In [2]:

    

import cv2
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
%matplotlib inline


    

In [17]:

    

afmdata = np.genfromtxt('../NaRWHAL/Data/BackgroundedTXTFiles/500umNetwork-SubstrateArea')
afmdata= afmdata*(10**9)
height, width = afmdata.shape
afmimg=np.zeros((height, width, 3))


    

In [18]:

    

factor=(255)/(afmdata.max()-afmdata.min())
for i in range(height):
    for j in range(width):
        intensity=np.int((afmdata[i][j]-afmdata.min())*factor)
        #afmimg[i][j]=np.array([np.int((afmdata[i][j]-afmdata.min())*factor),0,0])
        afmimg[i][j]=np.array([intensity, intensity, intensity])


    

In [19]:

    

afmimg = np.uint8(afmimg)
grayscaled = cv2.cvtColor(afmimg,cv2.COLOR_BGR2GRAY)
plt.matshow(grayscaled, origin = 'lower', cmap = 'viridis')


    

    
Out[19]:





<matplotlib.image.AxesImage at 0x112d40f98>






    

In [20]:

    

background = cv2.adaptiveThreshold(grayscaled, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 71, 3)
plt.matshow(background, origin = 'lower', cmap = 'viridis')


    

    
Out[20]:





<matplotlib.image.AxesImage at 0x112ce69b0>






    

In [14]:

    

backgrounded = grayscaled - background
plt.matshow(backgrounded, origin = 'lower', cmap = 'viridis')


    

    
Out[14]:





<matplotlib.image.AxesImage at 0x11391d5f8>






    

In [15]:

    

retval2,threshold2 = cv2.threshold(backgrounded, 0, 255, cv2.THRESH_BINARY+cv2.THRESH_OTSU)
plt.matshow(threshold2, origin = 'lower', cmap = "viridis")


    

    
Out[15]:





<matplotlib.image.AxesImage at 0x1138e8e80>






    

In [16]:

    

img = backgrounded
equ = cv2.equalizeHist(img)
res = np.hstack((img,equ)) #stacking images side-by-side
plt.matshow(res, cmap = 'viridis')
plt.title('Backgrounded Image with and without Histogram Equalization')


    

    
Out[16]:





<matplotlib.text.Text at 0x1135b25c0>






    

In [26]:

    

def histogram_equalization(source):
    """
    Takes in a uint8 image and equalizes the histogram
    """
    img = source
    equ = cv2.equalizeHist(img)
    
    return equ


    

In [27]:

    

test1 = histogram_equalization(backgrounded)


    

In [28]:

    

plt.matshow(test1)


    

    
Out[28]:





<matplotlib.image.AxesImage at 0x1121215f8>






    

In [ ]: