lire cette page en :

Learning in bioimage analysis using ilastik

Heidelberg Collaboratory for Image Processing, University of Heidelberg, Germany
le 28/03/2014 à 14:15


Bioimaging data have become so big that automatic analysis is the only option. However, traditional hand-crafted algorithms are very expensive and do not adapt to variations in the experimental setup. To solve this problem, our group's major open-source project "ilastik" ( offers generic image analysis methods (pixel and object classification, interactive segmentation, and tracking) in up to five dimensions (space, time and spectral) which can be adapted to new experiments by means of modern machine learning. ilastik's intuitive user interface and immediate feedback on all interactions enable biologists to train these methods themselves, without consulting an image analyis expert. The talk will give an overview over ilastik's basic principles and an online demonstration of the software's capabilities.