Pywayne Geometric Hull Calculator

This module computes geometric hulls (convex and concave) for 2D point sets.

Quick Start

from pywayne.cv.geometric_hull_calculator import GeometricHullCalculator
import numpy as np

# Create calculator with random points
points = GeometricHullCalculator.generate_random_points(num_points=50)
calculator = GeometricHullCalculator(points, algorithm='concave-hull')

# Get results
print(f"MBR: {calculator.get_mbr()}")
print(f"Convex Hull: {calculator.get_convex_hull()}")
print(f"Concave Hull: {calculator.get_concave_hull()}")

# Visualize with matplotlib
calculator.visualize_matplotlib()

Initialization

# algorithm options: 'concave-hull' or 'alphashape'
# use_filtered_pts: Enable point filtering based on radius
calculator = GeometricHullCalculator(
    points=your_points,
    algorithm='alphashape',
    use_filtered_pts=True
)

Supported Algorithms

Hull Types

Properties

Visualization

OpenCV Visualization

calculator.visualize_opencv()

Displays: All input points, MBR, center, filter circle (if enabled), concave hull (green), convex hull (red).

Matplotlib Visualization

calculator.visualize_matplotlib()

Displays: All input points (red), MBR (blue), center, filter radius circle (if enabled), concave hull (orange), convex hull (purple).

Requirements

• numpy - Array operations
• cv2 (OpenCV) - For OpenCV visualization and MBR computation
• matplotlib - For matplotlib visualization
• scipy - For convex hull computation
• concave_hull - Concave hull algorithm
• alphashape - Alphashape algorithm
• shapely - Polygon operations for area calculation

Notes

• Point filtering uses radius = 30% of shorter MBR edge length
• MBR computed using OpenCV's minAreaRect
• Convex hull uses scipy's ConvexHull
• Supports both single Polygon and MultiPolygon from alphashape results