Face Recognition in Python and sending message through Whatsapp once face is succesfully recognized
5 min readJun 24, 2021
Dear Readers,
This article includes the code and explaination for training model for face recognition, face detection and sending message through whatsapp once face is recognized.
Collection of Samples
import cv2
import numpy as npface_classifier = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
# Load haar classifier
# Load functions
def face_extractor(img):
# Function detects faces and returns the cropped face
# If no face detected, it returns the input image
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
# We convert our color image to grayscale format for faster calculation
faces = face_classifier.detectMultiScale(gray)
# Detecting face
if faces is ():
#Empty round brackets represent empty tuple
return None
#If there are no faces detected return None i.e. resetting it back to original posititon here
for (x,y,w,h) in faces:
# Cropping all possible captured faces
cropped_face = img[ x:x+w , y:y+h]
# We store the complete image inside cropped_face array
# .detectMultiScale() returns x,y,w,h where (x,y) are top-left edge of a rectangle , w is width , h is height
# faces[0][0] = x ; faces[0][1] = y ; faces[0][2] = w ; faces[0][3] = h
# x1 = x ; y1 = y ; x2 = x1 + w (as x1 + width = x2) ; y2 = y1 + h (as y1 + height = y2)
print(type(cropped_face))
return cropped_face
# returning cropped_face to img in face_extractor function cap = cv2.VideoCapture(0)
# start webcam
count = 0
# Collect 100 samples of your face from webcam input
while True:
ret, frame = cap.read()
if face_extractor(frame) is not None:
# If there is cropped image ready and not empty, then true and get inside loop
count += 1
# Count increases by 1 if image captured
face = cv2.resize(face_extractor(frame), (200, 200))
# Image is resized into size (200,200)
face = cv2.cvtColor(face, cv2.COLOR_BGR2GRAY)
# Converting image into grayscale
file_name_path = './faces/user/' + str(count) + '.jpg'
# Save file in specified directory with unique name
cv2.imwrite(file_name_path, face)
# Image is written on that path with name mentioned
# Put count on images and display live count
cv2.putText(face, str(count), (50, 50), cv2.FONT_HERSHEY_COMPLEX, 1, (0,255,0), 2)
cv2.imshow('Face Cropper', face)
# Show the window
else:
print("Face not found")
pass if cv2.waitKey(1) == 13 or count == 100:
# 13 is the Enter Key
break
cap.release()
# Close the camera
cv2.destroyAllWindows()
print("Collecting Samples Complete")
print(frame.shape)
Training Model
import cv2
import numpy as np
from os import listdir
from os.path import isfile, join
# Get the training data we previously made
data_path = './faces/user/' #our main folder
onlyfiles = [f for f in listdir(data_path) if isfile(join(data_path, f))]# listdir(datapath) is used to list directories/files in the datapath folder/directory.
# f for f in listdir(datapath) means listing all labels in that directory.
# isfile returns either true or false
# join(datapath,f) joins the file in the datapath.
# overall this line means that onlyfiles will store files that are present in the directory mentioned# Create arrays for training data and labels
Training_Data, Labels = [], []# Open training images in our datapath
# Create a numpy array for training data
for i, files in enumerate(onlyfiles):
# Doing transactions one by one on <onlyfiles>
image_path = data_path + onlyfiles[i]
# Storing Data Path combined with the image
images = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
# Reading the image at the path
Training_Data.append(np.asarray(images, dtype=np.uint8))
# Appending the training data and converting input to array(asarray)
Labels.append(i)
# Appending Labels array similarly
# Create a numpy array for both training data and labels
Labels = np.asarray(Labels, dtype=np.int32)
# Converting labels array into numpy format
# Initialize facial recognizer
# model = cv2.face.createLBPHFaceRecognizer()
# NOTE: For OpenCV 3.0 use cv2.face.createLBPHFaceRecognizer()
# pip install opencv-contrib-python
# model = cv2.createLBPHFaceRecognizer()
facemodel = cv2.face_LBPHFaceRecognizer.create()
# Creating the model
facemodel.train(np.asarray(Training_Data), np.asarray(Labels))
# Training the model with the Training Data and corresponding label
print("Model trained sucessefully")
# Facemodel.save('face_recognition_model.h5')
# If you want to save the model in your directory
Face Recognition
import cv2
import numpy as np
import osface_classifier = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')def face_detector(img, size=0.5):
# Convert image to grayscale
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
faces = face_classifier.detectMultiScale(gray, 1.3, 5)
if faces is ():
# if faces returns no value, return the same image screen available
return img, []
# return image and empty list since we have no values of x,y,w,h2
# creating a rectangle on the border of the face
for (x,y,w,h) in faces:
cv2.rectangle(img,(x,y),(x+w,y+h),(0,255,255),2)
roi = img[y:y+h, x:x+w]
roi = cv2.resize(roi, (200, 200))
return img, roi
# return image and the cropped photograph
# Open Webcam
cap = cv2.VideoCapture(0)
while True:
ret, frame = cap.read()
image, face = face_detector(frame)
# function called
try:
face = cv2.cvtColor(face, cv2.COLOR_BGR2GRAY)
# Pass face to prediction model
# "results" comprises of a tuple containing the label and the confidence value
results = facemodel.predict(face)
# prediction using the cropped image
print(results)
# harry_model.predict(face)
confidence = int( 100 * (1 - (results[1])/400) ) # formula to find confidence score
display_string = str(confidence) + '% Confident it is User' # variable declared to store confidence percentage
print(confidence)
cv2.putText(image, display_string, (100, 120), cv2.FONT_HERSHEY_COMPLEX, 1, (255,120,150), 2)
if confidence > 80:
cv2.putText(image, "Hey Yash", (250, 450), cv2.FONT_HERSHEY_COMPLEX, 1, (0,255,0), 2)
cv2.imshow('Face Recognition', image )
break
# os.system("chrome https://www.google.com/)
# os.system("wmplayer c:\<filename>.mp3")
# break
else:
cv2.putText(image, "I dont know, how r u", (250, 450), cv2.FONT_HERSHEY_COMPLEX, 1, (0,0,255), 2)
cv2.imshow('Face Recognition', image )
except:
cv2.putText(image, "No Face Found", (220, 120) , cv2.FONT_HERSHEY_COMPLEX, 1, (0,0,255), 2)
cv2.putText(image, "looking for face", (250, 450), cv2.FONT_HERSHEY_COMPLEX, 1, (0,0,255), 2)
cv2.imshow('Face Recognition', image )
pass
if cv2.waitKey(1) == 13:
# 13 is the Enter Key
break
cap.release()
cv2.destroyAllWindows()
Whatsapp Module
# Here, I have used twilio for sending message through Whatsapp.from twilio.rest import Client
account_sid = 'Account SID'
auth_token = 'Authentication Token'
client = Client(account_sid, auth_token)
message = client.messages.create(
from_='whatsapp:+1234567890',
body='Message from Python',
to='whatsapp:+1234567890'
)
print(message.sid)
OR
pywhatkit.sendwhatmsg_instantly(phone_no="RECEIVER’s NUMBER", message="Hi, I am Yash this is whatsapp message sent with Python.")
print("Whatsapp Message sent Successfully!!")Mailing Module
pywhatkit.send_mail(email_sender= "<email id>",
password= "<password>",
subject="Automated E-mail",
message="System generated mail using OpenCV face detection.",
email_receiver="<sender email id>")AWS Module
# Creating instance in AWS
os.system("aws ec2 run-instances --image-id ami-011c99152163a87ae --instance-type t2.micro --subnet-id subnet-06f7036d --count 1 --security-group-ids sg-5771be2b > ec2.txt")
print("Instance Launched")
# Creating volume of size 5gb
os.system("aws ec2 create-volume --availability-zone ap-south-1a --size 5 --volume-type gp2 --tag-specification ResourceType=volume,Tags=[{Key=face,Value=volume}] > ebs.txt")
print("Volume Created of size 5 gb")
print("Initiating in 120 seconds")
time.sleep(120)
ec2_id = open("ec2.txt", 'r').read().split(',')[3].split(':')[1].split('"')[1]
ebs_id = open("ebs.txt", 'r').read().split(',')[6].split(':')[1].split('"')[1]
os.system("aws ec2 attach-volume --instance-id " + ec2_id +" --volume-id " + ebs_id +" --device /dev/sdf")
print("Volume Successfully attached to the instance")