This study addresses the challenge of rapid identification and assessment of localized damage to building envelopes under resource-constrained conditions—specifically, the absence of specialized inspection equipment—with a particular focus on the detrimental effects of such damage on thermal performance and energy efficiency. An efficient detection methodology tailored to small-scale maintenance scenarios is proposed, leveraging the YOLOv11 object detection architecture to develop an intelligent system capable of recognizing common envelope defects in contemporary residential buildings, including cracks, spalling, and sealant failure. The system prioritizes the detection of anomalies that may induce thermal bridging, reduced airtightness, or insulation degradation. Defects are classified according to severity and their potential impact on thermal behavior, enabling a graded, integrated repair strategy that holistically balances structural safety, thermal restoration, and façade aesthetics. By explicitly incorporating energy performance recovery as a core objective, the proposed approach not only enhances the automation of spatial data processing but also actively supports the green operation and low-carbon retrofitting of existing urban building stock. Characterized by low cost, high efficiency, and ease of deployment, this method offers a practical and scalable technical pathway for the intelligent diagnosis of thermal anomalies and the enhancement of building energy performance. It aligns with the principles of high-quality architectural development and sustainable building governance, while concretely advancing operational energy reduction in the built environment and contributing meaningfully to energy conservation goals.
Luo et al. (Sun,) studied this question.