Python插值时间

Question

Novice in python so I hope I am asking correctly.

I have a huge set of data that I would like to interpolate to every one second and fill in the gaps with the appropriate latitude and longitude provided:

  Lat     Long    Time
-87.10   30.42    16:38:49
                      .
                      .
                      .
-87.09   30.40    16:39:22
                      .
                      .
                      .
-87.08   30.40    16:39:30

So I would like to generate a new latitude and longitude every second.

I have already plotted the corresponding latitude and longitude and would like to fill in the gaps with the interpolated data with points possibly.

Answer 1

If linear interpolation is good enough for you, you can use the numpy.interp function with the time array that you want to use for interpolation and the time and longitude/latitude data points that are read from the input file (the time must be increasing so you may need to sort the data).

To read the data for the file you can use the numpy.loadtxt function adding a converter to transform the time to an increasing number:

import numpy as np
from matplotlib.dates import strpdate2num

lon, lat, time = np.loadtxt('data.txt', skiprows=1,
        converters={2:strpdate2num('%H:%M:%S')}, unpack=True)

Then you can interpolate the longitude and latitude values using the interp function. The last argument to the linspace function gives the number of points in the time interpolated data.

interp_time = np.linspace(time[0], time[-1], 100)
interp_lon = np.interp(interp_time, time, lon)
interp_lat = np.interp(interp_time, time, lat)

For something more complicated that linear interpolation there are several facilities in scipy.interpolation .