Diagnosing Shortcut Learning in CNN-Based Photovoltaic Fault Recognition from RGB Images: A Multi-Method Explainability Audit

Convolutional neural networks (CNNs) can achieve high accuracy in photovoltaic (PV) fault recognition from RGB imagery, yet their decisions may rely on shortcut cues induced by heterogeneous backgrounds, viewpoints, and class imbalance. This work presents a multi-method explainability audit on the K...

ver descrição completa

Detalhes bibliográficos

Autor principal:	Bogdan Marian Diaconu
Formato:	Artigo
Idioma:	Inglês
Publicado em:	MDPI AG 2026-03-01
Colecção:	AI
Assuntos:	explainable AI shortcut learning transfer learning photovoltaic panel fault detection
Acesso em linha:	https://www.mdpi.com/2673-2688/7/3/94
Tags:	Adicionar Tag Sem tags, seja o primeiro a adicionar uma tag!