face_recognition.py

import numpy as np
import cv2
import os

def distance(v1, v2):
    # Eucledian
    return np.sqrt(((v1-v2)**2).sum())

def knn(train, test, k=5):
    dist = []

    for i in range(train.shape[0]):
        #Get the vector and label
        ix = train[i, :-1]
        iy = train[i, -1]
        # Computhe the distance from test point
        d = distance(test, ix)
        dist.append([d, iy])
    # Sort based on distance and get top k
    dk = sorted(dist, key=lambda x: x[0])[:k]
    # Retrieve only the labels
    labels = np.array(dk)[:, -1]

    # Get frequencies of each label
    output = np.unique(labels, return_counts=True)
    # Find max frequency and corresponding label
    index = np.argmax(output[1])
    return output[0][index]

cap = cv2.VideoCapture(0)
face_cascade = cv2.CascadeClassifier("haarcascade_frontalface_alt.xml") ## Taking the instructions from opencv

dataset_path = "./face_dataset/" ## Path to get the face data

face_data = []
labels = []
class_id = 0 ## Labels for every given file
names = {} ## Mapping between if and name


# Dataset preparation
for fx in os.listdir(dataset_path): ## Loop on the faces data folder.
    if fx.endswith('.npy'):
        names[class_id] = fx[:-4]
        data_item = np.load(dataset_path + fx)
        face_data.append(data_item)

        target = class_id * np.ones((data_item.shape[0],))
        class_id += 1
        labels.append(target)

face_dataset = np.concatenate(face_data, axis=0)
face_labels = np.concatenate(labels, axis=0).reshape((-1, 1))
print(face_labels.shape) 
print(face_dataset.shape)

trainset = np.concatenate((face_dataset, face_labels), axis=1)
print(trainset.shape)

font = cv2.FONT_HERSHEY_SIMPLEX

while True:
    ret, frame = cap.read()
    if ret == False:
        continue

    # Convert frame to grayscale
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    # Detect multi faces in the image
    faces = face_cascade.detectMultiScale(gray, 1.3, 5)

    for face in faces:
        x, y, w, h = face

        # Get the face RDI
        offset = 5
        face_section = frame[y-offset:y+h+offset, x-offset:x+w+offset]
        face_section = cv2.resize(face_section, (100, 100))

        out = knn(trainset, face_section.flatten())

        # Draw rectangle in hte original image
        cv2.putText(frame, names[int(out)],(x,y-10), cv2.FONT_HERSHEY_SIMPLEX, 1,(255,0,0),2,cv2.LINE_AA)
        cv2.rectangle(frame, (x,y), (x+w,y+h), (255,255,255), 2)
    
    cv2.imshow("Faces", frame)

    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows() ## Close the camera.