再也不怕別人動電腦了！用Python實時監控

作者：美圖博客

https://www.meitubk.com/zatan/386.html

前言

最近突然有個奇妙的想法，就是當我對着電腦屏幕的時候，電腦會先識別屏幕上的人臉是否是本人，如果識別是本人的話需要回答電腦說的暗語，答對了纔會解鎖並且有三次機會。如果都沒答對就會發送郵件給我，通知有人在動我的電腦並上傳該人頭像。

過程

環境是 win10 代碼我使用的是 python3 所以在開始之前需要安裝一些依賴包，請按順序安裝否者會報錯

pip install cmake -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install dlib -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install face_recognition -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install opencv-python -i https://pypi.tuna.tsinghua.edu.cn/simple

接下來是構建識別人臉以及對比人臉的代碼

import face_recognition
import cv2
import numpy as np

video_capture = cv2.VideoCapture(0)
my_image = face_recognition.load_image_file("my.jpg")
my_face_encoding = face_recognition.face_encodings(my_image)[0]
known_face_encodings = [
    my_face_encoding
]
known_face_names = [
    "Admin"
]

face_names = []
face_locations = []
face_encodings = []
process_this_frame = True

while True:
    ret, frame = video_capture.read()
    small_frame = cv2.resize(frame, (0, 0), fx=0.25, fy=0.25)
    rgb_small_frame = small_frame[:, :, ::-1]
    if process_this_frame:
        face_locations = face_recognition.face_locations(rgb_small_frame)
        face_encodings = face_recognition.face_encodings(rgb_small_frame, face_locations)
        face_names = []
        for face_encoding in face_encodings:
            matches = face_recognition.compare_faces(known_face_encodings, face_encoding)
            name = "Unknown"
            face_distances = face_recognition.face_distance(known_face_encodings, face_encoding)
            best_match_index = np.argmin(face_distances)
            if matches[best_match_index]:
                name = known_face_names[best_match_index]
            face_names.append(name)

    process_this_frame = not process_this_frame
    for (top, right, bottom, left), name in zip(face_locations, face_names):
        top *= 4
        left *= 4
        right *= 4
        bottom *= 4
        font = cv2.FONT_HERSHEY_DUPLEX
        cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 255), 2)
        cv2.rectangle(frame, (left, bottom - 35), (right, bottom), (0, 0, 255), cv2.FILLED)
        cv2.putText(frame, name, (left + 6, bottom - 6), font, 1.0, (255, 255, 255), 1)

    cv2.imshow('Video', frame)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

video_capture.release()
cv2.destroyAllWindows()

其中 my.jpg 需要你自己拍攝上傳，運行可以發現在你臉上會出現 Admin 的框框，我去網上找了張圖片類似這樣子

識別功能已經完成了接下來就是語音識別和語音合成，這需要使用到百度AI來實現了，去登錄百度AI的官網到控制檯選擇左邊的語音技術，然後點擊面板的創建應用按鈕，來到創建應用界面

創建後會得到AppID、API Key、Secret Key記下來，然後開始寫語音合成的代碼。安裝百度AI提供的依賴包

pip install baidu-aip -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install playsound -i https://pypi.tuna.tsinghua.edu.cn/simple

然後是簡單的語音播放代碼，運行下面代碼可以聽到萌妹子的聲音

import sys
from aip import AipSpeech
from playsound import playsound

APP_ID = ''
API_KEY = ''
SECRET_KEY = ''

client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
result = client.synthesis('你好吖', 'zh', 1, {'vol': 5, 'per': 4, 'spd': 5, })

if not isinstance(result, dict):
    with open('auido.mp3', 'wb') as file:
        file.write(result)

filepath = eval(repr(sys.path[0]).replace('\\', '/')) + '//auido.mp3'
playsound(filepath)

有了上面的代碼就完成了檢測是否在電腦前（人臉識別）以及電腦念出暗語（語音合成）然後我們還需要回答暗號給電腦，所以還需要完成語音識別。

import wave
import pyaudio
from aip import AipSpeech

APP_ID = ''
API_KEY = ''
SECRET_KEY = ''

client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
CHUNK = 1024
FORMAT = pyaudio.paInt16
CHANNELS = 1
RATE = 8000
RECORD_SECONDS = 3
WAVE_OUTPUT_FILENAME = "output.wav"

p = pyaudio.PyAudio()
stream = p.open(format=FORMAT, channels=CHANNELS, rate=RATE, input=True, frames_per_buffer=CHUNK)

print("* recording")
frames = []
for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)):
    data = stream.read(CHUNK)
    frames.append(data)
print("* done recording")

stream.stop_stream()
stream.close()
p.terminate()
wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
wf.setnchannels(CHANNELS)
wf.setsampwidth(p.get_sample_size(FORMAT))
wf.setframerate(RATE)
wf.writeframes(b''.join(frames))


def get_file_content():
    with open(WAVE_OUTPUT_FILENAME, 'rb') as fp:
        return fp.read()


result = client.asr(get_file_content(), 'wav', 8000, {'dev_pid': 1537, })
print(result)

