English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Crystalline defects, such as line-like dislocations, play an important rolefor the performance and reliability of many metallic devices. Their interactionand evolution still poses a multitude of open questions to materials scienceand materials physics. In-situ TEM experiments can provide important insightsinto how dislocations behave and move. During such experiments, the dislocationmicrostructure is captured in form of videos. The analysis of individual videoframes can provide useful insights but is limited by the capabilities ofautomated identification, digitization, and quantitative extraction of thedislocations as curved objects. The vast amount of data also makes manualannotation very time consuming, thereby limiting the use of DeepLearning-based, automated image analysis and segmentation of the dislocationmicrostructure. In this work, a parametric model for generating synthetictraining data for segmentation of dislocations is developed. Even though domainscientists might dismiss synthetic training images sometimes as too artificial,our findings show that they can result in superior performance, particularlyregarding the generalizing of the Deep Learning models with respect todifferent microstructures and imaging conditions. Additionally, we propose anenhanced deep learning method optimized for segmenting overlapping orintersecting dislocation lines. Upon testing this framework on four distinctreal datasets, we find that our synthetic training data are able to yieldhigh-quality results also on real images-even more so if fine-tune on a fewreal images was done.

Deep Learning of Crystalline Defects from TEM images: A Solution for the Problem of "Never Enough Training Data"