def addCoords(self, gpsfile):
self.info("Adding GPS coordinates from file: %s ..." % gpsfile)
gps = GPS(projection='UTM', utmzone='33')
if os.path.splitext(gpsfile)[1] == '.mat':
gps.readMAT(gpsfile)
else:
gps.readGPS(gpsfile)
gps.interpolateProj(self.traces * self.nstacks)
if self.nstacks > 1:
gps.stackCoords(self.traces)
self.coords = gps
self.hascoords = True
self.done()
def addCoords(self, projection="PS71", utmzone=None, gpsfile=None):
'''Convenient function:
Reads GPS coordinates from file and
interpolates x,y,z coordinates in the specified projection to each trace.
'''
if gpsfile == None:
gpsfile = self.filename + '.cor'
cor = GPS(projection, utmzone=utmzone)
cor.readCOR(gpsfile, correction='HUBRA')
cor.interpolateProj(self.traces * self.nstacks)
if self.nstacks > 1:
cor.stackCoords(self.traces)
self.coords = cor
# if projection == "PS71":
# proj_describe = 'Polar Stereographic Coordinates (lat_ts=-71)'
# elif projection == "UTM":
# proj_describe = "Universal Trans-Mercator, Zone %s" % utmzone
# else:
# proj_describe = "Lat/Long Degrees"
self.hascoords = True
class GPR:
def __init__(self):
'''
Initialize GPR class and set global variables.
'''
self.eps0 = 8.85418782e-12 #: permittivity of free space [m-3 kg-1 s4 A2]
self.mu0 = 1.2566371e-6 #: permeability of free space [H/m]
self.c0 = 0.299792458 #: speed of light in m/ns
self.cmed = 0.23 #: speed in bulk medium in m/ns
self.data = None
self.samples = None
self.deltaT = 0
self.ylabel = "Time"
self.nstacks = 1
self.hascoords = False
self.filename = ""
self.cmin = 0.0
self.cmax = 1.0
self.procext = '_org' #: added to the filename when saved to avoid file deletion
self.procstring = ""
self.proclog = [] #: lists the processing steps in detail
self.parameters()
def parameters(self):
pass
def info(self, text):
'''
Writes info text to standard out.
@todo: Set up for logging and H5.history.
'''
sys.stdout.write(text)
self.procstring = text[:-4]
def done(self):
sys.stdout.write(' Done!')
sys.stdout.flush()
print ""
self.proclog.append(self.procstring)
def log(self):
HomeDir = os.path.expanduser('~')
log_file = os.path.join(HomeDir, "processing.log")
logging.basicConfig(filename=log_file, level=logging.INFO)
def writeBin(self, filename):
self.info('Writing data to file %s ...' % filename)
self.data.tofile(filename) # data will be written in 'double' format
self.done()
def writeAscii(self, filename):
from scipy.io import write_array
self.info('Writing data to file: ' + filename + '.asc')
write_array(filename + ".asc", np.transpose(self.data[1:-1, :]), separator=' ', linesep='\n')
self.done()
def writeH5(self, filename):
self.info('Writing data to file: %s.h5 ...' % filename)
if np.dtype(self.data[0, 0]) != 'float32':
self.data = self.data.astype('float32')
h5file = tables.openFile(filename + '.h5', mode='w', title='GPR data file')
profile = h5file.createGroup(h5file.root, 'profile', 'GPR profile ' + filename)
h5file.createArray(profile, 'traces', self.data, 'GPR traces')
if self.hascoords:
coordinates = h5file.createGroup(h5file.root, 'coordinates', 'Polar Stereographic coordinates (lat_ts=-71)')
h5file.createArray(coordinates, 'Y', self.coords.y, 'Trace y location')
h5file.createArray(coordinates, 'X', self.coords.x, 'Trace x location')
h5file.createArray(coordinates, 'Elevation', self.coords.elev, 'Trace elevation')
h5file.createArray(coordinates, 'Distance', self.coords.dist,
'Distance from previous trace') # NEEDS TO BE RECALCULATED AFTER SATACK
h5file.root.profile._v_attrs.polarity = self.PP
h5file.flush()
print h5file
h5file.close()
self.done()
def saveH5(self, filename):
from processgpr.core.h5file import H5FILE
h5file = H5FILE()
h5file.new(filename, self.data.shape)
h5file.set_metadata(self)
h5file.append_earray_data(self.data)
if self.hascoords:
h5file.append_earray_gps(self.coords)
h5file.close()
def readH5(self, filename, tracemin=0, tracemax=-1,
samplemin=0, samplemax=-1, roffset=0.0):
'''
Reads GPR data files in HDF5 format that have been created by L{writeH5}.
