def clusters():
"""
Отображает AJAX-версию страницы с кластеризацией. Предназначено для замены
display_clusters() после тестирования.
"""
person = request.values.get('filter')
print(person)
Face._index.refresh()
total = Face.search().count()
named = Face.search().filter("exists", field="person").count()
status = "{:.1%} ({} out of {}) faces are named. Clusters count: {}".format(
named / total, named, total, Cluster.search().count())
a = A("terms", field="person.raw", size=10000)
ps = Search()
ps.aggs.bucket("persons", a)
psr = ps.execute()
persons = [b.key for b in psr.aggs['persons']]
if person:
s = Cluster.search().filter("prefix", person=person).sort("-face_count")
results = s[0:10000].execute()
else:
s = Cluster.search().exclude("exists", field="person")
s.query = FunctionScore(query=s.query,
functions=[SF('random_score', weight=100),
SF('field_value_factor',
field="face_count", weight=1)],
score_mode="avg", boost_mode="replace")
results = s[0:50].execute()
return render_template('clusters.html', clusters=results, persons=persons,
status=status)
def extract_faces(file_name, image, output_folder, named):
image_np = np.array(image)
locations = face_recognition.face_locations(image_np)
encodings = face_recognition.face_encodings(image_np, locations)
face_count = 0
for location, features in zip(locations, encodings):
face = Face()
face.file_name = file_name
face.features = base64.b64encode(features).decode()
face.position.top = location[0]
face.position.right = location[1]
face.position.bottom = location[2]
face.position.left = location[3]
if named:
face.person = os.path.splitext(os.path.basename(file_name))[0]
face.save()
face_count += 1
if output_folder:
face_image = image_np[location[0]:location[2],
location[3]:location[1]]
pil_image = Image.fromarray(face_image)
pil_image.save(join(output_folder, face.meta.id + '.jpg'),
"JPEG",
quality=90,
optimize=True,
progressive=True)
return face_count
def shuffle():
print_faces_stats()
faces = [face for face in Face.search().scan()]
clst = approximate_rank_order_clustering(faces)
generate_clusters_index(clst)
clusters_with_person = Cluster.search().filter("exists", field="person").scan()
for cls in clusters_with_person:
cls.update_faces_index()
print_faces_stats()
def generate_images_index():
Photo._index.delete(ignore=404)
Photo.init()
results = Face.search().filter("exists", field="person").exclude('terms', person=['ignored']).scan()
groups = defaultdict(list)
for face in results:
groups[face.file_name].append(face.person)
bulk(Elasticsearch(args.elastic), (Photo(group[0], group[1]).to_dict(True) for group in groups.items()))
def display_faces(person):
Face._index.refresh()
s = Face.search().filter("term", person__raw=person)
status = "{}: {}".format(person, s.count())
results = s[0:1000].execute()
return render_template('faces.html', faces=results, status=status)
def display_image(face_id):
face = Face().get(face_id)
img_src = os.path.splitdrive(face.file_name)[1]
status = ""
return render_template('image.html', img_src=img_src, status=status)
def print_faces_stats():
Face._index.refresh()
total = Face.search().count()
named = Face.search().filter("exists", field="person").count()
print("{:.1%} ({} out of {}) faces are named.".format(named / total, named, total))
"--user",
dest="user",
help="RabbitMQ user name",
metavar="USER",
default='guest')
parser.add_argument("-p",
"--password",
dest="password",
help="RabbitMQ user password",
metavar="PASSWORD",
default='guest')
args = parser.parse_args()
connections.create_connection(hosts=[args.elastic])
Face.init()
print(' [*] Waiting for messages. To exit press CTRL+BREAK')
credentials = PlainCredentials(args.user, args.password)
connection = pika.BlockingConnection(
pika.ConnectionParameters(host=args.host, credentials=credentials))
channel = connection.channel()
channel.queue_declare(queue='img_queue', durable=True)
def callback(ch, method, properties, body):
file_name = body.decode()
try:
image = Image.open(file_name)
faces_extracted = extract_faces(
file_name, image, args.folder,