Python3 利用face_recognition实现人脸识别的方法


Posted in Python onMarch 13, 2020

前言

之前实践了下face++在线人脸识别版本,这回做一下离线版本。github 上面有关于face_recognition的相关资料,本人只是做个搬运工,对其中的一些内容进行搬运,对其中一些例子进行实现。

官方描述:

face_recognition是一个强大、简单、易上手的人脸识别开源项目,并且配备了完整的开发文档和应用案例,特别是兼容树莓派系统。本项目是世界上最简洁的人脸识别库,你可以使用Python和命令行工具提取、识别、操作人脸。本项目的人脸识别是基于业内领先的C++开源库 dlib中的深度学习模型,用Labeled Faces in the Wild人脸数据集进行测试,有高达99.38%的准确率。但对小孩和亚洲人脸的识别准确率尚待提升。

(关于兼容树莓派,以后有板子了再做一下)

下面两个链接划重点

https://github.com/ageitgey/face_recognition/blob/master/README_Simplified_Chinese.md
https://face-recognition.readthedocs.io/en/latest/face_recognition.html

环境配置

  • ubuntu16.04(其他环境的安装可以参考第一个链接,官方有说明)
  • pycharm(可忽略,怎么舒服怎么来)
  • python3
  • opencv(我的是4.1.2,三点几的版本应该也一样)

实际上只需要安装face_recognition,当然,没有opencv的也需要安装一下opencv

pip3 install face_recognition

图片准备

由于需要做一些图片的比对,因此需要准备一些图片,本文图片取自以下链接

https://www.zhihu.com/question/314169580/answer/872770507

接下来开始操作

官方还有提供命令行的操作(这个没去做),本文不做这个,我们只要是要在python中用face_recognition,因此定位到这一块。

Python3 利用face_recognition实现人脸识别的方法

这个api文档地址就是上面的第二个链接。进去之后可以看到:

Python3 利用face_recognition实现人脸识别的方法

part1.识别图片中的人是谁

代码

# part1
# 识别图片中的人是谁
import face_recognition
known_image = face_recognition.load_image_file("lyf1.jpg")
unknown_image = face_recognition.load_image_file("lyf2.jpg")

lyf_encoding = face_recognition.face_encodings(known_image)[0]
unknown_encoding = face_recognition.face_encodings(unknown_image)[0]

results = face_recognition.compare_faces([lyf_encoding], unknown_encoding)	
# A list of True/False values indicating which known_face_encodings match the face encoding to check

print(type(results))
print(results)

if results[0] == True:
  print("yes")
else:
  print("no")

结果

<class 'list'>
[True]
yes

part2.从图片中找到人脸

代码

# part2
# 从图片中找到人脸(定位人脸位置)

import face_recognition
import cv2

image = face_recognition.load_image_file("lyf1.jpg")

face_locations_useCNN = face_recognition.face_locations(image,model='cnn')
# model ? Which face detection model to use. “hog” is less accurate but faster on CPUs.
# “cnn” is a more accurate deep-learning model which is GPU/CUDA accelerated (if available). The default is “hog”.

face_locations_noCNN=face_recognition.face_locations(image)
# A list of tuples of found face locations in css (top, right, bottom, left) order
# 因为返回值的顺序是这样子的,因此在后面的for循环里面赋值要注意按这个顺序来

print("face_location_useCNN:")
print(face_locations_useCNN)
face_num1=len(face_locations_useCNN)
print(face_num1)    # The number of faces


print("face_location_noCNN:")
print(face_locations_noCNN)
face_num2=len(face_locations_noCNN)
print(face_num2)    # The number of faces
# 到这里为止,可以观察两种情况的坐标和人脸数,一般来说,坐标会不一样,但是检测出来的人脸数应该是一样的
# 也就是说face_num1 = face_num2; face_locations_useCNN 和 face_locations_noCNN 不一样


org = cv2.imread("lyf1.jpg")
img = cv2.imread("lyf1.jpg")
cv2.imshow("lyf1.jpg",img) # 原始图片

# Go to get the data and draw the rectangle
# use CNN
for i in range(0,face_num1):
  top = face_locations_useCNN[i][0]
  right = face_locations_useCNN[i][1]
  bottom = face_locations_useCNN[i][2]
  left = face_locations_useCNN[i][3]

  start = (left, top)
  end = (right, bottom)

  color = (0,255,255)
  thickness = 2
  cv2.rectangle(img, start, end, color, thickness)  # opencv 里面画矩形的函数