@param filename: The name of the input file.
@param tracemin: The first trace to used.
@param tracemax: The last trace to be used.
@param samplemin: The first sample to be used.
@param samplemax: The last sample to be used.
@param roffset: The offset in meters between I{samplemin} and the\
antenna.
'''
self.info('Reading data from file: %s ...' % filename)
''' Get data '''
self.filename = os.path.splitext(filename)[0]
h5file = tables.openFile(filename, mode='r')
try:
self.totalsamples, self.totaltraces = h5file.root.data.traces.shape
except AttributeError:
self.totalsamples, self.totaltraces = h5file.root.profile.traces.shape
print "\n Total", self.totalsamples, self.totaltraces
if tracemax == -1:
tracemax = self.totaltraces
if samplemax == -1:
samplemax = self.totalsamples
self.samplendx = np.arange(samplemin, samplemax)
# print self.samplendx.shape
self.tracendx = np.arange(tracemin, tracemax + 1)
# print self.tracendx
try:
self.data = h5file.root.data.traces[samplemin:samplemax,
tracemin:tracemax]
self.samples, self.traces = np.shape(self.data)
except AttributeError:
self.data = h5file.root.profile.traces[samplemin:samplemax,
tracemin:tracemax] # old format
self.samples, self.traces = np.shape(self.data)
if h5file.root.__contains__('coordinates'):
self.hascoords = True
self.coords = GPS()
self.coords.y = h5file.root.coordinates.Y[tracemin:tracemax]
self.coords.x = h5file.root.coordinates.X[tracemin:tracemax]
self.coords.elev = h5file.root.coordinates.Elevation[tracemin:tracemax]
self.coords._getDistance()
self.d = np.linspace(0, np.add.reduce(self.coords.dist), self.traces)
else:
print "WARNING: No coordinates found"
try:
self.gatefreq = h5file.root.data.traces.attrs.gatefreq
except AttributeError:
print "WARNING: Missing attribute - Gating frequency not defined"
try:
self.nstacks = h5file.root.data.traces.attrs.nstacks
except AttributeError:
print "WARNING: Missing attribute - Number of stacks not defined"
try:
self.deltaT = h5file.root.data.traces.attrs.deltaT
except AttributeError:
print "WARNING: Missing attribute - Time increment not defined"
try:
self.cmed = h5file.root.data.traces.attrs.cmed
except AttributeError:
print "WARNING: Missing attribute - Medium velocity not defined"
h5file.close()
''' Set parameters '''
if samplemax == -1:
samplemax = self.samples
# self.t = self.deltaT*self.samplendx*1e9
self.t = np.arange(0.0, self.deltaT * self.samples, self.deltaT) * 1e9
self.r = (self.t * self.cmed / 2.0) + roffset
self.done()
def applyWindow(self, window="hanning", ww=0, cf=0):
'''
Apply window function to frequency domain data
cf: the frequency the window is centered over [Hz]
ww: the window width [Hz], if ww equals 0 the window covers the full range
'''
self.info("Applying %s window ..." % window)
if window == "hanning":
if ww == 0:
w = np.hanning(self.numfreq)
else:
pos = int((cf - self.lowF) / self.deltaF)
halfwidth = int(ww / (2.0 * self.deltaF))
w = np.zeros(self.numfreq)
w[pos - halfwidth:pos + halfwidth] = np.hanning(2 * halfwidth)
elif window == "hamming":
if ww == 0:
w = np.hamming(self.numfreq)
else:
pos = int((cf - self.lowF) / self.deltaF)
halfwidth = int(ww / (2.0 * self.deltaF))
w = np.zeros(self.numfreq)
w[pos - halfwidth:pos + halfwidth] = np.hamming(2 * halfwidth)
elif window == "blackman":
if ww == 0:
w = np.blackman(self.numfreq)
else:
pos = int((cf - self.lowF) / self.deltaF)
halfwidth = int(ww / (2.0 * self.deltaF))
w = np.zeros(self.numfreq)
w[pos - halfwidth:pos + halfwidth] = np.blackman(2 * halfwidth)
self.data = self.data * w
self.done()
def zeropad(self):
'''
Zero-pad to move samples to the correct position in the FD
'''
self.info("Zero-padding ...")
