Source code for foxes.utils.geom2d.polygon

import numpy as np
from matplotlib.path import Path
from matplotlib.patches import PathPatch
from scipy.spatial.distance import cdist
import matplotlib.pyplot as plt

from .area_geometry import AreaGeometry


[docs] class ClosedPolygon(AreaGeometry): """ This class represents a closed 2D polygon. Attributes ---------- points: numpy.ndarray The polygon points poly: matplotlib.path.Path The closed polygon geometry :group: utils.geom2d """
[docs] def __init__(self, points): """ Constructor. Parameters ---------- points: numpy.ndarray The polygon points, shape: (n_points, 2) """ self.points = points if not np.all(points[0] == points[-1]): self.points = np.append(self.points, points[[0]], axis=0) self.poly = Path(self.points, closed=True) self._pathp = None
[docs] def p_min(self): """ Returns minimal (x,y) point. Returns ------- p_min: numpy.ndarray The minimal (x,y) point, shape = (2,) """ return np.min(self.points, axis=0)
[docs] def p_max(self): """ Returns maximal (x,y) point. Returns ------- p_min: numpy.ndarray The maximal (x,y) point, shape = (2,) """ return np.max(self.points, axis=0)
[docs] def points_distance(self, points, return_nearest=False): """ Calculates point distances wrt boundary. Parameters ---------- points: numpy.ndarray The probe points, shape (n_points, 2) return_nearest: bool Flag for return of the nearest point on bundary Returns ------- dist: numpy.ndarray The smallest distances to the boundary, shape: (n_points,) p_nearest: numpy.ndarray, optional The nearest points on the boundary, if return_nearest is True, shape: (n_points, 2) """ dists = cdist(points, self.points[:-1]) if return_nearest: mini = np.argmin(dists, axis=1) dists = np.take_along_axis(dists, mini[:, None], axis=1)[:, 0] minp = self.points[mini] del mini else: dists = np.min(dists, axis=1) for pi in range(len(self.points) - 1): pA = self.points[pi] pB = self.points[pi + 1] n = pB - pA d = np.linalg.norm(n) if d > 0: n /= d q = points - pA[None, :] x = np.einsum("pd,d->p", q, n) sel = (x > 0) & (x < d) if np.any(sel): x = x[sel] y2 = np.maximum(np.linalg.norm(q[sel], axis=1) ** 2 - x**2, 0.0) dsel = dists[sel] dists[sel] = np.minimum(dsel, np.sqrt(y2)) if return_nearest: mini = np.argwhere(np.sqrt(y2) < dsel) hminp = minp[sel] hminp[mini] = pA[None, :] + x[mini, None] * n[None, :] minp[sel] = hminp del mini, hminp del y2, dsel del x, sel if return_nearest: return dists, minp else: return dists
[docs] def points_inside(self, points): """ Tests if points are inside the geometry. Parameters ---------- points: numpy.ndarray The probe points, shape (n_points, 2) Returns ------- inside: numpy.ndarray True if point is inside, shape: (n_points,) """ return self.poly.contains_points(points)
[docs] def add_to_figure( self, ax, show_boundary=True, fill_mode=None, pars_boundary={}, pars_distance={} ): """ Add image to (x,y) figure. Parameters ---------- ax: matplotlib.pyplot.Axis The axis object show_boundary: bool Add the boundary line to the image fill_mode: str, optional Fill the area. Options: dist, dist_inside, dist_outside, inside_<color>, outside_<color> pars_boundary: dict Parameters for boundary plotting command pars_distance: dict Parameters for distance plotting command """ if show_boundary: pars = dict(facecolor="none", edgecolor="darkblue", linewidth=1) pars.update(pars_boundary) pathpatch = PathPatch(self.poly, **pars) ax.add_patch(pathpatch) super().add_to_figure( ax, show_boundary, fill_mode, pars_boundary, pars_distance )
if __name__ == "__main__": points = np.array([[1.0, 1.0], [1.3, 6], [5.8, 6.2], [6.5, 0.8]]) N = 500 fig, ax = plt.subplots() g = ClosedPolygon(points) g.add_to_figure(ax, fill_mode="dist_inside") plt.show() plt.close(fig) fig, ax = plt.subplots() g = ClosedPolygon(points) g.add_to_figure(ax, fill_mode="dist_outside") plt.show() plt.close(fig) fig, ax = plt.subplots() g = ClosedPolygon(points).inverse() g.add_to_figure(ax, fill_mode="dist_inside") plt.show() plt.close(fig) fig, ax = plt.subplots() g = ClosedPolygon(points).inverse() g.add_to_figure(ax, fill_mode="dist_outside") plt.show() plt.close(fig)