使用dlib中的深度残差网络(ResNet)实现实时人脸识别

import dlib
import numpy as np
import cv2
import os
import json

detector = dlib.cnn_face_detection_model_v1('mmod_human_face_detector.dat')
sp = dlib.shape_predictor('shape_predictor_68_face_landmarks.dat')
facerec = dlib.face_recognition_model_v1('dlib_face_recognition_resnet_model_v1.dat')

imagePath = 'LocalImage/' #图像的目录
data = np.zeros((1,128)) #定义一个128维的空向量data
label = [] #定义空的list存放人脸的标签for file in os.listdir(imagePath): #开始一张一张索引目录中的图像
if '.jpg' in file or '.png' in file:
fileName = file
labelName = file.split('_')[0] #获取标签名
print('current image: ', file)
print('current label: ', labelName) img = cv2.imread(imagePath + file) #使用opencv读取图像数据
if img.shape[0]*img.shape[1] > 500000: #如果图太大的话需要压缩，这里像素的阈值可以自己设置
img = cv2.resize(img, (0,0), fx=0.5, fy=0.5)
dets = detector(img, 1) #使用检测算子检测人脸，返回的是所有的检测到的人脸区域
for k, d in enumerate(dets):
rec = dlib.rectangle(d.rect.left(),d.rect.top(),d.rect.right(),d.rect.bottom())
shape = sp(img, rec) #获取landmark
face_descriptor = facerec.compute_face_descriptor(img, shape) #使用resNet获取128维的人脸特征向量
faceArray = np.array(face_descriptor).reshape((1, 128)) #转换成numpy中的数据结构
data = np.concatenate((data, faceArray)) #拼接到事先准备好的data当中去
label.append(labelName) #保存标签
cv2.rectangle(img, (rec.left(), rec.top()), (rec.right(), rec.bottom()), (0, 255, 0), 2) #显示人脸区域
cv2.waitKey(2)
cv2.imshow('image', img)data = data[1:, :] #因为data的第一行是空的128维向量，所以实际存储的时候从第二行开始
np.savetxt('faceData.txt', data, fmt='%f') #保存人脸特征向量合成的矩阵到本地labelFile=open('label.txt','w')
json.dump(label, labelFile) #使用json保存list到本地
labelFile.close()cv2.destroyAllWindows() #关闭所有的窗口

detector = dlib.get_frontal_face_detector()

import dlib
import numpy as np
import cv2
import json

detector = dlib.cnn_face_detection_model_v1('mmod_human_face_detector.dat')
sp = dlib.shape_predictor('shape_predictor_68_face_landmarks.dat')
facerec = dlib.face_recognition_model_v1('dlib_face_recognition_resnet_model_v1.dat')
threshold = 0.54

def findNearestClassForImage(face_descriptor, faceLabel):
temp = face_descriptor - data
e = np.linalg.norm(temp,axis=1,keepdims=True)
min_distance = e.min()
print('distance: ', min_distance)
if min_distance > threshold:
return 'other'
index = np.argmin(e)
return faceLabel[index]

def recognition(img):
dets = detector(img, 1)
for k, d in enumerate(dets): print("Detection {}: Left: {} Top: {} Right: {} Bottom: {}".format(
k, d.rect.left(), d.rect.top(), d.rect.right(), d.rect.bottom()))
rec = dlib.rectangle(d.rect.left(),d.rect.top(),d.rect.right(),d.rect.bottom())
print(rec.left(),rec.top(),rec.right(),rec.bottom())
shape = sp(img, rec)
face_descriptor = facerec.compute_face_descriptor(img, shape) class_pre = findNearestClassForImage(face_descriptor, label)
print(class_pre)
cv2.rectangle(img, (rec.left(), rec.top()+10), (rec.right(), rec.bottom()), (0, 255, 0), 2)
cv2.putText(img, class_pre , (rec.left(),rec.top()), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0,255,0), 2, cv2.LINE_AA) cv2.imshow('image', img)

labelFile=open('label.txt','r')
label = json.load(labelFile) #载入本地人脸库的标签
labelFile.close()data = np.loadtxt('faceData.txt',dtype=float) #载入本地人脸特征向量cap = cv2.VideoCapture(0)
fps = 10
size = (640,480)
fourcc = cv2.VideoWriter_fourcc(*'XVID')
videoWriter = cv2.VideoWriter('video.MP4', fourcc, fps, size)while(1):
ret, frame = cap.read()
#frame = cv2.resize(frame, (0,0), fx=0.5, fy=0.5)
recognition(frame)
videoWriter.write(frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
breakcap.release()
videoWriter.release()
cv2.destroyAllWindows()