tmparray = np.zeros((self.traces, self.Nfft), dtype=np.float32)
for trace in xrange(self.traces):
tmparray[trace, self.nstart:self.nstart + self.numfreq] = self.data[trace, :]
self.data = tmparray
self.numfreqpadded = self.data.shape[1]
del (tmparray)
self.done()
def applyFFT(self, wdir=os.getcwd(), envelope=True):
'''
FFT transfers the FD data to TD data
'''
self.info("Applying FFT ...")
workfile = os.path.join(wdir, "tmp" + os.path.basename(self.filename) + ".dat")
fid = open(workfile, "wb")
if envelope:
for trace in xrange(self.traces):
np.abs(fft.fft(self.data[trace])).tofile(fid) # envelope
else:
for trace in xrange(self.traces):
np.real(fft.fft(self.data[trace])).tofile(fid) # incl. phase
fid.close()
fid = open(workfile, "rb")
try: # Zero-padding applied
self.data = np.zeros((self.numfreqpadded, self.traces), dtype=np.float32)
for i in range(self.traces):
fid.seek(self.numfreqpadded * i * 8)
self.data[:, i] = np.fromfile(fid, "<f8", self.numfreqpadded)
except AttributeError: # Zero-padding not applied
self.data = np.zeros((self.numfreq, self.traces), dtype=np.float32)
for i in range(self.traces):
fid.seek(self.numfreq * i * 8)
self.data[:, i] = np.fromfile(fid, "<f8", self.numfreq)
fid.close()
self.samples = self.data.shape[0]
self.traces = self.data.shape[1]
self.t = np.arange(0, self.deltaT * self.samples, self.deltaT) * 1e9 #: time axis in ns
self.r = self.t * self.cmed / 2.0 #: range axis in m
self.done()
def stackTraces(self, nstacks=2):
'''
Stack "nstacks" traces in time domain.
'''
self.info('Stacking profile over %s traces ...' % nstacks)
newtraces = int(self.traces / nstacks)
S = np.zeros((self.samples, newtraces), np.dtype(self.data[0, 0]))
n = 0
for i in range(newtraces):
S[:, i] = np.mean(self.data[:, n:n + nstacks], axis=1)
if n + nstacks <= self.traces:
n = n + nstacks
self.data = S
self.traces = newtraces
try:
tracendxmin = int(self.tracendx.min() / nstacks)
tracendxmax = int(self.tracendx.min() / nstacks) + newtraces
self.tracendx = np.arange(tracendxmin, tracendxmax + 1)
except AttributeError:
self.tracendx = np.arange(self.traces)
if self.hascoords:
self.coords.stackCoords(self.traces)
self.procext += '_stk'
self.nstacks = nstacks
self.done()
def prestackTraces(self, nstacks=2):
'''
Pre-stack "nstacks" traces in frequency domain.
'''
self.info('Pre-stacking profile over %s traces ...' % nstacks)
newtraces = int(self.traces / nstacks)
S = np.zeros((newtraces, self.numfreq), np.dtype(self.data[0, 0]))
n = 0
for i in range(newtraces):
S[i, :] = np.mean(self.data[n:n + nstacks, :], axis=0)
if n + nstacks <= self.traces:
n = n + nstacks
self.data = S
self.traces = newtraces
# if self.hascoords:
# self.coords.stackCoords(self.traces)
self.procext += '_prestk'
self.nstacks = nstacks
self.done()
def clip(self):
'''
Clips unwanted neg. values.
