Enabling Real-Time, Cost-Efficient, and Lightweight High-Speed Crack Segmentation using Self-Supervised Attention Mechanism

Abstract This study proposes a lightweight and high-performance crack segmentation framework based on a MobileNet U-Net architecture, enhanced through self-supervised learning (SSL) and a multi-head attention mechanism. The encoder is pretrained using a self-supervised image inpainting task on grays...

Descripción completa

Detalles Bibliográficos

Autores principales:	Wei Wei, Muhammad Saqib, Haleema Ehsan, Abdullah I Al-Mansour, Tong Xinzhe, Ali Alhawiti, Shi Qiu, Qasim Zaheer
Formato:	Artigo
Lenguaje:	Inglês
Publicado:	Springer 2026-04-01
Colección:	International Journal of Computational Intelligence Systems
Materias:	Crack segmentation Crack detection Pavement cracks Efficient model Cost effective
Acceso en línea:	https://doi.org/10.1007/s44196-026-01306-y
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!