PyTorchFire: A GPU-accelerated wildfire simulator with Differentiable Cellular Automata
PyTorchFire is an open-access, PyTorch-based software utilizing GPU acceleration and a differentiable wildfire Cellular Automata model to achieve rapid, accurate, and realistic wildfire simulations, significantly outperforming traditional methods and enhancing real-time prediction and management capabilities.
Apr 1, 2025
Accurate Identification and Measurement of the Precipitate Area by Two-Stage Deep Neural Networks in Novel Chromium-Based Alloy
The study introduces DT-SegNet, a deep learning scheme combining YOLOv5 and SegFormer, to enhance object detection and segmentation in electron microscopy images, significantly improving the development of chromium-based superalloys by addressing limitations in traditional methods.
May 9, 2023