'''
self.data = np.clip(self.data, 0.0, self.data.max())
def reverse_profile(self):
''' Flips the profile left to right '''
self.info('Flipping profile...')
self.data = np.fliplr(self.data)
self.done()
def power_phase(self, data):
mask = data < 0.0
tmpdata = data ** 2
tmpdata[mask] = -tmpdata[mask]
return tmpdata
def backgroundRemoval(self):
self.info('Applying background removal ...')
med = np.mean(self.data, axis=1)
for i in range(self.traces):
self.data[:, i] = self.data[:, i] - med
self.procext = '_bkg'
self.done()
def subtractMean(self, Lhalf=50):
self.info('Subtracting mean over %4.0f traces...' % (Lhalf * 2))
reftrace = self.data[:, 70]
tmp_data = np.zeros(self.data.shape)
med_first = np.mean(self.data[:, 0:2 * Lhalf], axis=1)
med_last = np.mean(self.data[:, self.traces - (2 * Lhalf):self.traces], axis=1)
med = []
for i in range(self.traces):
if i < Lhalf:
tmp_data[:, i] = self.data[:, i] - med_first
elif i > self.traces - Lhalf:
tmp_data[:, i] = self.data[:, i] - med_last
elif Lhalf <= i <= (self.traces - Lhalf):
print i - Lhalf, i, i + Lhalf
med.append(np.mean(self.data[:, (i - Lhalf):(i + Lhalf + 1)], axis=1))
tmp_data[:, i] = self.data[:, i] - med[-1]
self.data = tmp_data
self.procext = '_sbt'
import pylab
pylab.plot(reftrace)
# pylab.plot(med_first)
# pylab.plot(med_last)
# pylab.plot(med[50])
pylab.plot(self.data[:, 70])
pylab.plot(med[70])
pylab.show()
self.done()
def linearGain(self, maxgain=1.2):
self.info('Applying linear gain (Max. Gain = %2.2f) ...' % maxgain)
gain = np.linspace(1.0, maxgain, self.samples)
for i in range(self.traces):
self.data[:, i] *= gain
self.procext = '_lg'
self.done()
def compensateSpread(self, a=2.0, power=True):
self.info('Compensating for spreading losses (simple) ...')
r = self.r
r.shape = (r.shape[0], 1)
if power:
self.data = (self.data ** 2) * r ** a
else:
self.data = self.data * np.sqrt(r ** a)
self.done()
def compensateVol(self, h0=0.65, ba=61.0, power=True, view=False):
''' From Davis and Poznyak, IEEE Trans., 1993 '''
self.info('Compensating for volume spreading ...')
beamangle = np.deg2rad(ba) / 2.0
dr = self.deltaT * self.cmed * (1.0e9 / 2.0) # depth increment
Vs = 2 * np.pi * (1 - np.cos(beamangle)) / 3 * ((h0 + self.r + dr) ** 3 - (h0 + self.r) ** 3)
r = self.r
r.shape = (r.size, 1)
Vs.shape = (Vs.size, 1)
gain = 1.0 / Vs
if view == True:
from pylab import show, plot, xlabel, ylabel
plot(self.t, gain)
xlabel("Time [ns]")
ylabel("Volume spreading correction")
show()
# One has to make sure that the traces start at the surface and not at the antenna
# otherwise the initial height h0 needs to be corrected.
if power:
self.data = (self.data ** 2 * gain)
else:
self.data = (self.data * gain)
self.done()
def compensateVolSpread(self, h0=0.65, ba=61.0, power=True, view=False):
''' From Davis and Poznyak, IEEE Trans., 1993 '''
self.info('Compensating for volume spreading losses ...')
beamangle = np.deg2rad(ba) / 2.0
dr = self.deltaT * self.cmed * (1.0e9 / 2.0) # depth increment
Vs = 2 * np.pi * (1 - np.cos(beamangle)) / 3 * ((h0 + self.r + dr) ** 3 - (h0 + self.r) ** 3)
r = self.r
r.shape = (r.size, 1)
Vs.shape = (Vs.size, 1)
if view == True:
from pylab import show, plot, xlabel, ylabel
if power:
plot(self.t, ((h0 + r) ** 4 / Vs))
else:
plot(self.t, ((h0 + r) ** 2 / np.sqrt(Vs)))
xlabel("Time [ns]")
ylabel("Volume spreading correction")
show()
# One has to make sure that the traces start at the surface and not at the antenna
# otherwise the initial height h0 needs to be corrected.
if power:
self.data = (self.data ** 2 * ((h0 + r) ** 4 / Vs))
else:
self.data = (self.data * ((h0 + r) ** 2 / np.sqrt(Vs)))
self.done()
def equalizeTraces(self):
'''Calculates the mean value of the whole array and adjust each trace to have this mean value'''
self.info('Equalizing traces ...')
ndx = np.argwhere(self.data != np.nan)
sum = np.nansum(self.data)
s = np.shape(ndx)
allmean = sum / (s[0] * s[1])
print allmean
for i in range(self.traces):
tndx = np.argwhere(self.data[:, i] != np.nan)
tsum = np.nansum(self.data[:, i])
ts = np.shape(tndx)
tmean = tsum / (ts[0] * ts[1])
diff = allmean - tmean
# print allmean, tmean, diff, ts
self.data[:, i] += diff
self.procext = '_eqz'
self.done()
def distanceStack(self, distint=25.0):
'''
Bins traces with regard to there distance from each other.
