Clustering/CreateDatabase.py at master · Awish021/Clustering · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
import os
import xlrd
from DS import Crop
import pylab
import numpy as np
import math
import pickle
import cv2

def get_files_mb_pk():
    path = '/home/avishay/Project/RawData'
    images = []
    for root, dirs, files in os.walk(path):
        for name in files:
            if name.endswith(".png"):
                images.append(os.path.join(root, name))
       # for name in dirs:
        #   print name
    return images

def convert_to_int(item):
    if isinstance(item,basestring):
        return item
    else:
        return int(item)

def create_crops(path,mb_list):
    crops=[]
    wb = xlrd.open_workbook(path)
    sheet = wb.sheet_by_index(0)
    for row_num in range(sheet.nrows)[1:]:
        row_value = sheet.row_values(row_num)
        mb= str(int(row_value[8]))
        for record in mb_list:
            reversed = record[::-1]
            if mb == (reversed[reversed.find(".")+1:reversed.find("_")])[::-1]:
                url=record
        row_value.append(url)
        row_value.append(url.replace('RawData','BoundingBoxes'))
        row_value=map(convert_to_int,row_value[1::])
        crops.append(Crop(*row_value))
    return crops


def get_circle(im):
    x,y = im.shape[:2]
    return (x/2,y/2,(x+y)/10)


def generate_gaussian_func(mean,sigma):
    def gaussian_func(x):
        var = float(sigma) ** 2
        pi = np.pi
        denom = (2 * pi * var) ** .5
        num = math.exp(-(float(x) - float(mean)) ** 2 / (2 * var))
        return num / denom
    return gaussian_func

def pixel_value(xn,yn,cx,cy,x,y):
    return math.ceil(xn(x)*cx*yn(y)*cy)


def get_gaussian(im):
    cenx, ceny, rad = get_circle(im)
    xn = generate_gaussian_func(cenx,rad)
    cx = math.sqrt(10)/xn(cenx)
    yn = generate_gaussian_func(ceny,rad)
    cy = math.sqrt(10)/yn(ceny)
    return (xn,yn,cx,cy)

def get_im_array(im):
    arr = []
    sobelx = cv2.Sobel(im, cv2.CV_64F, 1, 0, ksize=5)
    sobely = cv2.Sobel(im, cv2.CV_64F, 0, 1, ksize=5)
    xn,yn,cx,cy= get_gaussian(im)
    pixx,pixy=im.shape[:2]
    for x in range(pixx):
        for y in range(pixy):
            lst = [x,y]
            lst.extend(im[x, y])
            lst.extend(sobelx[x, y])
            lst.extend(sobely[x, y])

            times = int( pixel_value(xn,yn,cx,cy,x,y))
            arr.extend([lst]*times)
    return np.asarray(arr)

def image_to_matrix(im):
    arr = get_im_array(im)
    arr.astype(float)
    xs = arr[:,0].ravel()
    ys = arr[:,1].ravel()
    r = arr[:,2].ravel()
    g = arr[:,3].ravel()
    b = arr[:,4].ravel()
    # drdx= arr[:,5].ravel()
    # dgdx = arr[:, 6].ravel()
    # dbdx = arr[:, 7].ravel()
    # drdy = arr[:, 8].ravel()
    # dgdy = arr[:, 9].ravel()
    # dbdy = arr[:, 10].ravel()
    #
    # f= np.vstack([xs,ys,r,g,b,drdx,dgdx,dbdx,drdy,dgdy,dbdy])
    #
    # f = np.dot(np.diag([1,1,3,3,3,5,5,5,5,5,5]),f)
    f = np.vstack([xs,ys,r,g,b])
    S = np.cov(f)
    return S

def image_to_vector(im):
    mat = image_to_matrix(im)
    arr = []
    for x in range(mat.shape[1]):
        for y in range(mat.shape[1]):
            if y<=x:
                arr.extend([mat[x,y]])
    return np.asarray(arr)


# def createBoundedBoxes(Objs):
#     count = 0
#     for obj in objs:
#         count=count+1
#
#         im =  pylab.imread(obj.url)
#         x,y=im.shape[:2]
#
#         print im.shape
#         print str(x) + ',' + str(y) +','+obj.url
#         boundedBox=im[x/3:2*x/3,y/4:3*y/4,:]
#         pylab.imsave(obj.bburl,boundedBox)
# createBoundedBoxes(objs)

objs = create_crops('/home/avishay/Project/crop_metadata.xlsx',get_files_mb_pk())

sum =0

path = '/home/avishay/Project/pickle-files/'
for obj in objs:
    sum=sum+1
    print sum
    im = cv2.imread(obj.url)
    obj.data=image_to_vector(im).tolist()
    filename = path + str(obj.probe_fk) + "_" + str(obj.mb_pk) + ".p"
    pickle.dump(obj, open(filename, "wb"))

# im = pylab.imread(objs[149].url)
#
# cenx,ceny,rad = get_circle(im)
# x,y = im.shape[:2]
# xn,yn,cx,cy = get_gaussian(im)
# print pixel_value(xn,yn,cx,cy,0,0)
#
# arr = image_to_vector(im)
#