github/deepface
1
1
Fork 0
mirror of https://github.com/serengil/deepface.git synced 2025-06-06 11:35:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #1186 from serengil/feat-task-1304-creating-more-utils

Browse Source 
creating image utils
This commit is contained in:
Sefik Ilkin Serengil 2024-04-13 08:27:41 +01:00 committed by GitHub
commit 52e7796b7d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
10 changed files with 166 additions and 162 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
deepface/DeepFace.py
View File

@ -10,7 +10,6 @@ os.environ["TF_USE_LEGACY_KERAS"] = "1"
# pylint: disable=wrong-import-position
# 3rd party dependencies
import cv2
import numpy as np
import pandas as pd
import tensorflow as tf
@ -26,6 +25,7 @@ from deepface.modules import (
demography,
detection,
streaming,
preprocessing,
)
from deepface import __version__
@ -548,5 +548,5 @@ def detectFace(
extracted_face = None
if len(face_objs) > 0:
extracted_face = face_objs[0]["face"]
extracted_face = cv2.resize(extracted_face, target_size)
extracted_face = preprocessing.resize_image(img=extracted_face, target_size=target_size)
return extracted_face

149
deepface/commons/image_utils.py Normal file
View File

@ -0,0 +1,149 @@
# built-in dependencies
import os
import io
from typing import List, Union, Tuple
import hashlib
import base64
from pathlib import Path
# 3rd party dependencies
import requests
import numpy as np
import cv2
from PIL import Image
def list_images(path: str) -> List[str]:
"""
List images in a given path
Args:
path (str): path's location
Returns:
images (list): list of exact image paths
"""
images = []
for r, _, f in os.walk(path):
for file in f:
exact_path = os.path.join(r, file)
_, ext = os.path.splitext(exact_path)
ext_lower = ext.lower()
if ext_lower not in {".jpg", ".jpeg", ".png"}:
continue
with Image.open(exact_path) as img: # lazy
if img.format.lower() in ["jpeg", "png"]:
images.append(exact_path)
return images
def find_image_hash(file_path: str) -> str:
"""
Find the hash of given image file with its properties
finding the hash of image content is costly operation
Args:
file_path (str): exact image path
Returns:
hash (str): digest with sha1 algorithm
"""
file_stats = os.stat(file_path)
# some properties
file_size = file_stats.st_size
creation_time = file_stats.st_ctime
modification_time = file_stats.st_mtime
properties = f"{file_size}-{creation_time}-{modification_time}"
hasher = hashlib.sha1()
hasher.update(properties.encode("utf-8"))
return hasher.hexdigest()
def load_image(img: Union[str, np.ndarray]) -> Tuple[np.ndarray, str]:
"""
Load image from path, url, base64 or numpy array.
Args:
img: a path, url, base64 or numpy array.
Returns:
image (numpy array): the loaded image in BGR format
image name (str): image name itself
"""
# The image is already a numpy array
if isinstance(img, np.ndarray):
return img, "numpy array"
if isinstance(img, Path):
img = str(img)
if not isinstance(img, str):
raise ValueError(f"img must be numpy array or str but it is {type(img)}")
# The image is a base64 string
if img.startswith("data:image/"):
return load_image_from_base64(img), "base64 encoded string"
# The image is a url
if img.lower().startswith("http://") or img.lower().startswith("https://"):
return load_image_from_web(url=img), img
# The image is a path
if os.path.isfile(img) is not True:
raise ValueError(f"Confirm that {img} exists")
# image must be a file on the system then
# image name must have english characters
if img.isascii() is False:
raise ValueError(f"Input image must not have non-english characters - {img}")
img_obj_bgr = cv2.imread(img)
# img_obj_rgb = cv2.cvtColor(img_obj_bgr, cv2.COLOR_BGR2RGB)
return img_obj_bgr, img
def load_image_from_base64(uri: str) -> np.ndarray:
"""
Load image from base64 string.
Args:
uri: a base64 string.
Returns:
numpy array: the loaded image.
"""
encoded_data_parts = uri.split(",")
if len(encoded_data_parts) < 2:
raise ValueError("format error in base64 encoded string")
encoded_data = encoded_data_parts[1]
decoded_bytes = base64.b64decode(encoded_data)
# similar to find functionality, we are just considering these extensions
# content type is safer option than file extension
with Image.open(io.BytesIO(decoded_bytes)) as img:
file_type = img.format.lower()
if file_type not in ["jpeg", "png"]:
raise ValueError(f"input image can be jpg or png, but it is {file_type}")
nparr = np.fromstring(decoded_bytes, np.uint8)
img_bgr = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
# img_rgb = cv2.cvtColor(img_bgr, cv2.COLOR_BGR2RGB)
return img_bgr
def load_image_from_web(url: str) -> np.ndarray:
"""
Loading an image from web
Args:
url: link for the image
Returns:
img (np.ndarray): equivalent to pre-loaded image from opencv (BGR format)
"""
response = requests.get(url, stream=True, timeout=60)
response.raise_for_status()
image_array = np.asarray(bytearray(response.raw.read()), dtype=np.uint8)
img = cv2.imdecode(image_array, cv2.IMREAD_COLOR)
return img

