The digital recognition and preservation of historical architectural heritage has become a critical challenge in cultural inheritance and sustainable urban development. While deep learning methods show promise in architectural classification, existing models often struggle to achieve ideal results due to the complexity and uniqueness of historical buildings, particularly the limited data availability in remote areas. Focusing on the study of Chinese historical architecture, this research proposes an innovative architectural recognition framework that integrates the Swin Transformer backbone with a custom-designed Global Channel and Spatial Attention (GCSA) mechanism, thereby substantially enhancing the model’s capability to extract architectural details and comprehend global contextual information. Through extensive experiments on a constructed historical building dataset, our model achieves an outstanding performance of over 97.8% in key metrics including accuracy, precision, recall, and F1 score (harmonic mean of the precision and recall), surpassing traditional CNN (convolutional neural network) architectures and contemporary deep learning models. To gain deeper insights into the model’s decision-making process, we employed comprehensive interpretability methods including t-SNE (t-distributed Stochastic Neighbor Embedding), Grad-CAM (gradient-weighted class activation mapping), and multi-layer feature map analysis, revealing the model’s systematic feature extraction process from structural elements to material textures. This study offers substantial technical support for the digital modeling and recognition of architectural heritage in historical buildings, establishing a foundation for heritage damage assessment. It contributes to the formulation of precise restoration strategies and provides a scientific basis for governments and cultural heritage institutions to develop region-specific policies for conservation efforts.

The built-environment morphology of traditional settlements is a spatial outcome shaped by mountain–water patterns, architectural fabric, street–alley networks, and long-term human adaptation. To reveal the morphological differentiation and typological spectrum of Dong traditional settlements in southwestern Hunan, this study examined 33 settlements and developed a multi-scale identification framework integrating the terrain-constraint scale and the morphological-structure scale. Using unified boundary delineation, indicator extraction, standardization, hierarchical clustering, cross-tabulation analysis, and association tests, three terrain-constraint types were identified: the high-elevation dissected and water-distant type, the footslope transitional and water-adjacent type, and the low-valley and water-attached type. Three morphological-structure types were further identified: the compact-clustered type, the elongated-loose type, and the aggregation-enhanced type. The results show a significant and moderate-to-strong association between the two type systems. Based on the dominant T–M combinations, three integrated typological types were synthesized: the high-elevation dissected–compact-clustered type, the footslope transitional–elongated composite type, and the low-valley–aggregation-enhanced type. The findings not only provide a basis for typological identification and morphological comparison of traditional Dong settlements but also support the selection of representative settlement types for subsequent microclimate analyses. By distinguishing the characteristics and management priorities of different settlement types, the proposed framework provides a scientific basis for differentiated conservation planning, landscape management, and climate-adaptive renewal of traditional Dong settlements.