def createClassifer(id_list: list):
faces, labels = testGatherTrainingData(id_list)
print("Total faces: ", len(faces))
print("Total labels: ", len(labels), labels)
face_recognizer = face.LBPHFaceRecognizer_create()
face_recognizer.train(np.array(faces), np.array(labels))
return face_recognizer
def showVideo():
url = "https://tianfeng-510finalproject.appspot.com/form"
webbrowser.open_new_tab(url)
haar_face_cascade = cv2.CascadeClassifier(
'data/haarcascade_frontalface_alt.xml')
recognizer = face.LBPHFaceRecognizer_create()
recognizer.read("trainner.yml")
lables = {"person_name": 1}
with open('lables.pickle', 'rb') as f:
og_lables = pickle.load(f)
lables = {v: k for k, v in og_lables.items()}
cap = cv2.VideoCapture(0)
cap.set(3, 320)
cap.set(4, 240)
cv2.namedWindow('frame', cv2.WINDOW_NORMAL)
cv2.resizeWindow('frame', (600, 600))
while (True):
ret, frame = cap.read()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = haar_face_cascade.detectMultiScale(gray,
scaleFactor=1.1,
minNeighbors=5)
for (x, y, w, h) in faces:
# print(x , y, w, h)
roi_gray = gray[y:y + h, x:x + w]
roi_color = frame[y:y + h, x:x + w]
id_, conf = recognizer.predict(roi_gray)
if conf >= 45 and conf <= 85:
name = lables[id_]
cv2.putText(frame, name, (x, y), cv2.FONT_HERSHEY_SIMPLEX, 1,
(255, 0, 0), 1, cv2.LINE_AA)
img_item = "face/my-image.png"
cv2.imwrite(img_item, roi_gray)
color = (255, 0, 0)
stroke = 1
cv2.rectangle(frame, (x, y), (x + w, y + h), color, stroke)
new_data = {"Name": str(lables[id_]), "X": int(x), "Y": int(y)}
print(new_data)
db.update(new_data)
text = "Faces found:" + str(len(faces))
cv2.putText(frame, text, (10, 50), cv2.FONT_HERSHEY_SIMPLEX, 1,
(255, 255, 255), 1)
cv2.imshow('frame', frame)
if cv2.waitKey(5) & 0xFF == ord('q'):
break
# When everything is done, release the capture
cap.release()
cv2.destroyAllWindows()
def face_rec():
names = ['BAO BAO ', 'BA BA BA ', 'Bei Bei', 'MMMMMMM'] # 标签对应的名字0,baobao,1，bbb,2，beibei....
[x, y] = read_img("D:") # 调读取函数，返回图像、和标签列表
y = np.asarray(y, dtype=np.int32) # 转为NUMPY的ARRAY
# CV自带的三种算法，现用LBPH算法。此处有坑，坑的我差点放弃，原来叫createLBPHFaceRecognizer，为什么我下载的这模样
# model=fc.EigenFaceRecognizer_create()
# model = fc.FisherFaceRecognizer_create()
model = fc.LBPHFaceRecognizer_create()
# 训练，此处应把训练结果保存，再用到时直接读取结果，效率更高，xml?json?pickle?
model.train(np.asarray(x), np.asarray(y))
# 下面读取摄像头图像，用矩形标识检测到脸部和训练后结果比对，打印出对应标签所对应名字
camera = cv2.VideoCapture(0)
face_cascade = cv2.CascadeClassifier('D:\pythonweb\face_test/haarcascade_frontalface_default.xml')
while True:
read, img = camera.read()
faces = face_cascade.detectMultiScale(img, 1.3, 5)
for (x, y, w, h) in faces:
img = cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
try:
roi = cv2.resize(roi, (200, 200), interpolation=cv2.INTER_LINEAR)
params = model.predict(roi) # predict（）函数做比对，返回一个元祖格式值 （标签，系数）。系数和算法有关，
# 前2种算法值低于5000不可靠，LBPH低于50可靠，80-90不可靠，高于90纯蒙
# 此处有文章可做，通过单位时间内检测到的系数平均值，可以得到更准确结果
print(params)
# 打印标签对应名字，如cvtopil的灰度问题解决，可cvtopil函数替换
cv2.putText(img, names[params[0]], (x, y - 20), cv2.FONT_HERSHEY_SIMPLEX, 1, 255, 2)
except:
continue
cv2.imshow("abc", img)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cv2.destroyAllWindows()
import cv2
import cv2.face as fc
recongnizer = fc.LBPHFaceRecognizer_create()
recongnizer.read('FaceTrainer/trainer.yml')
cascadePath = 'Cascade/haarcascade_frontalface_default.xml'
faceCascades = cv2.CascadeClassifier(cascadePath)
font = cv2.FONT_HERSHEY_SIMPLEX
cam = cv2.VideoCapture(0)
minW = 0.1 * cam.get(3)
minH = 0.1 * cam.get(4)
id = 0
names = ['lhz', '哈哈哈']
while True:
ret, img = cam.read()
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = faceCascades.detectMultiScale(gray,
scaleFactor=1.2,
minNeighbors=5,
minSize=(int(minW), int(minH)))
for (x, y, w, h) in faces:
# 比对数据
id, confidence = recongnizer.predict(gray[y:y + h, x:x + w])
import multiprocessing
import numpy as np
from cv2 import face # OpenCV
# My imports.
import extraction_model as exmodel
from sort_database.utils import EMOTIONS_5, EMOTIONS_8
# Start the script.
script_name = os.path.basename(__file__) # The name of this script
print("\n{}: Beginning face recogniser tests...\n".format(script_name))
start = time.clock() # Start of the speed test. clock() is most accurate.
fisherface = face.FisherFaceRecognizer_create() # Fisherface classifier
eigenface = face.EigenFaceRecognizer_create() # Eigenface classifier
lbph = face.LBPHFaceRecognizer_create() # Local Binary Patterns classifier
def run_fisher_recognizer(X_train, y_train, X_test, y_test):
"""Train the fisherface classifier."""
print("\n***> Training fisherface classifier")
print("Size of the training set is {} images.".format(len(y_train)))
fisherface.train(X_train, np.array(y_train))
print("Predicting classification set.")
cnt = 0
correct = 0
incorrect = 0
for image in X_test:
pred, conf = fisherface.predict(image)
import cv2
from cv2 import face
import pickle
import numpy as np
from PIL import Image
# whereever the file is saved
# looking for the path of it
BASE_DIR = os.path.dirname(os.path.abspath(__file__))
# path of the images
image_dir = os.path.join(BASE_DIR, "images")
face_cascades = cv2.CascadeClassifier(
"cascades/data/haarcascade_frontalface_alt2.xml")
# face recognizer:
recognizer = face.LBPHFaceRecognizer_create()
current_id = 0
label_ids = {}
y_labels = []
x_train = []
for root, dir, files in os.walk(image_dir):
for file in files:
if file.endswith("jpg") or file.endswith("png") or file.endswith(
"jpeg"):
path = os.path.join(root, file)
label = os.path.basename(os.path.dirname(path)).replace(
" ", "-").lower()
#print(label, path)
# add the label into label_ids dic if it's not there already