Understanding and analyzing biopores and plant roots in Soil Cores using Semantic Segmentation of CT Images

Decorative Image for HI Collaboration for biopores and plant roots in soil cores using semantic segmentation of CT images

To understand how processes in ecosystems work and how they are connected the analysis of soil systems is essential. Since traditional computer vision methods for analysing soil cores reach their limits the next step is to integrate deep learning methods. Therefore a sufficient amount of labeled ground truth data is needed. Since labeling this large data is a tedious and time consuming procedure an efficient annotation strategy is developed and following a semantic segmentation model is trained. Thus, it is sufficient to label only 0.5% of the data to obtain high quality predictions. The trained model can then be used to label the missing 99.5% of the training data or to predict completely new images. The obtained information can be used to analyse the soil structure or the flow of water as well as to measure the biological activities.

Other Collaborations


Detecting Type-2 Diabetes in histopathological images for a better understanding of biological processes behind the disease

Type-2 diabetes is a chronic disease affecting about 500 million people worldwide. Despite extensive research over the last decades, exact biological processes leading to a deteriorating insulin production are not yet fully understood. By building models that are able to classify whether a patient has type-2 diabetes or not from whole slide images of the […]

Understanding and analyzing plant roots using semantic segmentation of MRI images

The optimization of plants has long focused on the above-ground parts. Recently, new efforts are being made to exploit the potential below ground. To this end, our partners at the FZJ have developed an imaging system which enables imaging the root system throughout the growth of the plants using MRI. Besides qualitative analysis, a precise […]


In the context of the Unisef project funded by Helmholtz AI, we implemented the prototype of a webservice to allow training and application of deep learning networks for segmentation running on the HPC infrastructure of DESY. The main idea here is to make DL based segmentation accessible and usable by non-experts, as well as complementing […]