11# Face Recognition with OpenCV built in classifier
2- # Aim: This script recognizes the face in an image using OpenCV's built-in recognizer. The vast application of this project is
3- # recognizing the images in the running videos.
4- # Feature: Easy and understandable code
2+ # Aim: This script recognizes the face in an image using OpenCV's built-in recognizer. The vast application of this project is
3+ # recognizing the images in the running videos.
4+ # Feature: Easy and understandable code
55
66import os
77import cv2 as cv
88import numpy as np
99
1010people = []
11- DIR = r'images' # DIR - path of the folder containing images
12- for i in os .listdir (DIR ): # Will append the names of the person in the people list
11+ DIR = r'images' # DIR - path of the folder containing images
12+ for i in os .listdir (DIR ): # Will append the names of the person in the people list
1313 people .append (i )
1414
1515haar_cascade = cv .CascadeClassifier ('haar_face.xml' )
1616
17- # For every face in the list the corresponding labels to the features are designated.
18- features = []
17+ # For every face in the list the corresponding labels to the features are designated.
18+ features = []
1919labels = []
2020
21+ 2122def train_face ():
2223 for person in people :
23- path = os .path .join (DIR ,person )
24+ path = os .path .join (DIR ,person )
2425 label = people .index (person )
2526
2627 for img in os .listdir (path ):
27- img_path = os .path .join (path ,img )
28+ img_path = os .path .join (path ,img )
2829
2930 img_arr = cv .imread (img_path )
30- gray = cv .cvtColor (img_arr ,cv .COLOR_BGR2GRAY )
31+ gray = cv .cvtColor (img_arr ,cv .COLOR_BGR2GRAY )
3132
32- face_rect = haar_cascade .detectMultiScale (gray , scaleFactor = 1.1 , minNeighbors = 4 )
33+ face_rect = haar_cascade .detectMultiScale (
34+ gray , scaleFactor = 1.1 , minNeighbors = 4 )
3335
34- for (x ,y , w , h ) in face_rect :
35- face_roi = gray [y :y + h , x :x + w ]
36+ for (x ,y , w , h ) in face_rect :
37+ face_roi = gray [y :y + h , x :x + w ]
3638 features .append (face_roi )
3739 labels .append (label )
3840
41+ 3942train_face ()
4043
4144# Feature and labels numpy array
42- features = np .array (features ,dtype = 'object' )
45+ features = np .array (features ,dtype = 'object' )
4346labels = np .array (labels )
4447
4548# Train our recognizer
4649face_recognizer = cv .face .LBPHFaceRecognizer_create ()
4750
4851# Trained on features and the labels list
49- face_recognizer .train (features ,labels )
52+ face_recognizer .train (features ,labels )
5053
51- # Just like haar cascade yml file, we can save this file in yml format so that
54+ # Just like haar cascade yml file, we can save this file in yml format so that
5255# we can easily access this trained model
5356face_recognizer .save ('face_trained.yml' )
5457
55- #### save the features and labels into the numpy file
58+ #save the features and labels into the numpy file
5659# simply remove the comments of the below 2 line
5760#
58- # np.save('features.npy', features)
59- # np.save('labels.npy',labels)
61+ # np.save('features.npy', features)
62+ # np.save('labels.npy',labels)
0 commit comments