# !/usr/bin/python3.6
# -*- coding: utf-8 -*-
# @author breeze
import time
import cv2
import face_recognition
import win32com.client
import encoding_images
import pandas as pd
import threading
# This is a demo of running face recognition on live video from your webcam. It's a little more complicated than the
# other example, but it includes some basic performance tweaks to make things run a lot faster:
# 1. Process each video frame at 1/4 resolution (though still display it at full resolution)
# 2. Only detect faces in every other frame of video.
# PLEASE NOTE: This example requires OpenCV (the `cv2` library) to be installed only to read from your webcam.
# OpenCV is *not* required to use the face_recognition library. It's only required if you want to run this
# specific demo. If you have trouble installing it, try any of the other demos that don't require it instead.
# Load a sample picture and learn how to recognize it.
# face_recognition.api.batch_face_locations(images, number_of_times_to_upsample=1, batch_size=128)[source]
# face_recognition.api.compare_faces(known_face_encodings, face_encoding_to_check, tolerance=0.6)
# face_recognition.api.face_distance(face_encodings, face_to_compare)[source]
# face_recognition.api.face_encodings(face_image, known_face_locations=None, num_jitters=1)[source]
# face_recognition.api.face_landmarks(face_image, face_locations=None)[source]
# face_recognition.api.face_locations(img, number_of_times_to_upsample=1, model='hog')[source]
# face_recognition.api.load_image_file(file, mode='RGB')[source]
# Get a reference to webcam #0 (the default one)
video_capture = cv2.VideoCapture(0)
# 语音模块 voice model
speaker = win32com.client.Dispatch("SAPI.SpVoice")
name = "Unknown"
current_names = [name]
last_time = time.time()
known_face_names = []
known_face_encodings = []
known_face_encodings, known_face_names = encoding_images.load_encodings()
# Initialize some variables
face_locations = []
face_encodings = []
face_names = []
process_this_frame = True #
TIME_DIFF = 20 # 持久化的时间间隔,当设置为 0 时候,每次识别的结果直接进行保存.
name_record = "./dataset/face_record.txt" # 持久化识别出的人脸结果
NAME_DF = pd.DataFrame(known_face_names, columns=["name"])
import datetime
import time
last_ts = time.time()
def myprint(log, ts=time.time()):
global last_ts
diff = ts - last_ts
print(log, '--------', diff)
last_ts = ts
def process_face_records(name):
"""
处理每一条识别的记录 ,并在一定时间之后将数据持久化到文件中
:param name:
:return:
"""
print('process_face_records start', time.time())
global current_names, last_time
# myprint("global current_names {}, last_time {}".format(current_names, last_time))
# 判断是不是在识别的列表中,不在的话就进行问候
if name not in current_names:
print("ts ====", last_time, time.time())
current_names.append(name)
myprint("Hello {}, nice to meet you! ".format(name))
# speaker.Speak("Hello {}, nice to meet you! ".format(name))
# 在一定时间内,清空已经识别的人, 并进行
if last_time < time.time() - TIME_DIFF: # 每隔一段时间清空一下检测到的人
last_time = time.time()
time_format = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
myprint(time_format + " update last_time and clear current_names.")
with open(name_record, 'a') as f:
if len(current_names) > 0:
f.writelines("{}:{} \n".format(time_format, str(current_names)))
print("======", current_names)
current_names = [] # clear()
current_names = [name]
myprint('process_face_records end', time.time())
def vote_class(face_encoding, tolerance=0.3, topN=5):
myprint('vote start ', time.time())
"""
当比较的结果小于tolerance的时候,有多个值,采用取topN 进行投票 ,决定最终的分类,此处没有对 distance 距离进行加权
:param face_encoding: face encoding
:param tolerance: 距离的阈值,越小越相似
:param topN: 参与投票的最大数量
:return: detect name
"""
# 计算出距离
distance_ = face_recognition.face_distance(known_face_encodings, face_encoding)
df = pd.DataFrame(distance_, columns=["dis"]) # 转换成 DataFrame
topDF = df[df['dis'] <= tolerance].nsmallest(topN, columns=['dis']) # 过滤结果集
namedf = NAME_DF.loc[topDF.index] # 从姓名列表中获取face距离对应的人脸名称
con = pd.concat([topDF, namedf], axis=1) # concat name and distance
# print('con', con)
group = con.groupby(["name"])['dis'].sum()
gp = group.reset_index()
print('vote -- ', gp)
if len(gp) == 0:
print("------unknown -----")
return "Unknown", 10
import numpy as np # TODO optimize
arr = np.array(gp)
name1 = arr[0, 0]
dis1 = arr[0, 1]
print("get top one:", name1, dis1)
myprint('vote end', time.time())
return name1, dis1
def face_process():
myprint("face process start",time.time())
# Find all the faces and face encodings in the current frame of video
# face_locations = face_recognition.face_locations(rgb_small_frame, model="cnn")
myprint('face_locations start', time.time())
face_locations = face_recognition.face_locations(rgb_small_frame, model="hog")
myprint('face_locations end', time.time())
myprint('face_encodings start', time.time())
face_encodings = face_recognition.face_encodings(rgb_small_frame, face_locations)
myprint('face_encodings end', time.time())
face_names = []
for face_encoding in face_encodings:
# optimize start 采用KNN 排名*权重, 在类别上进行叠加,然后排序取出top1
name, dis = vote_class(face_encoding)
# optimize end 采用 排名*权重, 在类别上进行叠加,然后排序取出top1
face_names.append(name) # 将人脸数据
# Display the results
for (top, right, bottom, left), name in zip(face_locations, face_names):
# Scale back up face locations since the frame we detected in was scaled to 1/4 size
top *= 4
right *= 4
bottom *= 4
left *= 4
myprint('putText start', time.time())
# Draw a box around the face
cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 255), 2)
# Draw a label with a name below the face
cv2.rectangle(frame, (left, bottom - 35), (right, bottom), (0, 0, 255), cv2.FILLED)
font = cv2.FONT_HERSHEY_DUPLEX
cv2.putText(frame, name, (left + 6, bottom - 6), font, 1.0, (255, 255, 255), 1)
myprint("putText end " + name, time.time())
# say hello and save record to file
myprint('process_face_records start', time.time())
process_face_records(name)
myprint('process_face_records end', time.time())
# Display the resulting image
cv2.imshow('Video', frame)
myprint("face process end", time.time())
while video_capture.isOpened():
# Grab a single frame of video
ret, frame = video_capture.read()
# Resize frame of video to 1/4 size for faster face recognition processing
small_frame = cv2.resize(frame, (0, 0), fx=0.25, fy=0.25)
# Convert the image from BGR color (which OpenCV uses) to RGB color (which face_recognition uses)
rgb_small_frame = small_frame[:, :, ::-1]
# Only process every other frame of video to save time
if process_this_frame:
face_process()
# t = threading.Thread(target=face_process, name='face_process') # 线程对象.
# t.start() # 启动.
process_this_frame = not process_this_frame
# Hit 'q' on the keyboard to quit!
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# Release handle to the webcam
video_capture.release()
cv2.destroyAllWindows()
