def train_recognizers(base64_path, models_path, sample_size=10):
if not os.path.exists(models_path):
os.mkdir(models_path)
# faces = paths.list_files(base64_path, validExts=('.txt',))
# loop over the input faces for training
for (i, path) in enumerate(glob.glob(base64_path + "/*.txt")):
fr = FaceRecognizer(cv2.createLBPHFaceRecognizer(radius=1, neighbors=8, grid_x=8, grid_y=8))
# labels = []
# extract the person from the file name,
name = path[path.rfind("/") + 1:].replace(".txt", "")
print("[INFO] training on '{}'".format(name))
# load the faces file, sample it, and initialize the list of faces
sample = open(path).read().strip().split("\n")
sample = random.sample(sample, min(len(sample), sample_size))
faces = []
# loop over the faces in the sample
for face in sample:
# decode the face and update the list of faces
faces.append(encodings.base64_decode_image(face))
# train the face detector on the faces and update the list of labels
fr.train(faces, np.array([i] * len(faces)))
# labels = name
# update the face recognizer to include the face name labels, then write the model to file
fr.setLabels((name,))
model_path = os.path.join(models_path, name + '.model')
fr.save(model_path)
def handle_input(line, sep="\t"):
# split the line into the imageID, path, and image, then decode the image
(imageID, path, image) = line.strip().split(sep)
image = encodings.base64_decode_image(image)
# return a tuple of the imageID, filename and decoded image
return (imageID, path, image)
def train_recognizers(base64_path, models_path, sample_size=10):
if not os.path.exists(models_path):
os.mkdir(models_path)
# faces = paths.list_files(base64_path, validExts=('.txt',))
# loop over the input faces for training
for (i, path) in enumerate(glob.glob(base64_path + "/*.txt")):
fr = FaceRecognizer(
cv2.createLBPHFaceRecognizer(radius=1,
neighbors=8,
grid_x=8,
grid_y=8))
# labels = []
# extract the person from the file name,
name = path[path.rfind("/") + 1:].replace(".txt", "")
print("[INFO] training on '{}'".format(name))
# load the faces file, sample it, and initialize the list of faces
sample = open(path).read().strip().split("\n")
sample = random.sample(sample, min(len(sample), sample_size))
faces = []
# loop over the faces in the sample
for face in sample:
# decode the face and update the list of faces
faces.append(encodings.base64_decode_image(face))
# train the face detector on the faces and update the list of labels
fr.train(faces, np.array([i] * len(faces)))
# labels = name
# update the face recognizer to include the face name labels, then write the model to file
fr.setLabels((name, ))
model_path = os.path.join(models_path, name + '.model')
fr.save(model_path)
def saveUser(self, arguments):
beginRecognizing = FaceAuthentication(
cv2.createLBPHFaceRecognizer(radius=1,
neighbors=8,
grid_x=8,
grid_y=8))
userName = []
print('Completed training for users:')
for faceDBfile, path in enumerate(
glob.glob("authrizedUser/usersFaces/" + "/*.txt")):
names = path[path.rfind("/") + 1:].replace(".txt", "")
print("{0} account".format(names))
userAccount = open(path).read().strip().split("\n")
userAccount = random.sample(userAccount,
min(len(userAccount), 100))
faces = []
for face in userAccount:
faces.append(encodings.base64_decode_image(face))
beginRecognizing.newOrUpdateTrain(
faces, np.array([faceDBfile] * len(faces)))
userName.append(names)
beginRecognizing.defindNames(userName)
beginRecognizing.saveInfo("authrizedUser/hushFaces/")
# author: Adrian Rosebrock
# website: http://www.pyimagesearch.com
# USAGE
# python encode_image.py --image ../demo_images/pyimagesearch_logo.jpg
# import the necessary packages
from __future__ import print_function
from imutils import encodings
import argparse
import cv2
# construct the argument parser and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True, help="path to input image")
args = ap.parse_args()
# load the input image
image = cv2.imread(args.image)
cv2.imshow("Original", image)
# encode the image and dump the whole mess to the console
encoded = encodings.base64_encode_image(image)
print(encoded)
# decode the image
decoded = encodings.base64_decode_image(encoded)
print("Original image shape: {}".format(image.shape))
print("Decoded image shape: {}".format(decoded.shape))
cv2.imshow("Decoded", decoded)
cv2.waitKey(0)
# website: http://www.pyimagesearch.com
# USAGE
# python encode_image.py --image ../demo_images/pyimagesearch_logo.jpg
# import the necessary packages
from __future__ import print_function
from imutils import encodings
import argparse
import cv2
# construct the argument parser and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True, help="path to input image")
args = ap.parse_args()
# load the input image
image = cv2.imread(args.image)
cv2.imshow("Original", image)
# encode the image and dump the whole mess to the console
encoded = encodings.base64_encode_image(image)
print(encoded)
# decode the image
decoded = encodings.base64_decode_image(encoded)
print("Original image shape: {}".format(image.shape))