27
deepface/commons/package_utils.py
View File

@ -1,7 +1,3 @@
# built-in dependencies
import os
import hashlib
# 3rd party dependencies
import tensorflow as tf
@ -29,29 +25,6 @@ def get_tf_minor_version() -> int:
return int(tf.__version__.split(".", maxsplit=-1)[1])
def find_hash_of_file(file_path: str) -> str:
"""
Find the hash of given image file with its properties
finding the hash of image content is costly operation
Args:
file_path (str): exact image path
Returns:
hash (str): digest with sha1 algorithm
"""
file_stats = os.stat(file_path)
# some properties
file_size = file_stats.st_size
creation_time = file_stats.st_ctime
modification_time = file_stats.st_mtime
properties = f"{file_size}-{creation_time}-{modification_time}"
hasher = hashlib.sha1()
hasher.update(properties.encode("utf-8"))
return hasher.hexdigest()
def validate_for_keras3():
tf_major = get_tf_major_version()
tf_minor = get_tf_minor_version()

4
deepface/modules/detection.py
View File

@ -7,9 +7,9 @@ import cv2
from PIL import Image
# project dependencies
from deepface.modules import preprocessing
from deepface.models.Detector import DetectedFace, FacialAreaRegion
from deepface.detectors import DetectorWrapper
from deepface.commons import image_utils
from deepface.commons import logger as log
logger = log.get_singletonish_logger()
@ -63,7 +63,7 @@ def extract_faces(
resp_objs = []
# img might be path, base64 or numpy array. Convert it to numpy whatever it is.
img, img_name = preprocessing.load_image(img_path)
img, img_name = image_utils.load_image(img_path)
if img is None:
raise ValueError(f"Exception while loading {img_name}")

97
deepface/modules/preprocessing.py
View File

@ -1,15 +1,9 @@
# built-in dependencies
import os
from typing import Union, Tuple
import base64
from pathlib import Path
import io
from typing import Tuple
# 3rd party
import numpy as np
import cv2
import requests
from PIL import Image
# project dependencies
from deepface.commons import package_utils
@ -22,95 +16,6 @@ elif tf_major_version == 2:
from tensorflow.keras.preprocessing import image
def load_image(img: Union[str, np.ndarray]) -> Tuple[np.ndarray, str]:
"""
Load image from path, url, base64 or numpy array.
Args:
img: a path, url, base64 or numpy array.
Returns:
image (numpy array): the loaded image in BGR format
image name (str): image name itself
"""
# The image is already a numpy array
if isinstance(img, np.ndarray):
return img, "numpy array"
if isinstance(img, Path):
img = str(img)
if not isinstance(img, str):
raise ValueError(f"img must be numpy array or str but it is {type(img)}")
# The image is a base64 string
if img.startswith("data:image/"):
return load_base64(img), "base64 encoded string"
# The image is a url
if img.lower().startswith("http://") or img.lower().startswith("https://"):
return load_image_from_web(url=img), img
# The image is a path
if os.path.isfile(img) is not True:
raise ValueError(f"Confirm that {img} exists")
# image must be a file on the system then
# image name must have english characters
if img.isascii() is False:
raise ValueError(f"Input image must not have non-english characters - {img}")
img_obj_bgr = cv2.imread(img)
# img_obj_rgb = cv2.cvtColor(img_obj_bgr, cv2.COLOR_BGR2RGB)
return img_obj_bgr, img
def load_image_from_web(url: str) -> np.ndarray:
"""
Loading an image from web
Args:
url: link for the image
Returns:
img (np.ndarray): equivalent to pre-loaded image from opencv (BGR format)
"""
response = requests.get(url, stream=True, timeout=60)
response.raise_for_status()
image_array = np.asarray(bytearray(response.raw.read()), dtype=np.uint8)
img = cv2.imdecode(image_array, cv2.IMREAD_COLOR)
return img
def load_base64(uri: str) -> np.ndarray:
"""Load image from base64 string.
Args:
uri: a base64 string.
Returns:
numpy array: the loaded image.
"""
encoded_data_parts = uri.split(",")
if len(encoded_data_parts) < 2:
raise ValueError("format error in base64 encoded string")
encoded_data = encoded_data_parts[1]
decoded_bytes = base64.b64decode(encoded_data)
# similar to find functionality, we are just considering these extensions
# content type is safer option than file extension
with Image.open(io.BytesIO(decoded_bytes)) as img:
file_type = img.format.lower()
if file_type not in ["jpeg", "png"]:
raise ValueError(f"input image can be jpg or png, but it is {file_type}")
nparr = np.fromstring(decoded_bytes, np.uint8)
img_bgr = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
# img_rgb = cv2.cvtColor(img_bgr, cv2.COLOR_BGR2RGB)
return img_bgr
def normalize_input(img: np.ndarray, normalization: str = "base") -> np.ndarray:
"""Normalize input image.

