A Multi-Scale Approach to Early Fire Detection in Smart Homes

Electronics, Год журнала: 2024, Номер 13(22), С. 4354 - 4354

Опубликована: Ноя. 6, 2024

In recent years, advancements in smart home technologies have underscored the need for development of early fire and smoke detection systems to enhance safety security. Traditional methods relying on thermal or sensors exhibit limitations terms response time environmental adaptability. To address these issues, this paper introduces multi-scale information transformer–DETR (MITI-DETR) model, which incorporates feature extraction transformer-based attention mechanisms, tailored specifically homes. MITI-DETR achieves a precision 99.00%, recall 99.50%, mean average (mAP) 99.00% custom dataset designed reflect diverse lighting spatial conditions Extensive experiments demonstrate that outperforms state-of-the-art models metrics, especially under challenging conditions. This work provides robust solution homes, combining high accuracy with real-time deployment feasibility.

Язык: Английский

Evaluation of synthetic data impact on fire segmentation models performance

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Май 14, 2025

Язык: Английский