print("Decoded image shape: {}".format(decoded.shape))
cv2.imshow("Decoded", decoded)
cv2.waitKey(0)
else:
fr = FaceRecognizer(
cv2.face.LBPHFaceRecognizer_create(radius=1,
neighbors=8,
grid_x=8,
grid_y=8))
# initialize the list of labels
labels = []
for (i, path) in enumerate(glob.glob(args["selfies"] + "/*.txt")):
# extract the person from the file name,
name = path[path.rfind("/") + 1:].replace(".txt", "")
print("[INFO] training on '{}'".format(name))
# load the faces file, sample it, and initialize the list of faces
sample = open(path).read().strip().split("\n")
sample = random.sample(sample, min(len(sample), args["sample_size"]))
faces = []
# loop over the faces in the sample
for face in sample:
# decode the face and update the list of faces
faces.append(encodings.base64_decode_image(face))
# train the face detector on the faces and update the list of labels
fr.train(faces, np.array([i] * len(faces)))
labels.append(name)
# update the face recognizer to include the face name labels, then write the model to file
fr.setLabels(labels)
fr.save(args["classifier"])
def gather_selfies(employee_id):
BasePath = Path(__file__).resolve(strict=True).parent
#print(path)
c_path = os.path.join('cascades',
'haarcascade_frontalface_default.xml') #newp)
cascade_file = os.path.join(
BasePath,
c_path) #'cascades/haarcascade_frontalface_default.xml') #newp)
#punit fd = FaceDetector(args["face_cascade"])
fd = FaceDetector(cascade_file) #newp
captureMode = False
color = (0, 255, 0)
# grab a reference to the webcam and open the output file for writing
camera = cv2.VideoCapture(0)
#punit f = open(args["output"], args["write_mode"])
o_path = os.path.join('output', 'faces') # newp
output_path = os.path.join(BasePath, o_path) #'output/faces') #newp
#output_path = os.path.join(BasePath,'output/faces') # newp
f = open(output_path + '/' + str(employee_id) + '.txt', "a+") #newp
total = 0
# loop over the frames of the video
while True:
# grab the current frame
(grabbed, frame) = camera.read()
# if the frame could not be grabbed, then we have reached the end of the video
if not grabbed:
print('reached end')
break
# resize the frame, convert the frame to grayscale, and detect faces in the frame
frame = imutils.resize(frame, width=500)
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faceRects = fd.detect(gray,
scaleFactor=1.1,
minNeighbors=9,
minSize=(100, 100))
# ensure that at least one face was detected
if len(faceRects) > 0:
# sort the bounding boxes, keeping only the largest one
(x, y, w, h) = max(faceRects, key=lambda b: (b[2] * b[3]))
# if we are in capture mode, extract the face ROI, encode it, and write it to file
if captureMode:
face = gray[y:y + h, x:x + w].copy(order="C")
f.write("{}\n".format(encodings.base64_encode_image(face)))
total += 1
# draw bounding box on the frame
cv2.rectangle(frame, (x, y), (x + w, y + h), color, 2)
# show the frame and record if the user presses a key
cv2.imshow("Frame", frame)
key = cv2.waitKey(1) & 0xFF
# if the `c` key is pressed, then go into capture mode
if key == ord("c"):
# if we are not already in capture mode, drop into capture mode
if not captureMode:
captureMode = True
color = (0, 0, 255)
# otherwise, back out of capture mode
else:
captureMode = False
color = (0, 255, 0)
# if the `q` key is pressed, break from the loop
elif key == ord("q"):
break
# close the output file, cleanup the camera, and close any open windows
print("[INFO] wrote {} frames to file".format(total))
f.close()
camera.release()
cv2.destroyAllWindows()
#output_path = os.path.join(BasePath, 'output\\faces') # newp
#output_path = 'output/faces' # newp
c_path = os.path.join('output', 'classifier') # newp
classifier_path = os.path.join(BasePath,
c_path) #'output/classifier') # newp
# initialize the face recognizer and the list of labels
if imutils.is_cv2():
fr = FaceRecognizer(
cv2.createLBPHFaceRecognizer(radius=1,
neighbors=8,
grid_x=8,
grid_y=8))
else:
fr = FaceRecognizer(
cv2.face.LBPHFaceRecognizer_create(radius=1,
neighbors=8,
grid_x=8,
grid_y=8))
# initialize the list of labels
labels = []
# loop over the input faces for training
for (i, path) in enumerate(glob.glob(output_path + "/*.txt")):
# extract the person from the file name,
name = path[path.rfind("/") + 1:].replace(".txt", "")
print("[INFO] training on '{}'".format(name))
# load the faces file, sample it, and initialize the list of faces
sample = open(path).read().strip().split("\n")
sample = random.sample(sample, min(len(sample), 200))
faces = []
# loop over the faces in the sample
for face in sample:
# decode the face and update the list of faces
faces.append(encodings.base64_decode_image(face))
# train the face detector on the faces and update the list of labels
fr.train(faces, np.array([i] * len(faces)))
labels.append(name)
# update the face recognizer to include the face name labels, then write the model to file
fr.setLabels(labels)
fr.save(classifier_path)
#file=os.path.isfile(output_path+'//'+str(employee_id)+'.txt')
#return file
#gather_selfies('221629')