# Show the result
cv2.imshow("useCNN",img)


# for face_location in face_locations_noCNN:
#
#   # Print the location of each face in this image
#   top, right, bottom, left = face_location
# # 等价于下面的这种写法

for i in range(0,face_num2):
  top = face_locations_noCNN[i][0]
  right = face_locations_noCNN[i][1]
  bottom = face_locations_noCNN[i][2]
  left = face_locations_noCNN[i][3]

  start = (left, top)
  end = (right, bottom)

  color = (0,255,255)
  thickness = 2
  cv2.rectangle(org, start, end, color, thickness)

cv2.imshow("no cnn ",org)

cv2.waitKey(0)
cv2.destroyAllWindows()

结果

face_location_useCNN:
[(223, 470, 427, 266)]
1
face_location_noCNN:
[(242, 489, 464, 266)]
1

图片效果大致是这样

Python3 利用face_recognition实现人脸识别的方法

part3.找到人脸并将其裁剪打印出来(使用cnn定位人脸)

代码

# part3
# 找到人脸并将其裁剪打印出来(使用cnn定位人脸)

from PIL import Image
import face_recognition

# Load the jpg file into a numpy array
image = face_recognition.load_image_file("lyf1.jpg")

face_locations = face_recognition.face_locations(image, number_of_times_to_upsample=0, model="cnn")

print("I found {} face(s) in this photograph.".format(len(face_locations)))

for face_location in face_locations:
  top, right, bottom, left = face_location
  print("A face is located at pixel location Top: {}, Left: {}, Bottom: {}, Right: {}".format(top, left, bottom, right))

  face_image = image[top:bottom, left:right]
  pil_image = Image.fromarray(face_image)
  pil_image.show()

结果

I found 1 face(s) in this photograph.
A face is located at pixel location Top: 205, Left: 276, Bottom: 440, Right: 512

图片效果大致是这样

Python3 利用face_recognition实现人脸识别的方法

part4.识别单张图片中人脸的关键点

代码

# part4 识别单张图片中人脸的关键点

from PIL import Image, ImageDraw
import face_recognition

# Load the jpg file into a numpy array
image = face_recognition.load_image_file("lyf1.jpg")

# Find all facial features in all the faces in the image
face_landmarks_list = face_recognition.face_landmarks(image)
# print(face_landmarks_list)

print("I found {} face(s) in this photograph.".format(len(face_landmarks_list)))

# Create a PIL imagedraw object so we can draw on the picture
pil_image = Image.fromarray(image)
d = ImageDraw.Draw(pil_image)

for face_landmarks in face_landmarks_list:

  # Print the location of each facial feature in this image
  for facial_feature in face_landmarks.keys():
    print("The {} in this face has the following points: {}".format(facial_feature, face_landmarks[facial_feature]))

  # Let's trace out each facial feature in the image with a line!
  for facial_feature in face_landmarks.keys():
    d.line(face_landmarks[facial_feature], width=5)

# Show the picture
pil_image.show()

结果

I found 1 face(s) in this photograph.
The left_eyebrow in this face has the following points: [(305, 285), (321, 276), (340, 277), (360, 281), (377, 288)]
The right_eye in this face has the following points: [(422, 313), (432, 303), (446, 302), (459, 305), (449, 312), (435, 314)]
The nose_bridge in this face has the following points: [(394, 309), (394, 331), (395, 354), (396, 375)]
The right_eyebrow in this face has the following points: [(407, 287), (424, 278), (442, 273), (461, 272), (478, 279)]
The bottom_lip in this face has the following points: [(429, 409), (419, 421), (408, 428), (398, 430), (389, 429), (377, 424), (364, 412), (370, 413), (389, 414), (398, 415), (407, 413), (423, 411)]
The chin in this face has the following points: [(289, 295), (291, 323), (296, 351), (303, 378), (315, 403), (332, 428), (353, 448), (376, 464), (400, 467), (422, 461), (441, 444), (459, 425), (473, 403), (484, 377), (490, 351), (493, 323), (493, 296)]
The top_lip in this face has the following points: [(364, 412), (377, 407), (389, 403), (397, 406), (406, 402), (417, 405), (429, 409), (423, 411), (406, 412), (397, 414), (389, 413), (370, 413)]
The left_eye in this face has the following points: [(327, 308), (339, 304), (353, 306), (364, 314), (352, 317), (338, 316)]
The nose_tip in this face has the following points: [(375, 383), (386, 387), (396, 390), (407, 385), (416, 381)]

图片效果

Python3 利用face_recognition实现人脸识别的方法

到此这篇关于Python3 利用face_recognition实现人脸识别的方法的文章就介绍到这了,更多相关Python3 人脸识别内容请搜索三水点靠木以前的文章或继续浏览下面的相关文章希望大家以后多多支持三水点靠木!

