# -*- coding: utf-8 -*-
# Author: Puyuan Du
from typing import Tuple, Optional
import numpy as np
try:
from pykdtree.kdtree import KDTree
except ImportError:
from scipy.spatial import KDTree
from cinrad.constants import deg2rad
from cinrad._typing import Number_T
class KDResampler(object):
def __init__(
self, data: np.ndarray, x: np.ndarray, y: np.ndarray, roi: Number_T = 0.02
):
x_ravel = x.ravel()
y_ravel = y.ravel()
self.tree = KDTree(np.dstack((x_ravel, y_ravel))[0])
self.data = data
self.roi = roi
def map_data(self, x_out: np.ndarray, y_out: np.ndarray) -> np.ndarray:
out_coords = np.dstack((x_out.ravel(), y_out.ravel()))[0]
_, indices = self.tree.query(out_coords, distance_upper_bound=self.roi)
invalid_mask = indices == self.tree.n
indices[invalid_mask] = 0
data = self.data.ravel()[indices]
data[invalid_mask] = np.nan
return data.reshape(x_out.shape)
[docs]def resample(
data: np.ndarray,
distance: np.ndarray,
azimuth: np.ndarray,
d_reso: Number_T,
a_reso: int,
) -> tuple:
r"""
Resample radar radial data which have different number of radials
in one scan into that of 360 radials
Args:
data (numpy.ndarray): Radar radial data.
distance (numpy.ndarray): Original distance.
azimuth (numpy.ndarray): Original azimuth.
Returns:
numpy.ndarray: Resampled radial data.
numpy.ndarray: Resampled distance.
numpy.ndarray: Resampled azimuth.
"""
# Target grid
Rrange = np.arange(d_reso, distance.max() + d_reso, d_reso)
Trange = np.linspace(0, 360, a_reso + 1) * deg2rad
dist, theta = np.meshgrid(Rrange, Trange)
# Original grid
d, t = np.meshgrid(distance, azimuth)
kds = KDResampler(data, d, t, 1)
r = kds.map_data(dist, theta)
return r, dist, theta
[docs]def grid_2d(
data: np.ndarray,
x: np.ndarray,
y: np.ndarray,
x_out: Optional[np.ndarray] = None,
y_out: Optional[np.ndarray] = None,
resolution: tuple = (1000, 1000),
) -> tuple:
r"""
Interpolate data in polar coordinates into geographic coordinates
Args:
data (numpy.ndarray): Original radial data.
x (numpy.ndarray): Original longitude data arranged in radials.
y (numpy.ndarray): Original latitude data arranged in radials.
resolution (tuple): The size of output.
Returns:
numpy.ndarray: Interpolated data in grid.
numpy.ndarray: Interpolated longitude in grid.
numpy.ndarray: Interpolated latitude in grid.
"""
r_x, r_y = resolution
if isinstance(x_out, type(None)):
x_out = np.linspace(x.min(), x.max(), r_x)
if isinstance(y_out, type(None)):
y_out = np.linspace(y.min(), y.max(), r_y)
t_x, t_y = np.meshgrid(x_out, y_out)
kds = KDResampler(data, x, y)
# TODO: Rewrite the logic for conversion between np.ma.masked and np.nan
result = kds.map_data(t_x, t_y)
return result, x_out, y_out