-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathutils.py
168 lines (141 loc) · 5.58 KB
/
utils.py
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
156
157
158
159
160
161
162
163
164
165
166
167
168
#!/usr/bin/env python2
# -*- coding: utf-8 -*-
"""
@author: liuyaqi
"""
import torch
import torch.nn as nn
from torch.autograd import Variable
import cv2
import random
import numpy as np
import dmac_vgg_skip as dmac_vgg
import os
import math
def load_pairs_csv( input_pairs_csv_file) :
'''Load input csv file for splicing detection and localization
'''
if ( not os.path.isfile( input_pairs_csv_file ) ) :
raise IOError, "ERROR: cannot locate input splicing task csv file %s" % input_pairs_csv_file
with open( input_pairs_csv_file, 'r' ) as IN :
lines = [ line.strip() for line in IN.readlines() ]
headers = [ h.lower() for h in lines.pop(0).split(',') ]
assert ( 'image1' in headers ) and ( 'image2' in headers ) and ( 'label' in headers) and ('gt1' in headers) and ('gt2' in headers), "ERROR: csv file error"
pair_list = []
for line in lines :
fields = line.split(',')
lut = dict( zip( headers, fields ) )
pair_list.append((lut['image1'], lut['image2' ], lut['label'], lut['gt1'], lut['gt2']))
return pair_list
def imreadtonumpy(data_path,piece,input_scale):
img1 = np.zeros((input_scale,input_scale,3))
img_temp = cv2.imread(os.path.join(data_path,piece)).astype(float)
img_original1 = img_temp
img_temp = cv2.resize(img_temp,(input_scale,input_scale)).astype(float)
img_temp[:,:,0] = img_temp[:,:,0] - 104.008
img_temp[:,:,1] = img_temp[:,:,1] - 116.669
img_temp[:,:,2] = img_temp[:,:,2] - 122.675
img1[:img_temp.shape[0],:img_temp.shape[1],:] = img_temp
return img1,img_original1
def chunker(seq, size):
return (seq[pos:pos+size] for pos in xrange(0,len(seq), size))
def flip(I,flip_p):
if flip_p>0.5:
return np.fliplr(I)
else:
return I
def get_data_from_chunk(data_path,chunk,dim):
images1 = np.zeros((dim,dim,3,len(chunk)))
images2 = np.zeros((dim,dim,3,len(chunk)))
labels = np.zeros((len(chunk)))
gt1 = np.zeros((dim,dim,1,len(chunk)))
gt2 = np.zeros((dim,dim,1,len(chunk)))
for i,piece in enumerate(chunk):
flip_p = random.uniform(0, 1)
img_temp = cv2.imread(os.path.join(data_path,piece[0])).astype(float)
img_temp = cv2.resize(img_temp,(dim,dim)).astype(float)
img_temp[:,:,0] = img_temp[:,:,0] - 104.008
img_temp[:,:,1] = img_temp[:,:,1] - 116.669
img_temp[:,:,2] = img_temp[:,:,2] - 122.675
img_temp = flip(img_temp,flip_p)
images1[:,:,:,i] = img_temp
img_temp = cv2.imread(os.path.join(data_path,piece[1])).astype(float)
img_temp = cv2.resize(img_temp,(dim,dim)).astype(float)
img_temp[:,:,0] = img_temp[:,:,0] - 104.008
img_temp[:,:,1] = img_temp[:,:,1] - 116.669
img_temp[:,:,2] = img_temp[:,:,2] - 122.675
img_temp = flip(img_temp,flip_p)
images2[:,:,:,i] = img_temp
label_tmp = int(piece[2])
labels[i] = label_tmp
if label_tmp == 1:
gt_temp = cv2.imread(os.path.join(data_path,piece[3]))[:,:,0]
gt_temp[gt_temp == 255] = 1
gt_temp = cv2.resize(gt_temp,(dim,dim) , interpolation = cv2.INTER_NEAREST)
gt_temp = flip(gt_temp,flip_p)
gt1[:,:,0,i] = gt_temp
gt_temp = cv2.imread(os.path.join(data_path,piece[4]))[:,:,0]
gt_temp[gt_temp == 255] = 1
gt_temp = cv2.resize(gt_temp,(dim,dim) , interpolation = cv2.INTER_NEAREST)
gt_temp = flip(gt_temp,flip_p)
gt2[:,:,0,i] = gt_temp
images1 = images1.transpose((3,2,0,1))
images1 = torch.from_numpy(images1).float()
images2 = images2.transpose((3,2,0,1))
images2 = torch.from_numpy(images2).float()
a = dmac_vgg.outS(dim)
gt1 = resize_label(gt1,a)
gt1 = gt1.transpose((3,2,0,1))
gt1 = torch.from_numpy(gt1).float()
gt2 = resize_label(gt2,a)
gt2 = gt2.transpose((3,2,0,1))
gt2 = torch.from_numpy(gt2).float()
labels = torch.from_numpy(labels).long()
flip_im = random.uniform(0, 1)
if flip_im > 0.5:
images = images1
gt = gt1
images1 = images2
gt1 = gt2
images2 = images
gt2 = gt
return images1, images2, labels, gt1, gt2
def resize_label(label, size):
label_resized = np.zeros((size,size,1,label.shape[3]))
interp = nn.UpsamplingBilinear2d(size=(size, size))
labelVar = Variable(torch.from_numpy(label.transpose(3, 2, 0, 1)))
label_resized[:, :, :, :] = interp(labelVar).data.numpy().transpose(2, 3, 1, 0)
return label_resized
def load_pairs(data_path, subpath_list):
pair_list = []
for subpath in subpath_list:
list_path = data_path + 'labelfiles/' + subpath
pair_list_ = load_pairs_csv(list_path)
pair_list.extend(pair_list_)
np.random.shuffle(pair_list)
print 'Pair list length:',len(pair_list)
return pair_list
def fast_hist(a, b, n):
k = (a >= 0) & (a < n)
return np.bincount(n * a[k].astype(int) + b[k], minlength=n**2).reshape(n, n)
def get_NMM(hist,gt):
gt_size = float(np.sum(gt))
tp_size = float(hist[1,1])
fn_size = float(hist[1,0])
fp_size = float(hist[0,1])
if gt_size == 0:
return 0
nmm = (tp_size - fn_size - fp_size)/gt_size
if nmm < -1.0:
nmm = -1.0
return nmm
def get_MCC(hist):
tp = float(hist[1,1])
tn = float(hist[0,0])
fn = float(hist[1,0])
fp = float(hist[0,1])
denominator = math.sqrt((tp+fp)*(tp+fn)*(tn+fp)*(tn+fn))
if denominator == 0:
return 0
mcc = (tp*tn - fp*fn)/denominator
return mcc