Skip to content
Related Articles

Related Articles

ML | Implement Face recognition using k-NN with scikit-learn
  • Difficulty Level : Medium
  • Last Updated : 15 Mar, 2019

k-Nearest Neighbors:

k-NN is one of the most basic classification algorithms in machine learning. It belongs to the supervised learning category of machine learning. k-NN is often used in search applications where you are looking for “similar” items. The way we measure similarity is by creating a vector representation of the items, and then compare the vectors using an appropriate distance metric (like the Euclidean distance, for example).

It is generally used in data mining, pattern recognition, recommender systems and intrusion detection.

Libraries used are:


Dataset used:
We used haarcascade_frontalface_default.xml dataset which is easily available online and also you can download it from this link.

scikit-learn provides a range of supervised and unsupervised learning algorithms via a consistent interface in Python.
This library is built upon SciPy that must be installed on your devices in order to use scikit_learn.

Face-Recognition :
This includes three Python files where the first one is used to detect the face and storing it in a list format, second one is used to store the data in ‘.csv’ file format and the third one is used recognize the face.

# this file is used to detect face 
# and then store the data of the face
import cv2
import numpy as np
# import the file where data is
# stored in a csv file format
import npwriter
name = input("Enter your name: ")
# this is used to access the web-cam
# in order to capture frames
cap = cv2.VideoCapture(0)
classifier = cv2.CascadeClassifier("../dataset/haarcascade_frontalface_default.xml")
# this is class used to detect the faces as provided
# with a haarcascade_frontalface_default.xml file as data
f_list = []
while True:
    ret, frame =
    # converting the image into gray
    # scale as it is easy for detection
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    # detect multiscale, detects the face and its coordinates
    faces = classifier.detectMultiScale(gray, 1.5, 5)
    # this is used to detect the face which
    # is closest to the web-cam on the first position
    faces = sorted(faces, key = lambda x: x[2]*x[3],
                                     reverse = True)
    # only the first detected face is used
    faces = faces[:1]  
    # len(faces) is the number of
    # faces showing in a frame
    if len(faces) == 1:   
        # this is removing from tuple format      
        face = faces[0]   
        # storing the coordinates of the
        # face in different variables
        x, y, w, h = face 
        # this is will show the face
        # that is being detected     
        im_face = frame[y:y + h, x:x + w] 
        cv2.imshow("face", im_face)
    if not ret:
    cv2.imshow("full", frame)
    key = cv2.waitKey(1)
    # this will break the execution of the program
    # on pressing 'q' and will click the frame on pressing 'c'
    if key & 0xFF == ord('q'):
    elif key & 0xFF == ord('c'):
        if len(faces) == 1:
            gray_face = cv2.cvtColor(im_face, cv2.COLOR_BGR2GRAY)
            gray_face = cv2.resize(gray_face, (100, 100))
            print(len(f_list), type(gray_face), gray_face.shape)
            # this will append the face's coordinates in f_list
            print("face not found")
        # this will store the data for detected
        # face 10 times in order to increase accuracy
        if len(f_list) == 10:
# declared in npwriter
npwriter.write(name, np.array(f_list)) 
cv2.destroyAllWindows() – Create/Update ‘.csv’: file

import pandas as pd
import numpy as np
import os.path
f_name = "face_data.csv"
# storing the data into a csv file
def write(name, data):
    if os.path.isfile(f_name):
        df = pd.read_csv(f_name, index_col = 0)
        latest = pd.DataFrame(data, columns = map(str, range(10000)))
        latest["name"] = name
        df = pd.concat((df, latest), ignore_index = True, sort = False)
        # Providing range only because the data
        # here is already flattened for when
        # it was store in f_list
        df = pd.DataFrame(data, columns = map(str, range(10000)))
        df["name"] = name
    df.to_csv(f_name) – Face-recognizer

# this one is used to recognize the 
# face after training the model with
# our data stored using knn
import cv2
import numpy as np
import pandas as pd
from npwriter import f_name
from sklearn.neighbors import KNeighborsClassifier
# reading the data
data = pd.read_csv(f_name).values
# data partition
X, Y = data[:, 1:-1], data[:, -1]
print(X, Y)
# Knn function calling with k = 5
model = KNeighborsClassifier(n_neighbors = 5)
# fdtraining of model, Y)
cap = cv2.VideoCapture(0)
classifier = cv2.CascadeClassifier("../dataset/haarcascade_frontalface_default.xml")
f_list = []
while True:
    ret, frame =
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    faces = classifier.detectMultiScale(gray, 1.5, 5)
    X_test = []
    # Testing data
    for face in faces:
        x, y, w, h = face
        im_face = gray[y:y + h, x:x + w]
        im_face = cv2.resize(im_face, (100, 100))
    if len(faces)>0:
        response = model.predict(np.array(X_test))
        # prediction of result using knn
        for i, face in enumerate(faces):
            x, y, w, h = face
            # drawing a rectangle on the detected face
            cv2.rectangle(frame, (x, y), (x + w, y + h),
                                         (255, 0, 0), 3)
            # adding detected/predicted name for the face
            cv2.putText(frame, response[i], (x-50, y-50),
                              cv2.FONT_HERSHEY_DUPLEX, 2,
                                         (0, 255, 0), 3)
    cv2.imshow("full", frame)
    key = cv2.waitKey(1)
    if key & 0xFF == ord("q") :



My Personal Notes arrow_drop_up
Recommended Articles
Page :