Architectural Styles

Understanding regional architectural patterns is crucial for accurate building classification and feature extraction from IGN LiDAR HD data.

Overview

French architectural styles vary significantly by region, historical period, and urban vs rural context. This guide helps configure the processing pipeline for optimal results across different architectural contexts.

Regional Classifications

Ile-de-France (Paris Region)

Characteristics:

• Dense urban fabric with Haussmanian boulevards
• Uniform building heights (6-7 stories typical)
• Zinc roofing with characteristic slope angles
• Enclosed courtyards and consistent street alignment

Processing Configuration:

ile_de_france_config = {
    "building_detection": {
        "min_height": 8.0,  # Typical story height
        "max_height": 25.0,  # Haussmanian limit
        "roof_slope_range": [25, 45],  # Degrees
        "courtyard_detection": True
    },
    "architectural_features": {
        "mansard_roofs": True,
        "zinc_material": True,
        "balcony_detection": True
    }
}

Provence-Alpes-Côte d'Azur

Characteristics:

• Mediterranean flat roofs and terraces
• Stone and stucco construction
• Lower building heights
• Irregular urban patterns

Processing Configuration:

provence_config = {
    "building_detection": {
        "min_height": 3.0,
        "max_height": 15.0,
        "flat_roof_threshold": 5,  # Degrees max slope
        "terrace_detection": True
    },
    "materials": {
        "stone_detection": True,
        "tile_roofing": True,
        "stucco_surfaces": True
    }
}

Brittany (Bretagne)

Characteristics:

• Traditional granite construction
• Slate roofing with steep slopes
• Scattered rural settlements
• Maritime influence on building orientation

Processing Configuration:

brittany_config = {
    "building_detection": {
        "min_height": 4.0,
        "max_height": 12.0,
        "roof_slope_range": [35, 55],
        "wind_orientation": True
    },
    "materials": {
        "granite_detection": True,
        "slate_roofing": True,
        "chimney_prominence": True
    }
}

Historical Periods

Medieval Architecture (Pre-1500)

Features:

• Irregular building footprints
• Thick walls (>50cm typical)
• Small windows and openings
• Defensive characteristics

medieval_features = {
    "wall_thickness": {"min": 0.5, "typical": 0.8},
    "window_ratio": {"max": 0.15},  # Window to wall ratio
    "footprint_regularity": {"threshold": 0.3},
    "defensive_elements": True
}

Classical Architecture (1500-1800)

Features:

• Geometric regularity and symmetry
• Standardized proportions
• Formal gardens and courtyards
• Stone construction with carved details

classical_features = {
    "symmetry_detection": True,
    "proportion_analysis": True,
    "courtyard_geometry": "formal",
    "material_refinement": "high"
}

Industrial Architecture (1800-1950)

Features:

• Large span structures
• Brick and steel construction
• Repetitive bay systems
• Functional over decorative design

industrial_features = {
    "span_detection": {"min": 10.0, "max": 50.0},
    "bay_repetition": True,
    "material_types": ["brick", "steel", "concrete"],
    "chimney_detection": True
}

Contemporary Architecture (1950+)

Features:

• Diverse materials and forms
• Curtain wall systems
• Irregular geometries
• Mixed-use developments

contemporary_features = {
    "material_diversity": True,
    "geometric_complexity": "high",
    "curtain_wall_detection": True,
    "mixed_use_analysis": True
}

Building Typologies

Residential Buildings

Single-Family Homes

residential_single = {
    "footprint_area": {"min": 80, "max": 300},  # m²
    "height_range": {"min": 4, "max": 12},      # meters
    "roof_types": ["gable", "hip", "mansard"],
    "garden_detection": True
}

Multi-Family Housing

residential_multi = {
    "footprint_area": {"min": 200, "max": 2000},
    "height_range": {"min": 8, "max": 30},
    "balcony_detection": True,
    "courtyard_likelihood": 0.7
}

Commercial Buildings

Retail/Shops

commercial_retail = {
    "ground_floor_height": {"min": 3.5, "max": 6.0},
    "large_windows": True,
    "signage_detection": True,
    "street_frontage": True
}

Office Buildings

commercial_office = {
    "repetitive_floors": True,
    "curtain_walls": True,
    "regular_geometry": True,
    "parking_detection": True
}

Industrial Buildings

Manufacturing

industrial_manufacturing = {
    "large_spans": True,
    "high_ceilings": {"min": 8, "max": 25},
    "loading_docks": True,
    "minimal_windows": True
}

Configuration Examples

Urban Context Processing

from ign_lidar import Processor, ArchitecturalAnalyzer

# Initialize with urban architectural context
processor = Processor()
analyzer = ArchitecturalAnalyzer(
    region="ile_de_france",
    urban_context="dense_urban",
    historical_period="haussmanian"
)

# Process with architectural awareness
result = processor.process_tile(
    tile_path="paris_tile.las",
    architectural_context=analyzer,
    enable_style_classification=True
)

Rural Context Processing

analyzer = ArchitecturalAnalyzer(
    region="brittany",
    urban_context="rural",
    building_density="scattered"
)

result = processor.process_tile(
    tile_path="rural_tile.las",
    architectural_context=analyzer,
    preserve_vernacular_features=True
)

Style Classification Pipeline

Automatic Style Detection

def detect_architectural_style(building_features):
    """
    Automatically detect architectural style from extracted features
    """
    style_indicators = {
        "roof_slope": building_features["roof_slope_mean"],
        "wall_thickness": building_features["wall_thickness_mean"],
        "window_ratio": building_features["window_to_wall_ratio"],
        "regularity": building_features["geometric_regularity"],
        "height": building_features["building_height"]
    }

    # Style classification logic
    if style_indicators["roof_slope"] > 45 and style_indicators["wall_thickness"] > 0.6:
        return "traditional_rural"
    elif style_indicators["regularity"] > 0.8 and 15 < style_indicators["height"] < 25:
        return "haussmanian"
    elif style_indicators["window_ratio"] > 0.6:
        return "contemporary"
    else:
        return "mixed_urban"

Manual Style Configuration

# Define custom architectural style
custom_style = {
    "name": "art_deco_paris",
    "period": "1920-1940",
    "characteristics": {
        "stepped_facades": True,
        "ornamental_details": True,
        "vertical_emphasis": True,
        "mixed_materials": True
    },
    "detection_parameters": {
        "facade_complexity": {"min": 0.6},
        "height_variation": {"tolerance": 0.15},
        "material_transitions": {"detect": True}
    }
}

analyzer.add_custom_style(custom_style)

Performance Considerations

Memory Usage by Style Complexity

• Simple Rural: ~200MB per km²
• Urban Residential: ~500MB per km²
• Dense Urban: ~800MB per km²
• Mixed Architecture: ~1GB per km²

Processing Time Impact

• Style classification adds ~15% to processing time
• Detailed architectural analysis: +30-40%
• Historical pattern recognition: +20%

Validation and Accuracy

Ground Truth Comparison

• Manual architectural survey validation
• Historical cadastral data correlation
• Expert architectural assessment

Accuracy Metrics by Style

Style Type	Classification Accuracy	Feature Detection
Traditional Rural	89%	85%
Haussmanian Urban	94%	91%
Contemporary	78%	82%
Industrial	92%	88%

Related Documentation

• LOD3 Classification
• Building Feature Extraction
• Regional Processing Guide
• Historical Analysis Tools