def train():
from face_train import train
classifier = train("knn_examples/train",
model_save_path="trained_knn_model.clf",
n_neighbors=11)
return render_template('Train.html')
def EditInfo(self):
global luru
global mkpath
global number
name = self.ui.lineEdit.text()
number = self.ui.lineEdit_2.text()
cla = self.ui.lineEdit_3.text()
sex = self.ui.lineEdit_4.text()
age = self.ui.lineEdit_5.text()
if number:
mkpath = "./data/" + number
mkdir(mkpath)
if find_data(number):
# tim = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
#GUI嵌入摄像头
# self.openCamera()
# luru = 1
#重新打开摄像头窗口
reply = QMessageBox.question(self.ui, '信息已存在', '确认重新采集人脸？',
QMessageBox.Yes | QMessageBox.No,
QMessageBox.No)
if reply == QMessageBox.Yes:
face_save.generate(mkpath, number)
face_train.train(mkpath, number)
else:
QMessageBox.question(self.ui, '提示', '可直接签到',
QMessageBox.Yes)
else:
if name and number and cla and sex and age:
# tim = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
add_data(name, number, cla, sex, age)
#GUI嵌入摄像头
# self.openCamera()
# luru = 1
#重新打开摄像头窗口
face_save.generate(mkpath, number)
face_train.train(mkpath, number)
self.ui.label_8.setText(str(panduan.peple()))
else:
QMessageBox.question(self.ui, '提示', '请先输入信息',
QMessageBox.Yes)
else:
QMessageBox.question(self.ui, '提示', '未确认到信息输入', QMessageBox.Yes)
def start(name):
face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
a = 0
while True:
s, img = vid.read()
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
for (x, y, w, h) in faces:
cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
roi_gray = gray[y:y + h, x:x + w]
cv2.imwrite("images/" + name + "/" + str(a) + ".png", roi_gray)
a = a + 1
cv2.imshow('frame', img)
if cv2.waitKey(100) == 97 or a > 200:
break
face_train.train()
vid.release()
cv2.destroyAllWindows()
def clicked1(self):
face_train.train()
get1 = face_train.train()
self.label_3.setText("Status: " + get1)
self.update()
singular_image.append(list()) for j in range(60): if cur_line[j] != ' ': singular_image[i].append(1) else: singular_image[i].append(0); probability_per_class = list() for i in range(2): probability_per_class.append(get_prob(i, singular_image, digit_matrices, 1)) max_value = max(probability_per_class) max_index = probability_per_class.index(max_value) digit_class = int(line) global num_per_class global correct_count global correct_per_class global confusion_matrix confusion_matrix[digit_class][max_index] += 1 num_per_class[digit_class] +=1 if max_index == digit_class: correct_count+=1 correct_per_class[max_index]+=1 # print (max_index, digit_class) if __name__ == '__main__': test(rtrain.train())
def show_camera(self):
# global qflag 保存图片
# global Id
# mkpath = "./data/test"
global sampleNum
global luru
global mkpath
global number
global facecheck
# Id = self.ui.lineEdit_2.text()
# if Id:
# luru = 1
# mkpath = "./data/" + Id
# mkdir(mkpath)
# self.clear_text()
detector = cv2.CascadeClassifier(
'haarcascade/haarcascade_frontalface_default.xml')
flag, self.imagefan = self.ui.cap.read()
self.image = cv2.flip(self.imagefan, 1)
# print(luru)
if luru:
faces = detector.detectMultiScale(self.image, 1.3, 5)
for (x, y, w, h) in faces:
cv2.rectangle(self.image, (x, y), (x + w, y + h), (255, 0, 0),
2)
gray_than = (h) / (w)
f = cv2.resize(self.image[y:y + h, x:x + w],
(200, int(200 * gray_than)))
# incrementing sample number
sampleNum = sampleNum + 1
