# -*- coding: utf-8 -*-

"""

Created on Tue Apr 18 16:52:47 2017

@author: mimtiaz

"""

import cv2

import numpy as np

from skimage.feature import greycomatrix, greycoprops

import matplotlib.pyplot as plt

import matplotlib.patches as mpatches

with open('fileName.txt') as f: #save directories in 'fileName.txt' of all files

content = f.readlines()

#c = 0

for i in xrange(10, 11): #len(content)

img = cv2.imread((content[i])[:-1])

img = img[:,:,0]

# ret,thresh1 = cv2.threshold(img,127,1,cv2.THRESH_BINARY)

# cv2.imwrite('test1.bmp', thresh1)

# resizedImg = cv2.resize(thresh1, (4,4))

# cv2.imwrite('test2.bmp', resizedImg)

step = range(1,256)

step = np.asarray(step)

# step = [2]

angle = [0,np.pi/2]

coOccuranceMat = greycomatrix(img, step, angle, levels = 256, symmetric = True, normed = True )

# coOccuranceMat = greycomatrix(resizedImg, step, angle, levels = 2, symmetric = True)

#print coOccuranceMat[:,:,1,0]

ll = coOccuranceMat[:,:,4,0]

contrast = greycoprops(coOccuranceMat, 'contrast')

dissimilarity = greycoprops(coOccuranceMat, 'dissimilarity')

homogeneity = greycoprops(coOccuranceMat, 'homogeneity')

energy = greycoprops(coOccuranceMat, 'energy')

correlation = greycoprops(coOccuranceMat, 'correlation')

ASM = greycoprops(coOccuranceMat, 'ASM')

texturelist = {0: 'contrast', 1: 'dissimilarity', 2: ' homogeneity', 3: 'energy', 4: 'correlation', 5: 'ASM'}

marker = {0: 'b-o', 1: 'r-^'}

for j in range(0,6):

c = 1

for k in range(0, len(angle)): #len(angle)

plt.figure(j)

plt.plot(step, eval(texturelist[j])[:,k], marker[k], alpha = 0.8)

# plt.plot(step, eval(texturelist[j])[:,0], 'b-*')

plt.xlabel('distance from 1 ro 255')

plt.ylabel(texturelist[j])

plt.title(texturelist[j] + ' vs. distance')

redPatch = mpatches.Patch(color = 'red', label = 'Angle at 90')

bluePatch = mpatches.Patch(color = 'blue', label = 'Angle at 0')

plt.legend(handles = [bluePatch, redPatch])

# plt.xlim(0,256)

# plt.ylim(-1.5,1.5)

plt.show(j)

c = c +1