CN2VF-Net: A Hybrid Convolutional Neural Network and Vision Transformer Framework for Multi-Scale Fire Detection in Complex Environments

Fire detection remains a challenging task due to varying fire scales, occlusions, and complex environmental conditions. This paper proposes the CN2VF-Net model, a novel hybrid architecture that combines vision Transformers (ViTs) and convolutional neural networks (CNNs), effectively addressing these...

Cur síos iomlán

Sábháilte in:

Sonraí bibleagrafaíochta
Príomhchruthaitheoirí:	Naveed Ahmad, Mariam Akbar, Eman H. Alkhammash, Mona M. Jamjoom
Formáid:	Artigo
Teanga:	Inglês
Foilsithe / Cruthaithe:	MDPI AG 2025-05-01
Sraith:	Fire
Ábhair:	convolutional neural networks (CNNs) vision Transformers (ViTs) D-Fire Grad-CAM multi-scale attention mechanism
Rochtain ar líne:	https://www.mdpi.com/2571-6255/8/6/211
Clibeanna:	Cuir clib leis Níl clibeanna ann, Bí ar an gcéad duine le clib a chur leis an taifead seo!

CN2VF-Net: A Hybrid Convolutional Neural Network and Vision Transformer Framework for Multi-Scale Fire Detection in Complex Environments

Míreanna comhchosúla