Apply on raw data file that were converted to the time domain!
Inputs:
dist is an array that contains the distances between the traces - distEach output from GPSprofile.correctTraceNo()
distint is the distance interval over which traces should be stacked
'''
self.info('Averaging traces over %f meter ...' % distint)
averaged = []
avglat = []
avglong = []
avgelev = []
m = 0
stackint = 0.0
# print self.traces
for n in range(self.traces):
if stackint + self.coords.dist[n] < distint:
stackint = stackint + self.coords.dist[n]
# print stackint
else:
averaged.append(np.mean(self.data[:, m:n], axis=1))
avglat.append(np.mean(self.coords.y[m:n], axis=0))
avglong.append(np.mean(self.coords.x[m:n], axis=0))
avgelev.append(np.mean(self.coords.elev[m:n], axis=0))
m = n
stackint = 0.0
# print m, np.add.reduce(self.coords.dist[m:-1])
averaged.append(np.mean(self.data[:, m:-1], axis=1))
avglat.append(np.mean(self.coords.y[m:-1], axis=0))
avglong.append(np.mean(self.coords.x[m:-1], axis=0))
avgelev.append(np.mean(self.coords.elev[m:-1], axis=0))
self.data = np.transpose(np.asarray(averaged))
self.coords.y = np.asarray(avglat)
self.coords.x = np.asarray(avglong)
self.coords.elev = np.asarray(avgelev)
self.coords._getDistance()
self.traces = np.shape(self.data)[1]
self.procext = '_dstk'
self.done()
def removeTraces(self, remove, extend=True):
'''
Removes traces defined in array 'remove' in Tdata!
NEEDS to be updated to remove the correct traces in the HDF file, too.
'''
if extend:
remove = np.arange(remove[0], remove[1] + 1)
self.info('Removing %i traces ...' % len(remove))
old_tracendx = self.tracendx
R = np.delete(self.data, remove, axis=1)
if self.hascoords:
Rx = np.delete(self.coords.x, remove, axis=0)
Ry = np.delete(self.coords.y, remove, axis=0)
Relev = np.delete(self.coords.elev, remove, axis=0)
self.data = R
self.traces = np.shape(self.data)[1]
tracendxmin = int(self.tracendx.min())
tracendxmax = tracendxmin + self.traces
self.tracendx = np.arange(tracendxmin, tracendxmax + 1)
if self.hascoords:
self.coords.x = Rx
self.coords.y = Ry
self.coords.elev = Relev
self.coords._getDistance()
self.done()
print "Removed traces: %i-%i" % (old_tracendx[remove[0]], old_tracendx[remove[-1]])
def staticCorrection(self, s):
'''
Removes all samples above I{s}. Used to clip the data before the
direct wave.
@param s: Number of samples to be removed from top.
@type s: Integer
'''
self.info('Applying static correction ...')
self.data = self.data[s:-1, :]
self.samples = np.shape(self.data)[0]
self.t = np.arange(0, self.deltaT * self.samples, self.deltaT) * self.BW
self.r = self.t * self.cmed / 2.0
self.procext = '_scor'
self.done()
def envelope(self):
'''
NEEDS TESTING
Computes the envelope of the traces using the Hilbert transform.
'''
from scipy.fftpack import hilbert
self.info('Applying Hilbert transform ...')
for trace in range(self.traces):
self.data[:, trace] = np.abs(hilbert(self.data[:, trace]))
self.done()
def correctTopography(self, smoothit=False, sw=5):
'''
Places each trace at its correct elevation.
@param smoothit: Bool(false), decides if the elevation values will be\
smoothed.
