When you are fishing at work, leaders will always sneak up behind you. For example, Xiaobian once watched "Qingyin" sneakily and was caught. Today, we will use Python to crack this fish catching routine. The main idea is to use opencv to call the computer camera to detect and compare faces. When the leader appears behind, open the designated application to fish in troubled waters.
Install the module
Before writing the program, the camera module called by opencv and the face_locations face recognition module must be installed.
# opencv 模块
pip3 insatll opencv-python
# face_locations 模块
pip3 install cmake
pip3 install face_recognition
If you encounter difficulties in learning and want to find a python learning and communication environment, you can join our python skirt, follow the editor, and privately write "01" to enter the skirt and receive python learning materials, which will save a lot of time and reduce a lot Difficulties encountered.
Face recognition
First, encode the leader's face and put it into the memory so that it can be compared with the face recognition captured by the camera at any time.
For example, face encoding for the image below
import face_recognition
# 加载图片
pic_boss = face_recognition.load_image_file("/Users/xx/Desktop/face/0.png")
# 得到面部编码
boss_face_encoding = face_recognition.face_encodings(pic_boss)[0]
Example result
[-0.02630499 0.12300251 0.01698755 0.01275834 -0.07418888 -0.00981654
-0.03014973 -0.16349442 0.11407443 -0.03254088 0.26810074 -0.10167226
-0.15427223 -0.11180711 0.01873804 0.18030289 -0.14980686 -0.12194286
-0.02620432 -0.03438358 0.04720668 0.05182001 0.009987 0.09340413
-0.11347414 -0.26094455 -0.07678577 -0.09128669 0.08928929 -0.07264765
-0.11346096 0.1254302 -0.20916753 -0.10639326 0.09938065 0.05473833
-0.04935627 -0.06184902 0.17554277 -0.02231439 -0.19398358 0.01744412
0.10445927 0.26399308 0.21656345 0.05588599 0.00760998 -0.13855973
0.12407181 -0.09017442 0.09778374 0.11776066 0.08498169 0.07626694
-0.01237833 -0.20856641 0.02468708 0.06579788 -0.12179989 0.02987257
0.1338616 -0.07621925 0.01559527 -0.03452411 0.18915619 0.01698355
-0.04450341 -0.2167782 0.09793964 -0.11409818 -0.10012487 0.13745219
-0.17124982 -0.15017164 -0.34856451 -0.01826046 0.41654593 0.09037441
-0.21255262 0.04289294 0.01760755 -0.01859214 0.20364219 0.14642054
0.00619181 -0.02451363 -0.15138477 0.00500458 0.25368348 -0.02767867
-0.09737059 0.17870165 -0.02200082 0.03460512 0.03690759 0.06052291
-0.0686012 0.04330266 -0.15649761 -0.09057935 0.00870521 0.04586265
-0.04279764 0.18815981 -0.15697879 0.17292421 0.03271531 0.08370531
-0.04779428 -0.05095051 -0.08721299 0.01937558 0.10537415 -0.21216688
0.16163379 0.07646587 0.09025833 0.08259746 0.08811771 0.06535679
-0.01029789 -0.06432205 -0.25512969 -0.03111095 0.12503427 -0.02948561
0.15236887 0.03259711]
Call camera
Use the cv2.VideoCapture(0) method to turn up the camera, and perform face encoding on each frame of the camera, and finally use the face_recognition.compare_faces() method to compare faces.
import face_recognition
import cv2
import os
import time
# 调用摄像头，外部摄像头为：1
cap = cv2.VideoCapture(0)
while True:
# 按帧读取视频
# 其中ret是布尔值，如果读取帧是正确的则返回 True，
# frame就是每一帧的图像
ret, frame = cap.read()
# 进行面部编码
face_locations = face_recognition.face_locations(frame)
face_encodings = face_recognition.face_encodings(frame, face_locations)
for face_encoding in face_encodings:
# 人脸对比
results = face_recognition.compare_faces([boss_face_encoding], face_encoding)
if results[0]:
print("boss来了，快打开其他应用")
os.system('open /Applications/PyCharm.app')
time.sleep(300)
It should be noted that the cv2.VideoCapture(0) method will report: Abort trap: 6 error when called with terminal in VSCode.
There is no such problem when running in the terminal that comes with the system.
to sum up
You should fish well when you can fish at work, and never fish when you should be fighting hard.
[ Code acquisition method ]
I still want to recommend the Python learning group I built myself : 705933274. The group is all learning Python. If you want to learn or are learning Python, you are welcome to join. Everyone is a software development party and share dry goods from time to time (only Python software development related), including a copy of the latest Python advanced materials and zero-based teaching compiled by myself in 2021. Welcome friends who are in advanced and interested in Python to join!