# print(mkpath + "/User." + str(number) +'.' + str(sampleNum) + ".jpg")
# saving the captured face in the dataset folder
cv2.imwrite(mkpath + "/User." + str(number) + '.' +
str(sampleNum) + ".jpg", f) #
cv2.putText(self.image, str((x, y)), (x, y),
cv2.FONT_HERSHEY_SIMPLEX, 1, 255, 2)
cv2.putText(self.image, str(sampleNum), (x, y - 20),
cv2.FONT_HERSHEY_SIMPLEX, 1, 255, 2)
show = cv2.resize(self.image, (800, 600))
show = cv2.cvtColor(show, cv2.COLOR_BGR2RGB)
showImage = QImage(show.data, show.shape[1], show.shape[0],
QImage.Format_RGB888)
self.ui.label_6.setPixmap(QPixmap.fromImage(showImage))
if sampleNum >= 30:
self.closeCamera()
face_train.train(mkpath, number)
luru = 0
# cv2.imshow('frame', self.image)
# wait for 1 miliseconds
# self.image = self.imagefan
# elif facecheck == 1:
# recognizer = cv2.face.LBPHFaceRecognizer_create()
# # recognizer = cv2.createLBPHFaceRecognizer() # in OpenCV 2
# recognizer.read('trainner/trainner.xml')
# # recognizer.load('trainner/trainner.yml') # in OpenCV 2
# cascade_path = "haarcascade/haarcascade_frontalface_default.xml"
# face_cascade = cv2.CascadeClassifier(cascade_path)
# font = cv2.FONT_HERSHEY_SIMPLEX
# j = 1
# t = 3
# time_go=time.time()
# if j==1:
# time_start = time_go
# j = j - 1
# time_tal = int(time_go-time_start)
# gray = cv2.cvtColor(self.image, cv2.COLOR_BGR2GRAY)
# faces = face_cascade.detectMultiScale(gray, 1.3, 5)
# for (x, y, w, h) in faces:
# cv2.rectangle(self.image, (x - 50, y - 50), (x + w + 50, y + h + 50), (225, 0, 0), 2)
# my_id, conf = recognizer.predict(gray[y:y + h, x:x + w])
# print(my_id, conf)
# if conf > 0 and conf <50:
# # facedata = sqlTest.find_data(str(my_id)[1])
# img_id = "master" + str(my_id)
# t=1
# else:
# img_id = "Unknown"
# t=0
# # cv2.cv.PutText(cv2.cv.fromarray(im), str(Id), (x, y + h), font, 255)
# # print ("Label: %s, Confidence: %.2f" % (img_id, 100-conf))
# cv2.putText(self.image, str(img_id), (x, y - 20), font, 1, 255, 2)
# show = cv2.resize(self.image,(800,600))
# show = cv2.cvtColor(show, cv2.COLOR_BGR2RGB)
# showImage = QImage(show.data, show.shape[1],show.shape[0],QImage.Format_RGB888)
# self.ui.label_6.setPixmap(QPixmap.fromImage(showImage))
# if time_tal >= 300:
# t = 0
# self.closeCamera()
# facecheck = 0
# if t == 1:
# self.closeCamera()
# # dat1 = [[str(facedata[0]),str(facedata[1]),str(facedata[2]),str(facedata[3]),str(facedata[4]),time.strftime("%H:%M:%S",time.localtime())]]
# # common.append_csv("record.csv",dat1)
# print('hello')
else:
show = cv2.resize(self.image, (800, 600))
show = cv2.cvtColor(show, cv2.COLOR_BGR2RGB)
showImage = QImage(show.data, show.shape[1], show.shape[0],
QImage.Format_RGB888)
self.ui.label_6.setPixmap(QPixmap.fromImage(showImage))
singular_image.append(list())
for j in range(60):
if cur_line[j] != ' ':
singular_image[i].append(1)
else:
singular_image[i].append(0)
probability_per_class = list()
for i in range(2):
probability_per_class.append(
get_prob(i, singular_image, digit_matrices, 1))
max_value = max(probability_per_class)
max_index = probability_per_class.index(max_value)
digit_class = int(line)
global num_per_class
global correct_count
global correct_per_class
global confusion_matrix
confusion_matrix[digit_class][max_index] += 1
num_per_class[digit_class] += 1
if max_index == digit_class:
correct_count += 1
correct_per_class[max_index] += 1
# print (max_index, digit_class)
if __name__ == '__main__':
test(rtrain.train())
def menu():
ip = ([
ip for ip in socket.gethostbyname_ex(socket.gethostname())[2]
if not ip.startswith("127.")
