Analytische Chemie
Filtern
Erscheinungsjahr
- 2021 (4) (entfernen)
Dokumenttyp
- Zeitschriftenartikel (1)
- Vortrag (1)
- Sonstiges (1)
- Forschungsdatensatz (1)
Sprache
- Englisch (4)
Schlagworte
Organisationseinheit der BAM
- 1.2 Biophotonik (4) (entfernen)
Paper des Monats
- ja (1)
Eingeladener Vortrag
- nein (1)
We present a workflow for obtaining fully trained artificial neural networks that can perform automatic particle segmentations of agglomerated, non-spherical nanoparticles from scanning electron microscopy images “from scratch”, without the need for large training data sets of manually annotated images. The whole process only requires about 15 minutes of hands-on time by a user and can typically be finished within less than 12 hours when training on a single graphics card (GPU). After training, SEM image analysis can be carried out by the artificial neural network within seconds. This is achieved by using unsupervised learning for most of the training dataset generation, making heavy use of generative adversarial networks and especially unpaired image-to-image translation via cycle-consistent adversarial networks. We compare the segmentation masks obtained with our suggested workflow qualitatively and quantitatively to state-of-the-art methods using various metrics. Finally, we used the segmentation masks for automatically extracting particle size distributions from the SEM images of TiO2 particles, which were in excellent agreement with particle size distributions obtained manually but could be obtained in a fraction of the time.
This dataset accompanies the following publication, first published in Scientific Reports (www.nature.com/articles/s41598-021-84287-6):
B. Ruehle, J. Krumrey, V.-D. Hodoroaba, Scientific Reports, Workflow towards Automated Segmentation of Agglomerated, Non-Spherical Particles from Electron Microscopy Images using Artificial Neural Networks, DOI: 10.1038/s41598-021-84287-6
It contains electron microscopy micrographs of TiO2 particles, the corresponding segmentation masks, and their classifications into different categories depending on their visibility/occlusion. Please refer to the publication and its supporting information for more details on the acquisition and contents of the dataset, as well as the GitHub repository at https://github.com/BAMresearch/automatic-sem-image-segmentation
The publicly available document encapsulates the first version of the Catalogue of Services of the future EC4Safenano Centre (CoS 2019).
The CoS 2019 is structured in 12 Service Categories and 27 Service Topics, for each of the 12 categories considered. This architecture configures a 12 x 27 matrix that allows ordering the potential EC4Safenano offer in 324 types of services/groups of services.
Each type of service/group of services is described, in a simple and friendly way, by means of a specific service sheet: the EC4Safenano - Service Data Sheet (EC4-SDS). These EC4-SDSs allow structuring and summarizing the information of each service, providing the customer with a concise view of characteristics of the service and also the contact details with the service provider.
The CoS 2019 deploys a map of services consisting of a set of 100 EC4-SDSs, covering 7 of the 12 Service Categories and 17 of the 27 Service Topics.
The harmonization of services is visualized as a future necessary step in EC4Safenano, in order to strengthen the offer and provide added value to customers with a growing offer of harmonized services in future versions of the CoS.
The information contained in this document is structured in 3 main sections, as follows:
• Catalogue structure. This section describes in short the main characteristics of the CoS 2019.
• Catalogue content. This section represents the core part of the document and encapsulates the set of 100 SDSs displaying the offer proposed by the CoS 2019.
• Online Catalogue. This section describes the resources implemented by EC4Safenano to facilitate the on-line consultation of the CoS 2019 by customers and other interested parties.
We present a workflow for obtaining fully trained artificial neural networks that can perform automatic particle segmentations of agglomerated, non-spherical nanoparticles from electron microscopy images “from scratch”, without the need for large training data sets of manually annotated images. This is achieved by using unsupervised learning for most of the training dataset generation, making heavy use of generative adversarial networks and especially unpaired image-to-image translation via cycle-consistent adversarial networks. The whole process only requires about 15 minutes of hands-on time by a user and can typically be finished within less than 12 hours when training on a single graphics card (GPU). After training, SEM image analysis can be carried out by the artificial neural network within seconds, and the segmented images can be used for automatically extracting and calculating various other particle size and shape descriptors.