34
deepface/modules/recognition.py
View File

@ -8,10 +8,9 @@ import time
import numpy as np
import pandas as pd
from tqdm import tqdm
from PIL import Image
# project dependencies
from deepface.commons import package_utils
from deepface.commons import image_utils
from deepface.modules import representation, detection, verification
from deepface.commons import logger as log
@ -144,7 +143,7 @@ def find(
pickled_images = [representation["identity"] for representation in representations]
# Get the list of images on storage
storage_images = __list_images(path=db_path)
storage_images = image_utils.list_images(path=db_path)
if len(storage_images) == 0:
raise ValueError(f"No item found in {db_path}")
@ -161,7 +160,7 @@ def find(
if identity in old_images:
continue
alpha_hash = current_representation["hash"]
beta_hash = package_utils.find_hash_of_file(identity)
beta_hash = image_utils.find_image_hash(identity)
if alpha_hash != beta_hash:
logger.debug(f"Even though {identity} represented before, it's replaced later.")
replaced_images.append(identity)
@ -292,31 +291,6 @@ def find(
return resp_obj
def __list_images(path: str) -> List[str]:
"""
List images in a given path
Args:
path (str): path's location
Returns:
images (list): list of exact image paths
"""
images = []
for r, _, f in os.walk(path):
for file in f:
exact_path = os.path.join(r, file)
_, ext = os.path.splitext(exact_path)
ext_lower = ext.lower()
if ext_lower not in {".jpg", ".jpeg", ".png"}:
continue
with Image.open(exact_path) as img: # lazy
if img.format.lower() in ["jpeg", "png"]:
images.append(exact_path)
return images
def __find_bulk_embeddings(
employees: List[str],
model_name: str = "VGG-Face",
@ -360,7 +334,7 @@ def __find_bulk_embeddings(
desc="Finding representations",
disable=silent,
):
file_hash = package_utils.find_hash_of_file(employee)
file_hash = image_utils.find_image_hash(employee)
try:
img_objs = detection.extract_faces(

3
deepface/modules/representation.py
View File

@ -5,6 +5,7 @@ from typing import Any, Dict, List, Union
import numpy as np
# project dependencies
from deepface.commons import image_utils
from deepface.modules import modeling, detection, preprocessing
from deepface.models.FacialRecognition import FacialRecognition
@ -74,7 +75,7 @@ def represent(
)
else: # skip
# Try load. If load error, will raise exception internal
img, _ = preprocessing.load_image(img_path)
img, _ = image_utils.load_image(img_path)
if len(img.shape) != 3:
raise ValueError(f"Input img must be 3 dimensional but it is {img.shape}")

1
requirements.txt
View File

@ -1,3 +1,4 @@
requests>=2.27.1
numpy>=1.14.0
pandas>=0.23.4
gdown>=3.10.1

5
tests/test_extract_faces.py
View File

@ -6,6 +6,7 @@ import pytest
# project dependencies
from deepface import DeepFace
from deepface.modules import preprocessing
from deepface.commons import image_utils
from deepface.commons import logger as log
logger = log.get_singletonish_logger()
@ -60,12 +61,12 @@ def test_file_types_while_loading_base64():
img1_base64 = image_to_base64(image_path=img1_path)
with pytest.raises(ValueError, match="input image can be jpg or png, but it is"):
_ = preprocessing.load_base64(uri=img1_base64)
_ = image_utils.load_image_from_base64(uri=img1_base64)
img2_path = "dataset/img1.jpg"
img2_base64 = image_to_base64(image_path=img2_path)
img2 = preprocessing.load_base64(uri=img2_base64)
img2 = image_utils.load_image_from_base64(uri=img2_base64)
# 3 dimensional image should be loaded
assert len(img2.shape) == 3

4
tests/test_find.py
View File

@ -8,7 +8,7 @@ import pandas as pd
# project dependencies
from deepface import DeepFace
from deepface.modules import verification
from deepface.modules import recognition
from deepface.commons import image_utils
from deepface.commons import logger as log
logger = log.get_singletonish_logger()
@ -95,7 +95,7 @@ def test_filetype_for_find():
def test_filetype_for_find_bulk_embeddings():
imgs = recognition.__list_images("dataset")
imgs = image_utils.list_images("dataset")
assert len(imgs) > 0