Compare FWI result with true model for the Overthrust model
Daniel Köhn Kiel, 16/07/2016
Import Libraries
In [ ]:
import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
from matplotlib.colors import LightSource, Normalize
from matplotlib.pyplot import gca
from pylab import rcParams
from matplotlib import rc
from matplotlib.ticker import FormatStrFormatter
from mpl_toolkits.axes_grid1 import make_axes_locatable
import pickle
Import Colormap
In [ ]:
fp = open('cmap_overthrust.pkl', 'rb')
my_cmap_cm = pickle.load(fp)
fp.close()
FD grid dimensions
In [ ]:
DH = 25.0;
NX = 800;
NY = 186;
Wavefield clip value
In [ ]:
vpmin = 2360.0;
vpmax = 6000.0;
Define fonts
In [ ]:
FSize = 20
font = {'color': 'black',
'weight': 'normal',
'size': FSize}
mpl.rc('xtick', labelsize=FSize)
mpl.rc('ytick', labelsize=FSize)
rcParams['figure.figsize'] = 16, 8
Read FWI result and true model
In [ ]:
f = open("../start/overthrust_true.vp")
data_type = np.dtype ('float32').newbyteorder ('<')
mod_true = np.fromfile (f, dtype=data_type)
mod_true = mod_true.reshape(NX,NY)
mod_true = np.transpose(mod_true)
mod_true = np.flipud(mod_true)
In [ ]:
f = open("29_09_2016_lbfgs_app_Hessian/modelTest_vp_stage_8.bin")
#f = open("../start/overthrust_start_smooth2.vp")
data_type = np.dtype ('float32').newbyteorder ('<')
mod_fwi = np.fromfile (f, dtype=data_type)
mod_fwi = mod_fwi.reshape(NX,NY)
mod_fwi = np.transpose(mod_fwi)
mod_fwi = np.flipud(mod_fwi)
Define Axis
In [ ]:
x = np.arange(0.0, DH*NX, DH)
y = np.arange(0.0, DH*NY, DH)
x = np.divide(x,1000.0);
y = np.divide(y,1000.0);
Define SubPlot
In [ ]:
def do_plot(n, model, cm, an, title, vpmin, vpmax):
ax=plt.subplot(2, 1, n)
ax.set_xticks([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20])
ax.set_yticks([1, 2, 3, 4, 5])
#plt.rc('font',**{'family':'sans-serif','sans-serif':['Helvetica']})
rc('font',**{'family':'sans-serif','sans-serif':['Helvetica']})
## for Palatino and other serif fonts use:
#rc('font',**{'family':'serif','serif':['Palatino']})
#plt.rc('text', usetex=True)
rc('text', usetex=True)
# plt.pcolor(x, y, vp, cmap=cm, vmin=vpmin)
im1 = plt.imshow(model, cmap=cm, interpolation='none', extent=[0.0,NX*DH/1000.0,0.0,NY*DH/1000.0], vmin=vpmin, vmax=vpmax)
a = gca()
a.set_xticklabels(a.get_xticks(), font)
a.set_yticklabels(a.get_yticks(), font)
plt.axis('scaled')
plt.title(title, fontdict=font)
plt.ylabel('Depth [km]', fontdict=font)
if n==2:
plt.xlabel('Distance [km]', fontdict=font)
plt.gca().invert_yaxis()
# add annotation
plt.text(0.1, 0.4,an,fontdict=font,color='white')
# fit and label colorbar
divider = make_axes_locatable(ax)
cax = divider.append_axes("right", size="2.5%", pad=0.05)
cbar = plt.colorbar(im1, cax=cax)
cbar.set_label(r"Vp [m/s]", fontdict=font, labelpad=1)
Plot SubPlots
In [ ]:
plt.close('all')
plt.figure()
do_plot(1, mod_fwi, 'gray_r', '(a)', r"FWI result (f = 20.6 Hz)", vpmin, vpmax)
#do_plot(1, mod_fwi, 'gray_r', '(a)', r"Initial model", vpmin, vpmax)
do_plot(2, mod_true, 'gray_r', '(b)', r"True Overthrust model", vpmin, vpmax)
#plt.savefig('test.png', format='png', dpi=100)
plt.savefig('test.pdf', bbox_inches='tight', format='pdf')
plt.tight_layout()
plt.show()
In [ ]: