PulseWaves

Python3 module for reading uncompressed lidar full waveform PulseWaves files (PLS).

Install

git clone https://github.com/Rheinwalt/pulsewaves.git
cp pulsewaves/pulsewaves.py ~/your/python/path

Usage

from pulsewaves import PulseWaves

f = PulseWaves('sample.pls')
f.export()

This will export all Pulse Records to an HDF5 file with the same file name (e.g., sample.hdf). All Wave segments are stored in that file including the coordinates for all amplitude samples of outgoing and returning waveforms. Data is structured into three HDF group datasets:

import h5py

g = h5py.File('sample.hdf', 'r')
g.keys()
<KeysViewHDF5 ['Amplitude', 'Index', 'XYZ']>

The Amplitude dataset has two columns, the first for the actual amplitude values (integers), and the second for an integer specifying the segment number (0 for the outgoing wave, 1 for the first returning wave segment, 2 for the second returning wave segment, etc.). Each row corresponds to a sample. The same holds true for the dataset XYZ, except that this dataset has three columns; one for each coordinate (double). Which row corresponds to which Pulse Record is encoded by the Index dataset (integer). Hence, the following would plot the ten returning waveforms (all segments) from record 780 to 790 with their amplitudes on the y-axis and the elevation on the x-axis:

import h5py
from matplotlib import pyplot as pl

g = h5py.File('sample.hdf', 'r')
amp = g['Amplitude']
xyz = g['XYZ']
idx = g['Index']

fg = pl.figure(1, (8, 6))
ax = fg.add_subplot(111)

for i in range(780, 790):
    a = amp[idx[i]:idx[i+1], :]
    z = xyz[idx[i]:idx[i+1], 2]
    z = z[a[:,1] > 0]
    a = a[a[:,1] > 0, 0]
    ax.plot(z, a, label = 'pulse %i' % i)

ax.set_xlabel('Elevation [m]')
ax.set_ylabel('Amplitude')

pl.legend()
pl.tight_layout()
pl.show()

Showing the same waveforms in a 3D plot:

import h5py
from mpl_toolkits.mplot3d import Axes3D
import matplotlib.pyplot as pl

g = h5py.File('sample.hdf', 'r')
amp = g['Amplitude']
xyz = g['XYZ']
idx = g['Index']

a = amp[idx[780]:idx[790]]
p = xyz[idx[780]:idx[790]]
p = p[a[:,1] > 0, :]
a = a[a[:,1] > 0, 0]
x, y, z = p.T

fg = pl.figure(1, (8, 6))
ax = fg.add_subplot(111, projection = '3d')
ax.scatter(x, y, z, c = np.log10(a))
ax.set_xlabel('UTM X [m]')
ax.set_ylabel('UTM Y [m]')
ax.set_zlabel('Elevation [m]')

pl.tight_layout()
pl.show()

Alternatively, one can load individual Pulse Records and Waves. This is done as in the original Python2 version of this module:

from pulsewaves import PulseWaves

f = PulseWaves('sample.pls')

# get the first pulse record
r = f.get_pulse(0)

# get the corresponding wave segments
w = f.get_waves(0)
print(w.segments.keys())