In [16]:

    

%matplotlib notebook
import numpy as np
import matplotlib.pyplot as plt
import h5py


    

In [2]:

    

fpath = '/Users/lili/Research/Experiment/SRX_Aakriti/2015_11_13_22_26_pyxrf.h5'


    

In [3]:

    

with h5py.File(fpath, 'r') as f:
    d = f['xrfmap/detsum/counts'][:]


    

In [4]:

    

# 3D fluorescence data
# shape is [number of row, number of col, number of points in spectrum]
d.shape


    

    
Out[4]:





(10, 10, 4096)

In [11]:

    

# x axis as energy in kev
x = np.arange(4096)
x = 0.01*x             # approximately transfer channel number to kev

# choose two areas  
d1 = d[0:5,0:5,:]   # define y position from 0 to 5, x from 0 to 5, top left area
d2 = d[5:10,5:10,:] # define y position from 5 to 10, choose x from 5 to 10, bottom right area  

# sum spectrum in those two areas
y1 = np.sum(d1, axis=(0,1))   # sum along axis 0 and 1 to get summed 1D spectrum curve
y2 = np.sum(d2, axis=(0,1))


    

In [23]:

    

# compare spectrum in those two areas
fig, ax = plt.subplots()
ax.plot(x, y1, label='area1')
ax.plot(x, y2, label='area2')
ax.legend()


    

    















    












    
Out[23]:





<matplotlib.legend.Legend at 0x1199167d0>

In [24]:

    

# log plot
fig, ax = plt.subplots()
ax.semilogy(x, y1, label='area1')
ax.semilogy(x, y2, label='area2')
ax.legend()


    

    















    












    
Out[24]:





<matplotlib.legend.Legend at 0x11a344210>

In [25]:

    

# log plot and zoom in
fig, ax = plt.subplots()
ax.semilogy(x, y1, label='area1')
ax.semilogy(x, y2, label='area2')
ax.legend()
ax.set_xlim([1,13])


    

    















    












    
Out[25]:





(1, 13)

In [15]:

    

# choose a roi to plot 2D image, like Fe
d_fe = d[:, :, 610:660]
fe = np.sum(d_fe, axis=(2))   # sum along axis 2
fig, ax = plt.subplots()
ax.imshow(fe)


    

    















    












    
Out[15]:





<matplotlib.image.AxesImage at 0x11236d410>

In [ ]: