NATHANCoverserver for UAntwerpen
Oktober 2023 - Mei 2024
For Anet, the IT service behind the University of Antwerp, I trained an AI-model that can distinguish good and bad book cover photos.
As a result, hundreds of cover photos are checked every day in a dozen libraries.
The catalog of the Anet network contains about 2,500,000 records, including those of the library of the city of Antwerp and the KMSKA.
After training the model, I integrated it, which was challenging due to the complexity of the codebase, especially since I had no experience with the framework Django used.
Additionally, I developed a REST API that allows communication between the front-end and the AI model. All predictions generated by the model are stored in the database.
Cover photos can be annotated with comments such as 'too much white space' or 'blurry', with the aim of training a better model in the future. I also created two screens for the developers where the predictions are visually displayed.
