Railway construction and industrialization played an important role in reshaping the urban morphology of inland resource-based cities in Republican-period China. Taking Taiyuan during 1912–1937 as a case study, this paper examines the spatial association between railway corridors and industrial locations and discusses how this relationship contributed to the reorganization of urban space. Historical maps, railway maps, local gazetteers, archival materials, and existing literature were integrated into a historical GIS database. Kernel density estimation, railway buffer analysis, mean-center analysis, and standard deviational ellipse analysis were used to reconstruct the staged evolution of industrial nodes and railway corridors. The results show that Taiyuan’s industrial space evolved from a scattered point-based pattern to a railway-oriented axial-belt layout and then to a polycentric industrial corridor. Industrial nodes gradually moved closer to railway lines, while the mean center of industrial distribution shifted from the traditional city toward railway-related industrial districts. The findings suggest that Taiyuan followed one possible railway–resource–industry pathway shaped by railway construction, resource endowments, provincial industrial policies, land conditions, and the inherited urban fabric. This study provides historical spatial evidence for identifying railway–industrial heritage corridors and understanding inland industrial urban form.

Chinese local gazetteer maps contain visual evidence of historical cities and architecture, but their hand-drawn, low-texture, and small-sample characteristics make automatic recognition difficult. This study develops a transfer-learning workflow to classify gazetteer-derived architectural images and examines its usefulness for digital heritage image organization, indexing, and retrieval. A dataset of 685 images covering nine classification labels was constructed from Chinese local gazetteer maps. Five representative models, ResNet-18, MobileNetV3-Small, EfficientNet-B0, DenseNet-121, and ViT-B/16, were trained and compared under the same transfer-learning setting. Their performance was evaluated using accuracy, macro-average F1 score, class-level metrics, confusion matrices, difficult-sample analysis, and Grad-CAM visualization. CNN-based models outperformed the Vision Transformer on this task. EfficientNet-B0 achieved the best results, with a test accuracy of 87.38% and a macro-average F1 score of 86.60%; after disabling class weights, the macro-average F1 score increased to 86.81%. Misclassifications mainly occurred in visually similar or context-dependent category pairs, including gate tower and drum tower, bell tower and drum tower, and buildings inside and outside the city. The results show that compact CNNs can support classification, indexing, and retrieval of hand-drawn architectural images in gazetteer archives, although the workflow should be regarded as an auxiliary tool rather than direct evidence for specific extant buildings.