Navigation

Overview

Welcome to the Pattern Recognition Lab!

Researchers and students at Pattern Recognition Lab (LME) work on the development and implementation of algorithms to classify and analyze patterns like images or speech. The research is mostly interdisciplinary and is focussed on medical- and health engineering. The LME has close national and international collaborations with other universities, research institutes and industrial partners.

Research Areas

We are proud to announce that Bayerische Forschungsallianz will support a project between the McGill University and FAU Erlangen-Nuremberg on Dual Energy Imaging. Aim of the project is to enable easy access to dual energy image data and to further improve its clinical use. PIs are Reza Forghani at ...

It is a great pleasure to announce our new DFG Project on multimodal scanning of manuscripts. In the project, we will investigate joint scanning of books using X-ray, X-ray Darkfield, and Terahertz Imaging toward investigating whether their joint use enables further insights into inaccessible histo...

On January 24th, Prof. Martin Langner from the University of Göttingen will give a talk in our lab. Title: Mustererkennung als geisteswissenschaftliche Methode: Aktuelle Projekte des Instituts für Digital Humanities in GöttingenTime: 09:00 on January 24thRoom: H10 Abstract: Eine Reihe von Pro...

On December 19th, Prof. Colin Studholme from the University of Washington will give a talk in our lab. Title: Big Data From Small Brains: Computational Methods in Studies of Human Fetal Brain DevelopmentTime: 13:00 on December 19thRoom: H10 Abstract:This talk will cover some of the problem ar...

Camilo Vasquez, a PhD student of our lab was granted with the best paper award at the Iberoamerican conference on pattern recognition (CIARP 2019) that was held in Havana (Cuba) from 28.10.2019 to 31.10.2019. The award was given by the paper "Convolutional Neural Networks and a Transfer Learning Strategy to Classify Parkinson's Disease from Speech in Three Different Languages". The transfer learning scheme aims to improve the accuracy of the models when the weights of a CNN are initialized with utterances from a different language than the used for the test set.