only provides a low level of protection)

4) To enable sign detection in the web front end, install the cuneiform-sign-detection-code on the same machine and run the webapp back-end using $python detector_app.py. For instruction how to run the webapp back-end, refer to the readme provided in ./lib/webapp/.

Usage

Please refer to the video and the help texts provided throughout the web front-end.

References

The two example images of clay tablets included in this repo are from the collection of the Vorderasiatisches Museum Berlin which kindly granted us permission to use them for our research purposes.