-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathextract_bboxes.py
184 lines (147 loc) · 6.43 KB
/
extract_bboxes.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
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
#!/usr/bin/env python
import numpy as np
import os
import six.moves.urllib as urllib
import sys
import tarfile
import tensorflow as tf
import zipfile
import argparse
import csv
import time
from distutils.version import StrictVersion
from collections import defaultdict
from io import StringIO
from PIL import Image
sys.path.append("..")
from object_detection.utils import ops as utils_ops
from utils import label_map_util
from utils import visualization_utils as vis_util
MODEL_NAME = 'ssd_mobilenet_v1_coco_2017_11_17'
MODEL_FILE = MODEL_NAME + '.tar.gz'
DOWNLOAD_BASE = 'http://download.tensorflow.org/models/object_detection/'
PATH_TO_FROZEN_GRAPH = MODEL_NAME + '/frozen_inference_graph.pb'
PATH_TO_LABELS = os.path.join('data', 'mscoco_label_map.pbtxt')
detection_graph = tf.Graph()
with detection_graph.as_default():
od_graph_def = tf.GraphDef()
with tf.gfile.GFile(PATH_TO_FROZEN_GRAPH, 'rb') as fid:
serialized_graph = fid.read()
od_graph_def.ParseFromString(serialized_graph)
tf.import_graph_def(od_graph_def, name='')
category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True)
def load_image_into_numpy_array(image):
(im_width, im_height) = image.size
return np.array(image.getdata()).reshape((im_height, im_width, 3)).astype(np.uint8)
def run_inference_for_single_image(image, graph):
with graph.as_default():
with tf.Session() as sess:
ops = tf.get_default_graph().get_operations()
all_tensor_names = {output.name for op in ops for output in op.outputs}
tensor_dict = {}
for key in [
'num_detections', 'detection_boxes', 'detection_scores',
'detection_classes', 'detection_masks'
]:
tensor_name = key + ':0'
if tensor_name in all_tensor_names:
tensor_dict[key] = tf.get_default_graph().get_tensor_by_name(
tensor_name)
if 'detection_masks' in tensor_dict:
detection_boxes = tf.squeeze(tensor_dict['detection_boxes'], [0])
detection_masks = tf.squeeze(tensor_dict['detection_masks'], [0])
real_num_detection = tf.cast(tensor_dict['num_detections'][0], tf.int32)
detection_boxes = tf.slice(detection_boxes, [0, 0], [real_num_detection, -1])
detection_masks = tf.slice(detection_masks, [0, 0, 0], [real_num_detection, -1, -1])
detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(
detection_masks, detection_boxes, image.shape[0], image.shape[1])
detection_masks_reframed = tf.cast(
tf.greater(detection_masks_reframed, 0.5), tf.uint8)
tensor_dict['detection_masks'] = tf.expand_dims(
detection_masks_reframed, 0)
image_tensor = tf.get_default_graph().get_tensor_by_name('image_tensor:0')
output_dict = sess.run(tensor_dict,
feed_dict={image_tensor: np.expand_dims(image, 0)})
output_dict['num_detections'] = int(output_dict['num_detections'][0])
output_dict['detection_classes'] = output_dict[
'detection_classes'][0].astype(np.uint8)
output_dict['detection_boxes'] = output_dict['detection_boxes'][0]
output_dict['detection_scores'] = output_dict['detection_scores'][0]
if 'detection_masks' in output_dict:
output_dict['detection_masks'] = output_dict['detection_masks'][0]
return output_dict
parser = argparse.ArgumentParser()
parser.add_argument("--src_dir",default='/home/whiterab22bit/src/')
parser.add_argument("--res_dir", default='/home/whiterab22bit/res1/')
parser.add_argument("--csv",default='/home/whiterab22bit/info.csv')
parser.add_argument("--crop_dir", default='/home/whiterab22bit/crop1/')
args = parser.parse_args()
TEST_IMAGE_PATHS = []
tmp_listdir=sorted(os.listdir(args.src_dir))
for f in tmp_listdir:
if (os.path.isfile(os.path.join(args.src_dir, f)) and (f.split('.')[1] == "jpg")):
TEST_IMAGE_PATHS.append(os.path.join(args.src_dir, f))
RES_PATH = args.res_dir
CROP_PATH = args.crop_dir
header = ['Name', 'Classes','Scores']
wr_file = open(args.csv, mode='w')
f_writer = csv.writer(wr_file, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
f_writer.writerow(header)
wr_file.close()
tmpboxes = 0
tmpclasses = 0
tmpscores = 0
ms1 = time.time()*1000.0
for image_path in TEST_IMAGE_PATHS:
image = Image.open(image_path)
image_np = load_image_into_numpy_array(image)
output_dict = run_inference_for_single_image(image_np, detection_graph)
PERSON = 1
THRESH = 0.0
boxes= output_dict['detection_boxes']
classes = output_dict['detection_classes']
scores = output_dict['detection_scores']
MASK = (classes==PERSON) & (scores >= THRESH)
classes_ = np.reshape(classes[MASK],(-1,1))
scores_ = np.reshape(scores[MASK], (-1,1))
d = np.hstack((classes_,scores_))
name = os.path.basename(image_path)
wr_file1 = open(args.csv, mode='a')
f_writer1 = csv.writer(wr_file1, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
for l in range(d.shape[0]):
print ("writing to CSV file")
f_writer1.writerow([name]+list(d[l]))
wr_file1.close()
mage_np = load_image_into_numpy_array(image)
image_np_expanded = np.expand_dims(image_np, axis=0)
output_dict = run_inference_for_single_image(image_np, detection_graph)
tmpboxes = output_dict['detection_boxes']
tmpclasses = output_dict['detection_classes']
tmpscores = output_dict['detection_scores']
vis_util.visualize_boxes_and_labels_on_image_array(
image_np,
output_dict['detection_boxes'],
output_dict['detection_classes'],
output_dict['detection_scores'],
category_index,
instance_masks=output_dict.get('detection_masks'),
use_normalized_coordinates=True,
line_thickness=8)
full_pic = Image.fromarray(image_np)
print ("saving full picture to res_dir")
full_pic.save(RES_PATH + name)
(h,w,_) = image_np.shape
b=tmpboxes[MASK]
for i in range(b.shape[0]):
print ("finding dimensions of bounding box")
y0 = b[i][0]*h
y1= b[i][2]*h
x0= b[i][1]*w
x1= b[i][3]*w
croped = image.crop((x0,y0,x1,y1))
orig_name=name.split('.')
print ("saving croped picture to crop_dir")
new_f = CROP_PATH + orig_name[0]+'-' + str(i)+'.'+orig_name[1]
croped.save(new_f, 'JPEG')
ms2 = time.time()*1000
print (ms2-ms1)