OpenCV + python to achieve face recognition
hair-like
In layman's terms, it can be used as a face feature.
The Haar eigenvalue reflects the grayscale change of the image. For example, some features of the face can be simply described by rectangular features, such as: the eyes are darker than the cheeks, the sides of the bridge of the nose are darker than the bridge of the nose, the mouth is darker than the surrounding area, etc.
opencv api
To use opencv, you must first know what it can do and how to do it. So the importance of API will be reflected. As far as this example is concerned, there are very few functions used, that is, ordinary image reading, grayscale conversion, display image, and simple image editing.
as follows:
read image
Just give the path of the image to be manipulated.
import cv2
image = cv2.imread(imagepath)
Grayscale conversion
The effect of grayscale conversion is that the computational intensity of the image converted to grayscale is reduced.
import cv2
gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
draw
One embodiment of the power of opencv is that it can edit and process images arbitrarily.
The last parameter of the function below specifies the size of the brush.
import cv2
cv2.rectangle(image,(x,y),(x+w,y+w),(0,255,0),2)
display image
Edited images are either displayed directly or saved to a physical storage medium.
import cv2
cv2.imshow("Image Title",image)
Get face recognition training data
It seems complicated, but it is actually some description of the face features, so that after reading the data, opencv can perceive the features on the read pictures according to the sample data in the training, and then perform face recognition on the pictures.
import cv2
face_cascade = cv2.CascadeClassifier(r'./haarcascade_frontalface_default.xml')
The xml file shown here is the universally trained data shared by opencv on GitHub. We can use it directly.
Training data reference address:
https://github.com/opencv/opencv/tree/master/data/haarcascades
face detection
To put it bluntly, it is the process of recognizing new images based on the training data.
import cv2
# Detect faces in pictures
faces = face_cascade.detectMultiScale(
gray,
scaleFactor = 1.15,
minNeighbors = 5,
minSize = (5,5),
flags = cv2.cv.CV_HAAR_SCALE_IMAGE
)
We can arbitrarily specify the values of the parameters inside to achieve recognition with different precisions. The return value is the embodiment of the detection result of opencv on the image.
Process the results of face detection
After the face detection just now, we can get the return value for further processing. But this is not to say how complicated it will be, it is nothing more than adding some eigenvalues.
import cv2
print "Found {0} faces!".format(len(faces))
for(x,y,w,h) in faces:
cv2.rectangle(image,(x,y),(x+w,y+w),(0,255,0),2)
example
With the foundation just now, we can complete a simple small example of face recognition.
Picture material
The image below will serve as our detection basis.
face detection code
import cv2
import numpy as np
import sys,os,glob,numpy
from skimage import io
#Specify the face recognition of the picture and then store
img = cv2.imread("test.jpg")
color = (0, 255, 0)
grey = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
classfier = cv2.CascadeClassifier("C:\\Users\\22291_000\\Anaconda3\\Lib\\site-packages\\cv2\\data\\haarcascade_frontalface_alt2.xml")
faceRects = classfier.detectMultiScale(grey, scaleFactor=1.2, minNeighbors=3, minSize=(32, 32))
if len(faceRects) > 0: # If it is greater than 0, a face is detected
for faceRect in faceRects: # Frame each A face
x, y, w, h = faceRect
cv2.rectangle(img, (x - 10, y - 10), (x + w + 10, y + h + 10), color, 3) #5 control Thickness of green box
# write image
cv2.imwrite('output.jpg',img)
cv2.imshow("Find Faces!",img)
cv2.waitKey(0)
face detection result
Output picture:
