forked from MEDAL-IITB/Fast_WSI_Color_Norm
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
132 lines (108 loc) · 4.83 KB
/
main.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
# Please cite both papers if you use this code:
# @article{ramakrishnan_fast_2019,
# title = {Fast {GPU}-{Enabled} {Color} {Normalization} for {Digital} {Pathology}},
# url = {http://arxiv.org/abs/1901.03088},
# urldate = {2019-01-11},
# journal = {arXiv:1901.03088 [cs]},
# author = {Ramakrishnan, Goutham and Anand, Deepak and Sethi, Amit},
# month = jan,
# year = {2019},
# note = {arXiv: 1901.03088},
# keywords = {Computer Science - Computer Vision and Pattern Recognition},
# file = {arXiv\:1901.03088 PDF:/home/deepakanand/Zotero/storage/PBAERTSZ/Ramakrishnan et al. - 2019 - Fast GPU-Enabled Color Normalization for Digital P.pdf:application/pdf;arXiv.org Snapshot:/home/deepakanand/Zotero/storage/FH5SW7R3/1901.html:text/html}
# }
# @inproceedings{Vahadane2015ISBI,
# Author = {Abhishek Vahadane and Tingying Peng and Shadi Albarqouni and Maximilian Baust and Katja Steiger and Anna Melissa Schlitter and Amit Sethi and Irene Esposito and Nassir Navab},
# Booktitle = {IEEE International Symposium on Biomedical Imaging},
# Date-Modified = {2015-01-31 17:49:35 +0000},
# Title = {Structure-Preserved Color Normalization for Histological Images},
# Year = {2015}}
# input image should be color image
# Python implementation by: Goutham Ramakrishnan, [email protected] and Deepak Anand, [email protected]
import glob
import sys
import os
from Run_StainSep import run_stainsep
from Run_ColorNorm import run_colornorm, run_batch_colornorm
#setting tensorflow verbosity
os.environ['TF_CPP_MIN_LOG_LEVEL']='3'
#to use cpu instead of gpu, uncomment the below line
os.environ['CUDA_VISIBLE_DEVICES'] = '1' #use only GPU-0
#os.environ['CUDA_VISIBLE_DEVICES'] = '' #use only CPU
import tensorflow as tf
gpu_options=tf.GPUOptions(per_process_gpu_memory_fraction=1)
# config = tf.ConfigProto(device_count={'GPU': 1},log_device_placement=False,gpu_options=gpu_options)
config = tf.ConfigProto(log_device_placement=False,gpu_options=gpu_options)
#Parameters
nstains=2 #number of stains
lamb=0.01 #default value sparsity regularization parameter
# lamb=0 equivalent to NMF
# 1= stain separation of all images in a folder
# 0= stain separation of single image
# 2= color normalization of one image with one target image
# 3= color normalization of all images in a folder with one target image
# 4= color normalization of one image with multiple target images
# 5= color normalization of all images in a folder with multiple target images individually
op=0
if op==0:
filename="/home/deepakanand/Downloads/registered_data_for_color_normalization/1HE.tif"
# filename="./experiment_CN/PrognosisTMABlock2_B_2_5_H&E1.tif"
# filename="./Target/TCGA-E2-A14V-01Z-00-DX1.tif"
# filename = "./Target/he.png"
print filename
run_stainsep(filename,nstains,lamb)
elif op==1:
input_direc="./deepak_he/"
output_direc="./stain_separated/"
if not os.path.exists(output_direc):
os.makedirs(output_direc)
file_type="*"
if len(sorted(glob.glob(input_direc+file_type)))==0:
print "No source files found"
sys.exit()
filenames=sorted(glob.glob(input_direc+file_type))
print filenames
for filename in filenames:
run_stainsep(filename,nstains,lamb,output_direc=output_direc)
elif op==2:
level=0
output_direc="./new_tests/"
if not os.path.exists(output_direc):
os.makedirs(output_direc)
source_filename="./Test/WSI/57_HE: 10000x11534.png"
# source_filename="./experiment_CN/PrognosisTMABlock3_A_5_6_H&E1.tif"
#source_filename="./Test0/20_HE(28016,22316).png"
#source_filename="./Source/TUPAC-TR-004.svs"
#source_filename="../../Downloads/01/no1_HE.ndpi"
#source_filename = "../../Documents/Shubham/Data/Test Data/78/78_HE.ndpi"
target_filename="./Target/TCGA-E2-A14V-01Z-00-DX1.tif"
# target_filename="b048.tif"
if not os.path.exists(source_filename):
print "Source file does not exist"
sys.exit()
if not os.path.exists(target_filename):
print "Target file does not exist"
sys.exit()
background_correction = True
run_colornorm(source_filename,target_filename,nstains,lamb,output_direc,level,background_correction,config=config)
elif op==3:
level=0
# input_direc="../experiment_CN/"
input_direc="../Test/WSI patches/"
output_direc="./new_tests_WSI/GPU/"
if not os.path.exists(output_direc):
os.makedirs(output_direc)
file_type="*.*"
#file_type="*.svs" #all of these file types from input_direc will be normalized
target_filename="../Target/TCGA-E2-A14V-01Z-00-DX1.tif"
if not os.path.exists(target_filename):
print "Target file does not exist"
sys.exit()
if len(sorted(glob.glob(input_direc+file_type)))==0:
print "No source files found"
sys.exit()
#filename format of normalized images can be changed in run_batch_colornorm
filenames=[target_filename]+sorted(glob.glob(input_direc+file_type))
background_correction = True
# background_correction = False
run_batch_colornorm(filenames,nstains,lamb,output_direc,level,background_correction,config)