Indoor landmarks play a crucial role in the process of indoor positioning and route planning for pedestrians or unmanned devices. Indoor structural landmarks, a type of indoor landmarks, can provide rich steering and semantic descriptions for indoor navigation services. However, most traditional indoor landmark extraction methods rely on indoor points of interest and indoor vector map data. These methods face the problem of difficult acquisition of indoor data and overlook the exploration of indoor structural landmarks. Therefore, this paper innovatively proposes a method for extracting indoor structural landmarks based on the commonly available indoor fire protection plan images. First, the HSV model is employed to eliminate noise from the original image, and vector data of indoor components is obtained using the constructed Canny operator. Subsequently, the visibility is calculated based on the grids of indoor space segmentation. Finally, the identification and extraction of indoor structural landmarks are achieved through grid visibility classification, directional clustering analysis, and spatial proximity verification. This approach opens up new ideas for indoor landmark extraction methods. The experimental results show that the method proposed in this paper can effectively extract indoor structural landmarks, the extraction accuracy of indoor structural landmarks reaches over 90%, verifying the feasibility of using indoor fire protection plan data for landmark extraction and expanding the data sources for indoor landmark extraction.

High-precision, multi-tier categorical soil classification faces critical bottlenecks, including the neglect of spatial context by conventional pixel-based models, the error cascade propagation phenomenon in multi-level classification networks, and the disruption of geophysical directional anisotropy by traditional geometric data augmentation. To address these challenges, in this study, we propose a multi-level soil classification model based on MTSC-ResNet-Trans, which organically couples residual convolutional blocks with a 3-layer Transformer encoder to model long-range spatial dependencies. The framework integrates a geospatial-safe data augmentation pipeline to preserve the topological fidelity of absolute geographic coordinates alongside four decoupled parallel multi-task classification heads to substantially suppress inter-level error propagation. Evaluated in the Southern Foothills of Qilian Mountains using 18 environmental covariates, the framework achieves an Overall Accuracy of 0.8931 at the Great Group level under conventional random splitting. Under a distance-stratified spatial evaluation—which isolates the contribution of spatial autocorrelation to accuracy estimates—the framework maintains robust performance, with MTSC-ResNet-Trans consistently outperforming pixel-based baselines (Random Forest) by approximately 3.7 percentage points even at spatial separation distances exceeding 400 m. This protocol transparently decomposes predictive accuracy into a component attributable to spatial proximity and a component reflecting reduced spatial proximity performance. Across the four taxonomic levels, accuracy decay is suppressed to 5.11%. Although spatial-block cross-validation indicates a lower regional extrapolation accuracy (OA = 0.7389 ± 0.0671), the decoupled parallel framework provides an effective and robust baseline for high-resolution regional digital soil mapping.