@param sw: int(5) is the windowlength of the Hanning window
'''
self.info('Applying topography correction ...')
if smoothit:
from processgpr.util.smooth import smooth
self.coords.elev = smooth(self.coords.elev, window_len=sw, window='hanning')
if self.hascoords:
zmin = self.coords.elev.min() - self.r[-1]
zmax = self.coords.elev.max()
dz = self.r[1] - self.r[0]
zsamples = int((zmax - zmin) / dz + 1)
tmpdata = np.ones((zsamples, self.traces)) * -9999.0
for n in range(self.traces):
zspos1 = int((zmax - self.coords.elev[n]) / dz)
zspos2 = zspos1 + self.samples
tmpdata[zspos1:zspos2, n] = self.data[:, n]
self.data = tmpdata
self.samples = zsamples
self.e = np.flipud(np.arange(zmin, zmax, dz))
self.ylabel = "Elevation"
else:
print "No elevation data found!"
self.done()
def get_phasecenter(self, amp_threshold=1.0, freq_range=[500, 3000],
stacked=1, save=False, view=False):
'''
Calculates the phase center along the profile.
The phase center is -delta_phi/(2 delta_k)
delta_k = delta_f * c0 / (2pi)
The factor 2 in front of delta_k is due to the measuring geometry - the wave travels the distance twice
'''
self.info("Calculating the phase center for each trace ...")
self.phasecenter = np.zeros((self.traces,))
for n in range(self.traces):
trace = TRACE(self.data[:, n], self.deltaT, Nfft=2 ** 15)
trace.get_phasecenter(amp_threshold, freq_range, self.cmed)
self.phasecenter[n] = trace.zphi
if self.hascoords:
scoords = self.coords
if stacked > 1:
from processgpr.util.congrid import congrid
newlength = int(self.phasecenter.shape[0] / stacked)
self.phasecenter = congrid(self.phasecenter, (newlength,))
scoords = self.coords
scoords.stackCoords(newlength)
if save:
if self.hascoords:
np.savez('zphi_coords_' + self.filename, zphi=self.phasecenter,
x=scoords.x, y=scoords.y, z=scoords.elev)
else:
np.savez('zphi_' + self.filename, zphi=self.phasecenter)
if view:
from pylab import show, plot
plot(self.phasecenter)
show()
self.done()
def get_penetrationdepth(self, save=False):
"""
Calculates the penetration depth for each trace.
"""
self.info("Calculating the penetration depth along profile ...")
zp = []
for trace in range(self.traces):
T = TRACE(self.data[:, trace], self.deltaT)
T.get_penetrationdepth(self.r)
zp.append(T.zp)
self.zp = np.asarray(zp, np.float32)
if save:
if self.hascoords:
np.savez('zp_' + self.filename, zp=self.zp,
x=self.coords.x, y=self.coords.y, z=self.coords.elev)
else:
np.savez('zp_' + self.filename, zp=self.zp)
self.done()
def lin2dB(self, power=True):
self.info('Converting to decibel ...')
if power:
self.data = (10 * np.log10(self.data))
else: # intensity needs to be squared: P = I^2
self.data = (10 * np.log10(self.data ** 2))
self.procext = '_dB'
self.done()
def dB2lin(self):
self.info('Converting to linear power scale ...')
self.data = 10. ** (self.data / 10.)
self.procext = '_lin'
self.done()
def norm2max(self):
self.info('Normalizing to maximum value (lin) ...')
for n in range(self.traces):
max_power = np.max(self.data[:, n])
self.data[:, n] = self.data[:, n] / max_power
self.procext = '_norm'
self.done()
def dBnorm2max(self):
self.info('Normalizing to maximum value (dB) ...')
for n in range(self.traces):
max_power = np.max(self.data[:, n])
self.data[:, n] = self.data[:, n] - max_power
self.procext = '_dBnorm'
self.done()
def normalize(self):
self.info('Normalizing linear scale ...')
for n in range(self.traces):
total_power = np.sum(self.data[:, n])
self.data[:, n] = self.data[:, n] / total_power
self.procext = '_norm'
self.done()
def dBnormalize(self):
self.info('Normalizing dB scale ...')
for n in range(self.traces):
total_power = 10 * np.log10(np.add.reduce(10 ** (self.data[:, n] / 10.0)))
self.data[:, n] = self.data[:, n] - total_power
self.procext = '_dBnorm'
self.done()
def powerDecay(self):
self.info('Estimating power decay ...')
for n in range(self.traces):
total_power = np.sum(self.data[:, n])
z_power = np.cumsum(self.data[:, n])
self.data[:, n] = (total_power - z_power) / total_power
self.procext = '_Pdecay'
self.done()
def compensateExtinction(self, k0, A):
self.info('Correcting extinction term ...')