][0])
ip_ask = input("this programme needs to use the ip of this computer y/n")
if (ip_ask == "y"):
print("we got the ip" + ip)
with open("setting.txt", "w") as fic:
fic.write("ip:" + str(ip) + "/")
if (ip_ask == "n"):
print(
"sorry but this version needs the ip, there his no sollution for this version"
)
microphone_ask = input(
"do you use a microphone compatible with the libary pyaudio |yes|no|don't know"
)
microphone_ask_find = (microphone_ask.find("know"))
if (microphone_ask_find is not -1):
FORMAT = pyaudio.paInt16
CHANNELS = 1
RATE = 44100
p = pyaudio.PyAudio()
stream = p.open(format=FORMAT,
channels=CHANNELS,
rate=RATE,
input=True,
output=True,
frames_per_buffer=CHUNK)
data = stream.read(CHUNK)
data_int = np.array(struct.unpack(str(2 * CHUNK) + 'B', data))
print(data_int)
password = input("what password do you want to setup?")
with open("setting.txt", "a") as fic:
fic.write("pas==" + password + "/")
with open("setting.txt", "r") as fic:
set_ = fic.read()
data2 = ("setting set" + str(set_))
socket_client.socket_client(data2, ip)
opencv = input("do you have a camera:")
if (opencv == "yes"):
import cv2
from PIL import Image
import numpy as np
import pickle
with open("setting.txt", "a") as fic:
fic.write("cam=yes")
print("ok we will take some picture.")
print("the admin pleas go infront of the camera alone")
found = (0)
while found == 0:
print(found)
print("found")
face_cascade = cv2.CascadeClassifier(
'cascade/data/haarcascade_frontalface_alt.xml')
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray,
scaleFactor=1.5,
minNeighbors=5)
for (x, y, w, h) in faces:
roi_gray = gray[y:y + h, x:x + w]
img_item = ("image/user1/ima10.png")
cv2.imwrite(img_item, roi_gray)
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
img_item = ("image/user1/ima" + "11.png")
cv2.imwrite(img_item, roi_gray)
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
img_item = ("image/user1/ima" + "11.png")
cv2.imwrite(img_item, roi_gray)
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
img_item = ("image/user1/ima12.png")
cap = cv2.VideoCapture(0)
cv2.imwrite(img_item, roi_gray)
ret, frame = cap.read()
img_item = ("image/user1/ima13.png")
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
cv2.imwrite(img_item, roi_gray)
img_item = ("image/user1/ima14.png")
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
cv2.imwrite(img_item, roi_gray)
img_item = ("image/user1/ima15.png")
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
cv2.imwrite(img_item, roi_gray)
found = (1)
d = input("pleas put another face")
found = (0)
while found == 0:
print(found)
print("found")
face_cascade = cv2.CascadeClassifier(
'cascade/data/haarcascade_frontalface_alt.xml')
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray,
scaleFactor=1.5,
minNeighbors=5)
for (x, y, w, h) in faces:
roi_gray = gray[y:y + h, x:x + w]
img_item = ("image/user2/ima10.png")
cv2.imwrite(img_item, roi_gray)
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
img_item = ("image/user2/ima" + "11.png")
cv2.imwrite(img_item, roi_gray)
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
img_item = ("image/user2/ima" + "11.png")
cv2.imwrite(img_item, roi_gray)
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
img_item = ("image/user2/ima12.png")
cap = cv2.VideoCapture(0)
cv2.imwrite(img_item, roi_gray)
ret, frame = cap.read()
img_item = ("image/user2/ima13.png")
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
cv2.imwrite(img_item, roi_gray)
img_item = ("image/user2/ima14.png")
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
cv2.imwrite(img_item, roi_gray)
img_item = ("image/user2/ima15.png")
cap = cv2.VideoCapture(0)
ret, frame = cap.read()
cv2.imwrite(img_item, roi_gray)
number_img = (0)
face_train.train()
print("we recomande you to restart the program")
found = 1
import numpy as np
import cv2
import pickle
import time
from tkinter import *
import tkMessageBox
import os
from face_train import train
# running the face train algorithm
train()
face_cascade = cv2.CascadeClassifier(
'cascades/data/haarcascades/haarcascade_frontalface_alt2.xml')
eye_cascade = cv2.CascadeClassifier(
'cascades/data/haarcascades/haarcascade_eye.xml')
recognizer = cv2.face.LBPHFaceRecognizer_create()
recognizer.read("trainner.yml")
labels = {"person_name": 1}
with open("labels.pickle", 'rb') as f:
# dump the label ids to a file
og_labels = pickle.load(f)
labels = {v: k for k, v in og_labels.items()}
cap = cv2.VideoCapture(0)
match = 0
mismatch = 0
timeout = time.time() + 10