Lightweight UMamba3D improves seismic fault detection in noisy 3D data
A lightweight 3D network combined competitive synthetic-data segmentation accuracy with stronger fault recovery under severe noise and more coherent fault ...
A lightweight 3D network combined competitive synthetic-data segmentation accuracy with stronger fault recovery under severe noise and more coherent fault ...
Researchers have developed an attention-enhanced residual network that converts structural vibration data into Markov transition field images and detects local ...
Deep learning models trained on noisy acoustic signals can classify and localize natural gas pipeline leaks with over 99% accuracy, ...
A new FCN-GRU-Attention deep learning model with a confidence correction mechanism identifies overlapping composite power quality disturbances with over 99 ...
A hybrid systematic review and experimental benchmark of five deep learning architectures reveals which neural networks can realistically detect and ...
A new study systematically benchmarks five CNN compression techniques across GPUs, CPUs, and noisy data, finding that the best choice ...
A controlled study of synthetic identity fraud detection shows that rule-anchored heterogeneous graph models win through complete identity relation structure ...
Researchers in China have developed a noise-resistant deep neural network that monitors the oil temperature of marine thrust bearings from ...
Researchers in the Czech Republic have unveiled an unambiguous variational quantum classifier that reaches 90 percent accuracy on a breast ...
Researchers have developed an aspect-centric, noise-resilient fusion framework that improves fine-grained sentiment analysis of text-image pairs by aligning aspects across ...
© 2025 Scienmag - Science Magazine
© 2025 Scienmag - Science Magazine