Deep learning framework tracks liquid sheet breakup dynamics

Researchers have developed a novel two-stage deep learning framework to analyze the complex dynamics of liquid sheet breakup into ligaments and droplets. The first stage utilizes a Faster R-CNN model with a ResNet-50 backbone and Feature Pyramid Network to detect and classify these fragments in high-speed imagery. This stage is enhanced by a morphology-preserving synthetic data generation strategy, achieving an F1 score of up to 0.872. The second stage employs a Transformer-augmented multilayer perceptron to track these objects across frames, accurately identifying continuations, one-to-many fragmentation events, and non-associations with 86.1% accuracy and perfect recall for fragmentation. AI

IMPACT This framework offers a new automated method for analyzing atomization processes, potentially improving spray analysis and fluid dynamics research.

RANK_REASON The cluster contains a research paper detailing a novel deep learning framework for scientific analysis. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• carbomer
• Faster R-CNN
• Feature Pyramid Networks for Object Detection
• ResNet-50
• transformer
• Vrushank Ahire

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →