Chinese Hospital Landscape Images with Character Classification and Visual Comfort Scores

488 images from three Chinese hospitals were evaluated by 252 respondents to create a labeled dataset linking landscape character to human perception. A Data-Efficient Image Transformer model developed for this data achieved 96.34% classification accuracy and a 0.055 MAE for visual comfort prediction. The dataset, authored by Yue Li and last updated in May 2026, is shared under a CC-BY-4.0 license.

Use Cases

• Train image classification models for landscape character types based on the labeled hospital images.
• Develop regression models to predict continuous visual comfort scores from landscape imagery.
• Benchmark model performance against reported results for ResNet-50 and Vision Transformer.
• Analyze correlations between specific landscape characteristics and human perception outcomes in healthcare settings.
• Create decision support tools for systematic therapeutic landscape optimization in hospital planning.

Strengths

• Model performance metrics are reported, with 96.34% classification accuracy and a mean absolute error of 0.055 for visual comfort prediction.
• The dataset creation involved 252 respondents evaluating 30 representative images from a collection of 488, providing a basis for perception scoring.
• The associated research demonstrates the model outperformed ResNet-50 and a standard Vision Transformer in both accuracy and convergence speed.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset's 1.4 MB size suggests it may contain metadata or a sample rather than the full image collection.

Provenance

Source

figshare

Collection Method

488 images collected across three Chinese hospitals, with 30 representative images evaluated by 252 respondents to assign perception scores.

Time Range

null

Freshness

Last updated 2026-05-08 05:50:22; freshness should be verified.

Geography

Chinese hospital settings