E = np.zeros_like(self.data)
for n in range(self.traces):
E[:, n] = np.exp((2 * k0[n] * self.r[:] + A[n] * self.r[:] ** 2))
from pylab import show, imshow
imshow(E, aspect='auto')
show()
self.data = 10.0 * np.log10(10.0 ** (self.data / 10.0) * E)
self.done()
def set_numfreq(self, val):
self.numfreq = val
self.deltaF = self.BW / (self.numfreq - 1) #: sampling step in the frequency domain
self.nstart = int(self.lowF / self.deltaF)
self.deltaT = 1.0 / (self.deltaF * self.Nfft) #: sampling in the time domain in [ns]
if self.samples:
self.t = np.arange(0, self.deltaT * self.samples, self.deltaT) * 1e9 #: time axis in ns
self.r = self.t * self.cmed / 2.0 #: range axis in m
def set_cmed(self, val):
self.cmed = val
self.r = self.t * self.cmed / 2.0
def readH5(self, filename, tracemin=0, tracemax=-1,
samplemin=0, samplemax=-1, roffset=0.0):
'''
Reads GPR data files in HDF5 format that have been created by L{writeH5}.
@param filename: The name of the input file.
@param tracemin: The first trace to used.
@param tracemax: The last trace to be used.
@param samplemin: The first sample to be used.
@param samplemax: The last sample to be used.
@param roffset: The offset in meters between I{samplemin} and the\
antenna.
'''
self.info('Reading data from file: %s ...' % filename)
''' Get data '''
self.filename = os.path.splitext(filename)[0]
h5file = tables.openFile(filename, mode='r')
try:
self.totalsamples, self.totaltraces = h5file.root.data.traces.shape
except AttributeError:
self.totalsamples, self.totaltraces = h5file.root.profile.traces.shape
print "\n Total", self.totalsamples, self.totaltraces
if tracemax == -1:
tracemax = self.totaltraces
if samplemax == -1:
samplemax = self.totalsamples
self.samplendx = np.arange(samplemin, samplemax)
# print self.samplendx.shape
self.tracendx = np.arange(tracemin, tracemax + 1)
# print self.tracendx
try:
self.data = h5file.root.data.traces[samplemin:samplemax,
tracemin:tracemax]
self.samples, self.traces = np.shape(self.data)
except AttributeError:
self.data = h5file.root.profile.traces[samplemin:samplemax,
tracemin:tracemax] # old format
self.samples, self.traces = np.shape(self.data)
if h5file.root.__contains__('coordinates'):
self.hascoords = True
self.coords = GPS()
self.coords.y = h5file.root.coordinates.Y[tracemin:tracemax]
self.coords.x = h5file.root.coordinates.X[tracemin:tracemax]
self.coords.elev = h5file.root.coordinates.Elevation[tracemin:tracemax]
self.coords._getDistance()
self.d = np.linspace(0, np.add.reduce(self.coords.dist), self.traces)
else:
print "WARNING: No coordinates found"
try:
self.gatefreq = h5file.root.data.traces.attrs.gatefreq
except AttributeError:
print "WARNING: Missing attribute - Gating frequency not defined"
try:
self.nstacks = h5file.root.data.traces.attrs.nstacks
except AttributeError:
print "WARNING: Missing attribute - Number of stacks not defined"
try:
self.deltaT = h5file.root.data.traces.attrs.deltaT
except AttributeError:
print "WARNING: Missing attribute - Time increment not defined"
try:
self.cmed = h5file.root.data.traces.attrs.cmed
except AttributeError:
print "WARNING: Missing attribute - Medium velocity not defined"
h5file.close()
''' Set parameters '''
if samplemax == -1:
samplemax = self.samples
# self.t = self.deltaT*self.samplendx*1e9
self.t = np.arange(0.0, self.deltaT * self.samples, self.deltaT) * 1e9
self.r = (self.t * self.cmed / 2.0) + roffset
self.done()
