Siina

Description

Python library for Ground Penetrating Radar (GPR) data processing: IO, filters and visualization.

Tested with Python 3.6.

Installation

siina can be installed with pip

pip install siina

Latest Github version

Either clone the repo and install with setup.py

git clone https://github.com/ahartikainen/siina  
cd siina
python setup.py install

or with a pip

python -m pip install git+https://github.com/ahartikainen/siina

Underlying datastructures

Header information is saved as a dictionary: obj.header Measurement data is saved as a list of ndarrays: obj.data_list Main channel can be accessed with .data -method

Example usage

import siina

# create RadarFile object
meas = siina.Radar()

# read in the data
meas.read_file("./example_path/example_file.DZT")

# set the center frequency for GPR (in Hertz) if not done
if meas.header.get('frequency', None) is None:
    meas.header['frequency'] = 1e9 # 1 GHz

# print dimensions for the data
print("points in samples={}, samples={}, channels={}".format(meas.nrows, meas.ncols, meas.nchan)

# strip markers (important step with .DZT files)
meas.read_markers()

# center each sample (for each trace do func(trace[500:])
meas.func_dc(start=500)

# apply lowpass filter with cutoff= 6 * frequency
#     if cutoff is float -> cutoff = cutoff
#     if cutoff is str -> cutoff = float(cutoff) * frequency
meas.func_filter(cutoff='6')

import matplotlib.pyplot as plt

# plot mean function for the first channel
# all channels are found under obj.data_list
plt.plot(meas.data.mean(1))
plt.show()

# plot radargram with plt.imshow
# be careful with the profile size (meas.ncols < 5000)
plt.imshow(meas.data, aspect='auto')
plt.show()

Development

pip install -r requirements-test.txt

black siina
pylint siina
pydocstyle --convention=numpy siina