Python 相关文章推荐
Python中基础的socket编程实战攻略
Jun 01 Python
在python的类中动态添加属性与生成对象
Sep 17 Python
Python实现将MySQL数据库表中的数据导出生成csv格式文件的方法
Jan 11 Python
pycharm 在windows上编辑代码用linux执行配置的方法
Oct 27 Python
python微信撤回监测代码
Apr 29 Python
python3中的eval和exec的区别与联系
Oct 10 Python
python ubplot使用方法解析
Jan 10 Python
Python中私有属性的定义方式
Mar 05 Python
三步解决python PermissionError: [WinError 5]拒绝访问的情况
Apr 22 Python
Pythonic版二分查找实现过程原理解析
Aug 11 Python
Pandas中DataFrame交换列顺序的方法实现
Dec 14 Python
python 实现图片特效处理
Apr 03 Python
在django中使用post方法时,需要增加csrftoken的例子
Mar 13 #Python
python 安装教程之Pycharm安装及配置字体主题,换行,自动更新
Mar 13 #Python
详解用Python进行时间序列预测的7种方法
Mar 13 #Python
django-xadmin根据当前登录用户动态设置表单字段默认值方式
Mar 13 #Python
在django项目中导出数据到excel文件并实现下载的功能
Mar 13 #Python
Django choices下拉列表绑定实例
Mar 13 #Python
django model object序列化实例
Mar 13 #Python
You might like
对盗链说再见...
2006/10/09 PHP
坏狼的PHP学习教程之第2天
2008/06/15 PHP
Jquery优化效率 提升性能解决方案
2010/09/06 Javascript
使用 TypeScript 重新编写的 JavaScript 坦克大战游戏代码
2015/04/07 Javascript
简介JavaScript中的getSeconds()方法的使用
2015/06/10 Javascript
jquery.validate 自定义验证方法及validate相关参数
2016/01/18 Javascript
AngularJS在IE下取数据总是缓存问题的解决方法
2016/08/05 Javascript
vue自定义指令实现v-tap插件
2016/11/03 Javascript
原生js实现图片放大缩小计时器效果
2017/01/20 Javascript
jquery横向纵向鼠标滚轮全屏切换
2017/02/27 Javascript
微信小程序实现给循环列表添加点击样式实例
2017/04/26 Javascript
JS兼容所有浏览器的DOMContentLoaded事件
2018/01/12 Javascript
JavaScript模拟实现自由落体效果
2018/08/28 Javascript
微信小程序实现tab左右切换效果
2020/11/15 Javascript
Vue中的Props(不可变状态)
2018/09/29 Javascript
js/jquery遍历对象和数组的方法分析【forEach,map与each方法】
2019/02/27 jQuery
JS数组方法push()、pop()用法实例分析
2020/01/18 Javascript
JavaScript TAB栏切换效果的示例
2020/11/05 Javascript
简单谈谈Python中的闭包
2016/11/30 Python
Python探索之ModelForm代码详解
2017/10/26 Python
python利用标准库如何获取本地IP示例详解
2017/11/01 Python
Python的CGIHTTPServer交互实现详解
2018/02/08 Python
python-opencv获取二值图像轮廓及中心点坐标的代码
2019/08/27 Python
使用Python解析Chrome浏览器书签的示例
2020/11/13 Python
python如何获得list或numpy数组中最大元素对应的索引
2020/11/16 Python
美国体育用品商店:Academy Sports + Outdoors
2020/01/04 全球购物
材料成型专业个人求职信范文
2013/09/25 职场文书
业务副厂长岗位职责
2014/01/03 职场文书
电脑专业个人求职信范文
2014/02/04 职场文书
房地产资料员岗位职责
2014/07/02 职场文书
党员干部反四风民主生活会对照检查材料思想汇报
2014/10/12 职场文书
公司员工违纪检讨书
2015/05/05 职场文书
商场广播稿范文
2015/08/19 职场文书
生日宴会家属答谢词
2015/09/29 职场文书
MySQL的存储过程和相关函数
2022/04/26 MySQL
JavaScript parseInt0.0000005打印5原理解析
2022/07/23 Javascript