DS_LAB_Project/ImageDatabaseBuilder_MultiThread.py at master · grimmdaniel/DS_LAB_Project · 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
import numpy as np
import cv2
import sys
import datetime
import multiprocessing as mp

window_size = 106
border_size = 3

def readImageWindowLocationsFile():
	file = open('train_set/window_location.csv','r')
	return file

def createWindowLocations(line):
	splittedLine = line.split(',')
	return(splittedLine[0],(int(splittedLine[1]),int(splittedLine[2])),(int(splittedLine[4]),int(splittedLine[5])))

def openImageWindow(image):
	image_name = image[0]
	img = cv2.imread('train_set/raw_images_train/' + image_name)
	fix_img = cv2.cvtColor(img,cv2.COLOR_BGR2RGB)

	window_inside_x = image[1][0]
	window_inside_y = image[1][1]

	window_outside_x = image[2][0]
	window_outside_y = image[2][1]

	window_inside = fix_img[window_inside_y:window_inside_y+window_size,window_inside_x:window_inside_x+window_size]
	window_outside = fix_img[window_outside_y:window_outside_y+window_size,window_outside_x:window_outside_x+window_size]
	return (window_inside,window_outside)

def processImageWindows(q,window,is_inside):
	for i in range(border_size,window_size-border_size):
		for j in range(border_size,window_size-border_size):
			pixel_region = window[i-border_size:i+border_size+1,j-border_size:j+border_size+1]
			result = processPixelsWithRegions(pixel_region)
			if result != '':
				writeResultToFile(q,result,is_inside)


def processPixelsWithRegions(pixel_region):
	#image processing methods here...
	# means of R G B channels
	red_channel = np.reshape(pixel_region[:,:,0], -1)
	green_channel = np.reshape(pixel_region[:,:,1], -1)
	blue_channel = np.reshape(pixel_region[:,:,2], -1)

	center_of_pixel_region = pixel_region[border_size:border_size+1,border_size:border_size+1,:]
	center_red_pixel = center_of_pixel_region[:,:,0:1].item()
	center_green_pixel = center_of_pixel_region[:,:,1:2].item()
	center_blue_pixel = center_of_pixel_region[:,:,2:3].item()

	std_red = np.std(red_channel)
	std_green = np.std(green_channel)
	std_blue = np.std(blue_channel)

	if std_red == 0 or std_green == 0 or std_blue == 0:
		return ''

	#correlation between Red and Green channels
	corr_rg = np.corrcoef(red_channel,green_channel)[0][1]

	#correlation between Red and Blue channels
	corr_rb = np.corrcoef(red_channel,blue_channel)[0][1]

	#correlation between Green and Blue channels
	corr_gb = np.corrcoef(green_channel,blue_channel)[0][1]

	#means of rgb channels + standard deviation of rgb channels
	result = str(center_red_pixel) + ',' + str(center_green_pixel) + ',' + str(center_blue_pixel) + ',' + str(np.mean(red_channel)) + ',' + str(np.mean(green_channel)) + ',' + str(np.mean(blue_channel)) + ',' + str(std_red) + ',' + str(std_green) + ',' + str(std_blue)  + ',' + str(corr_rg)  + ',' + str(corr_rb)  + ',' + str(corr_gb)
	return result

def createFileForOutput():
	o_file = open('train_set/output_train_set_multi.csv','w+')
	return o_file

def writeResultToFile(q,result,is_inside):
	if is_inside:
		q.put(result+',1\n')
	else:
		q.put(result+',0\n')

t1=datetime.datetime.utcnow()
output_file = createFileForOutput()
output_file.write('red_pixel,green_pixel,blue_pixel,mean_red,mean_green,mean_blue,std_red,std_green,std_blue,corr_rg,corr_rb,corr_gb,inside\n')
file = readImageWindowLocationsFile()

def processLine(line,q):
	currentImageWindows = createWindowLocations(line)
	window_inside,window_outside = openImageWindow(currentImageWindows)
	processImageWindows(q,window_inside,True)
	processImageWindows(q,window_outside,False)
	print('finished with line: ' + line)

def listener(q):
	global output_file

	while 1:
		m = q.get()
		if m == 'kill':
			break
		output_file.write(str(m))
		output_file.flush()


if __name__ == "__main__":

	manager = mp.Manager()
	q = manager.Queue()
	pool = mp.Pool(mp.cpu_count() + 2)

    #put listener to work first
	watcher = pool.apply_async(listener, (q,))

	jobs = []

	for line in file:
		job = pool.apply_async(processLine, (line,q))
		jobs.append(job)

	for job in jobs:
		job.get()

	q.put('kill')
	pool.close()
	pool.join()

	file.close()
	output_file.close()
	print('Database exported successfully')
	t2=datetime.datetime.utcnow()
	print('Elapsed time: ' + str(t2-t1))