-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathpredictor.py
132 lines (111 loc) · 5.46 KB
/
predictor.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
import os
import collections
from glob import glob
import numpy as np
import tensorflow as tf
from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_util
from PIL import Image
# What model to use
MODEL_NAME = 'ssd_inception_v2_coco_11_06_2017_udacity_real_inference_graph'
# Path to frozen detection graph. This is the actual model that is used for the object detection.
PATH_TO_CKPT = "./frozen_graphs/" + MODEL_NAME + '/frozen_inference_graph.pb'
# List of the strings that is used to add correct label for each box.
PATH_TO_LABELS = os.path.join('training_data', 'tl_label_map.pbtxt')
NUM_CLASSES = 4
#PATH_TO_TEST_IMAGES_DIR = 'test_images_real'
PATH_TO_TEST_IMAGES_DIR = '../../our-team/Capstone/data/traffic_light/loop_with_traffic_light/'
TEST_IMAGE_PATHS = glob(os.path.join(PATH_TO_TEST_IMAGES_DIR, '*.jpg'))
#TEST_IMAGE_PATHS = [os.path.join(PATH_TO_TEST_IMAGES_DIR, 'left0000.jpg'),
# os.path.join(PATH_TO_TEST_IMAGES_DIR, 'left0285.jpg'),
# os.path.join(PATH_TO_TEST_IMAGES_DIR, 'left0350.jpg')]
detection_graph = tf.Graph()
def load_frozen_model():
with detection_graph.as_default():
od_graph_def = tf.GraphDef()
with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
serialized_graph = fid.read()
od_graph_def.ParseFromString(serialized_graph)
tf.import_graph_def(od_graph_def, name='')
def load_label_map():
label_map = label_map_util.load_labelmap(PATH_TO_LABELS)
categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)
category_index = label_map_util.create_category_index(categories)
return categories, category_index
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 get_category(categories, index):
for category in categories:
if category['id'] == index:
return category
return None
def extract_box(box_tuple, image):
ymin, xmin, ymax, xmax = box_tuple[0]
im_width, im_height = image.size
(left, right, top, bottom) = (xmin * im_width,
xmax * im_width,
ymin * im_height,
ymax * im_height)
return (int(left), int(right), int(top), int(bottom))
def predict(categories, category_index):
with detection_graph.as_default():
with tf.Session(graph=detection_graph) as sess:
# Definite input and output Tensors for detection_graph
image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
# Each box represents a part of the image where a particular object was detected.
detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
# Each score represent how level of confidence for each of the objects.
# Score is shown on the result image, together with the class label.
detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')
detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')
num_detections = detection_graph.get_tensor_by_name('num_detections:0')
for image_path in TEST_IMAGE_PATHS:
image = Image.open(image_path)
# the array based representation of the image will be used later in order to prepare the
# result image with boxes and labels on it.
image_np = load_image_into_numpy_array(image)
# Expand dimensions since the model expects images to have shape: [1, None, None, 3]
image_np_expanded = np.expand_dims(image_np, axis=0)
# Actual detection.
(boxes, scores, classes, num) = sess.run(
[detection_boxes, detection_scores, detection_classes, num_detections],
feed_dict={image_tensor: image_np_expanded})
# Visualization of the results of a detection.
#vis_util.visualize_boxes_and_labels_on_image_array(
# image_np,
# np.squeeze(boxes),
# np.squeeze(classes).astype(np.int32),
# np.squeeze(scores),
# category_index,
# use_normalized_coordinates=True,
# line_thickness=8)
print(os.path.basename(image_path), end=",")
top_score = scores[0][0]
if top_score < 0.5:
print("Unknown")
else:
class_index = int(classes[0][0])
category = get_category(categories, class_index)
if category is not None:
print("{},{}".format(category['name'], top_score))
#print("\tBounding box for the highest score object (l,r,t,b):{}".format(extract_box(tuple(boxes[0].tolist()), image)))
#pick the highest score for now
#for i in range(3):
# score = scores[0][i]
# class_index = int(classes[0][i])
# category = get_category(categories, class_index)
# if category is not None:
# print("\tRank:{} label:{} class_index:{} prob:{:0.4f}".format(i, category['name'], class_index, score))
#counted = collections.Counter(classes.tolist()[0])
#most_common_class_index=int(counted.most_common()[0][0])
#category = get_category(categories, most_common_class_index)
#if category is not None:
# print("\tMost commonly identified label:{}".format(category['name']))
def main(_):
load_frozen_model()
categories, category_index = load_label_map()
predict(categories, category_index)
if __name__ == '__main__':
tf.app.run()