def find_matches_in_cluster(cluster_and_descriptor):
cluster, input_descriptors = cluster_and_descriptor
imgMatch = cluster[0]
print 'Matching ', os.path.dirname(imgMatch), '...'
cluster_dir = os.path.dirname(imgMatch)
cluster_files = glob.glob(os.path.join(cluster_dir, '*.png'))
cluster_files = [f for f in cluster_files if not f.endswith('template.png')]
cluster_descriptors = match.load_images_descriptors(cluster_files, os.path.join(cluster_dir, 'images.db'))
return match.find_match(input_descriptors, cluster_descriptors)
def find_card_in_clusters(img, resize_factor=0.25): alls = time() start = time() print 'Loading clusters descriptors...' template_descriptors = load_clusters_template_descriptors() print 'Loaded clusters in:', time() - start, 'seconds' start = time() print 'Loading input image descriptors...' input_descriptors = match.descriptors_for_input_image(img, resize_factor) print 'Loaded input image in:', time() - start, 'seconds' print 'Searching for top clusters...' start = time() cluster_matches = match.find_match(input_descriptors, template_descriptors, top=4) print 'Found top clusters in:', time() - start, 'seconds' start = time() print 'Searching for image in top clusters...' matches = find_matches_in_clusters(input_descriptors, cluster_matches, num_of_processors=4) print 'Found match in:', time() - start, 'seconds with score:', matches[0][1] print 'Total time:', time() - alls, 'seconds' return matches[0]
def __init__(self, no):
QtWidgets.QWidget.__init__(self)
self.setAutoFillBackground(True)
p = self.palette()
p.setColor(self.backgroundRole(), QtGui.QColor(102, 185, 51))
self.setPalette(p)
matches = match.find_match(profiles, no - 1)
self.setWindowTitle('We found you a match!')
self.resize(350, 580)
layout = QtWidgets.QGridLayout()
heading = QtGui.QFont()
heading.setBold(True)
heading.setPointSize(20)
detailsHead = QtWidgets.QLabel("We found you a match!")
detailsHead.setAlignment(QtCore.Qt.AlignCenter)
detailsHead.setFont(heading)
layout.addWidget(detailsHead)
for i in range(1, no):
cur = matches[i]
frame = QtWidgets.QFrame()
frame.setFrameShape(0x3)
frame.setFrameShadow(0x30)
frame.setMaximumSize(350, 150)
frame.setAutoFillBackground(True)
p = frame.palette()
p.setColor(frame.backgroundRole(), QtCore.Qt.white)
frame.setPalette(p)
frameLayout = QtWidgets.QGridLayout()
bold = QtGui.QFont()
bold.setBold(True)
label = QtWidgets.QLabel('Matched With: \n')
label.setAlignment(QtCore.Qt.AlignCenter)
label.setFont(bold)
matchLabel = QtWidgets.QLabel('Name: ' + str(cur[0]) + '\nAge: ' +
str(cur[1]) + '\nLanguage: ' +
str(cur[2]))
matchLabel.setAlignment(QtCore.Qt.AlignCenter)
frameLayout.addWidget(label)
frameLayout.addWidget(matchLabel)
frame.setLayout(frameLayout)
layout.addWidget(frame)
frame = QtWidgets.QFrame()
frame.setMaximumSize(350, 70)
frameLayout = QtWidgets.QHBoxLayout()
frame.setLayout(frameLayout)
buttonReject = QtWidgets.QPushButton('Reject')
buttonReject.clicked.connect(self.close)
buttonChat = QtWidgets.QPushButton('Open Chat')
buttonChat.clicked.connect(self.close)
frameLayout.addWidget(buttonReject)
frameLayout.addWidget(buttonChat)
layout.addWidget(frame)
self.setLayout(layout)
def example():
# get face count of the group
face_count = get_face_count(group_similarity_src)
# select from database
print("\n\n\n############ Training Session ############")
datas = []
for cnn_build_time_range in cnn_build_time_range_set:
start_time = cnn_build_time_range[0]
end_time = cnn_build_time_range[1]
this_datas = get_data(start_time, end_time)
datas = datas + this_datas
results.clear()
# print import data result for banchmark
print("\n\n\n------ Original Result ------")
find_rate(datas, 'person_first_id')
# build automaton
root_c = build_model(datas)
# apply decision model
matched_face_table, unmatched_face_table, unmatched_validate_table, unmatched_face_objects = find_match(
root_c.next, face_count)
# convert data to vector for neuron network
unmatched_face_table, unmatched_validate_table, unmatched_face_objects = convert_to_vector(
matched_face_table, unmatched_face_table, unmatched_validate_table,
unmatched_face_objects)
# separate training and test data
division = int(unmatched_face_table.shape[0] * 0.7)
X = unmatched_face_table[:division]
y = unmatched_validate_table[:division]
X_test = unmatched_face_table[division:]
y_test = unmatched_validate_table[division:]
# build cnn, and save the model
model = conv_net(X, y, X_test, y_test)
# use the model to make prediction on the construct session
print("\n\n\n------ Unmatched Face (Corrected by CNN) ------")
cnn_result = make_prediction(model, unmatched_face_table,
unmatched_face_objects)
print("\n\n\n------ Overall Result (Corrected by CNN) ------")
all_result = list(results) + list(cnn_result)
find_rate(all_result, 'result_id')
print(predict_time_range_set)
for predict_time_range in predict_time_range_set:
print(predict_time_range)
# get data from different session for testing
print("\n\n\n############ Indipendent Testing Session ############")
start_time = predict_time_range[0]
end_time = predict_time_range[1]
datas = get_data(start_time, end_time)
results.clear()
# print import data result for banchmark
print("\n\n\n------ Original Result ------")
find_rate(datas, 'person_first_id')
# build automaton
root_c = build_model(datas)
# apply decision model
matched_face_table, unmatched_face_table, unmatched_validate_table, unmatched_face_objects = find_match(
root_c.next, face_count)
# convert data to vector for neuron network
unmatched_face_table, unmatched_validate_table, unmatched_face_objects = convert_to_vector(
matched_face_table, unmatched_face_table, unmatched_validate_table,
unmatched_face_objects)
# evaluate the model using data of the indipendent test session
score = model.evaluate(unmatched_face_table, unmatched_validate_table)
print("\nCross Session Accuracy: ", score[-1])
# use the model to make prediction on the indipendent test session
print("\n\n\n------ Unmatched Face ------")
find_rate(unmatched_face_objects, 'person_first_id')
print("\n\n\n------ Unmatched Face (Corrected by CNN) ------")
cnn_result = make_prediction(model, unmatched_face_table,
unmatched_face_objects)
print("\n\n\n------ Overall Result (Corrected by CNN) ------")
all_result = list(results) + list(cnn_result)
find_rate(all_result, 'result_id')