運行此代碼之前需要安裝 pyaudio 依賴包，由於在win10系統上安裝會報錯所以可以通過如下方式安裝。到這個鏈接 https://www.lfd.uci.edu/~gohlke/pythonlibs/#pyaudio 去下載對應的安裝包然後安裝即可。

運行後我說了你好，可以看到識別出來了。那麼我們的小模塊功能就都做好了接下來就是如何去整合它們。可以發現在人臉識別代碼中 if matches[best_match_index] 這句判斷代碼就是判斷是否爲電腦主人，所以我們把這個判斷語句當作main函數的入口。

if matches[best_match_index]:
    # 在這裏寫識別到之後的功能
    name = known_face_names[best_match_index]

那麼識別到後我們應該讓電腦發出詢問暗號，也就是語音合成代碼，然我們將它封裝成一個函數，順便重構下人臉識別的代碼。

import cv2
import time
import numpy as np
import face_recognition

video_capture = cv2.VideoCapture(0)
my_image = face_recognition.load_image_file("my.jpg")
my_face_encoding = face_recognition.face_encodings(my_image)[0]
known_face_encodings = [
    my_face_encoding
]
known_face_names = [
    "Admin"
]

face_names = []
face_locations = []
face_encodings = []
process_this_frame = True


def speak(content):
    import sys
    from aip import AipSpeech
    from playsound import playsound
    APP_ID = ''
    API_KEY = ''
    SECRET_KEY = ''
    client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
    result = client.synthesis(content, 'zh', 1, {'vol': 5, 'per': 0, 'spd': 5, })
    if not isinstance(result, dict):
        with open('auido.mp3', 'wb') as file:
            file.write(result)
    filepath = eval(repr(sys.path[0]).replace('\\', '/')) + '//auido.mp3'
    playsound(filepath)


try:
    while True:
        ret, frame = video_capture.read()
        small_frame = cv2.resize(frame, (0, 0), fx=0.25, fy=0.25)
        rgb_small_frame = small_frame[:, :, ::-1]
        if process_this_frame:
            face_locations = face_recognition.face_locations(rgb_small_frame)
            face_encodings = face_recognition.face_encodings(rgb_small_frame, face_locations)
            face_names = []
            for face_encoding in face_encodings:
                matches = face_recognition.compare_faces(known_face_encodings, face_encoding)
                name = "Unknown"
                face_distances = face_recognition.face_distance(known_face_encodings, face_encoding)
                best_match_index = np.argmin(face_distances)
                if matches[best_match_index]:
                    speak("識別到人臉，開始詢問暗號，請回答接下來我說的問題")
                    time.sleep(1)
                    speak("天王蓋地虎")
                    error = 1 / 0
                    name = known_face_names[best_match_index]
                face_names.append(name)
        process_this_frame = not process_this_frame
        for (top, right, bottom, left), name in zip(face_locations, face_names):
            top *= 4
            left *= 4
            right *= 4
            bottom *= 4
            font = cv2.FONT_HERSHEY_DUPLEX
            cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 255), 2)
            cv2.rectangle(frame, (left, bottom - 35), (right, bottom), (0, 0, 255), cv2.FILLED)
            cv2.putText(frame, name, (left + 6, bottom - 6), font, 1.0, (255, 255, 255), 1)

        cv2.imshow('Video', frame)
        if cv2.waitKey(1) & 0xFF == ord('q'):
            break
except Exception as e:
    print(e)
finally:
    video_capture.release()
    cv2.destroyAllWindows()

這裏有一點需要注意，由於 playsound 播放音樂的時候會一直佔用這個資源，所以播放下一段音樂的時候會報錯，解決方法是修改 ~\Python37\Lib\site-packages 下的 playsound.py 文件，找到如下代碼

在 sleep 函數下面添加 winCommand('close', alias) 這句代碼，保存下就可以了。運行發現可以正常將兩句話都說出來。那麼說出來之後就要去監聽了，我們還要打包一個函數。

def record():
    import wave
    import json
    import pyaudio
    from aip import AipSpeech

    APP_ID = ''
    API_KEY = ''
    SECRET_KEY = ''

    client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
    CHUNK = 1024
    FORMAT = pyaudio.paInt16
    CHANNELS = 1
    RATE = 8000
    RECORD_SECONDS = 3
    WAVE_OUTPUT_FILENAME = "output.wav"

    p = pyaudio.PyAudio()
    stream = p.open(format=FORMAT, channels=CHANNELS, rate=RATE, input=True, frames_per_buffer=CHUNK)

    print("* recording")
    frames = []
    for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)):
        data = stream.read(CHUNK)
        frames.append(data)
    print("* done recording")

    stream.stop_stream()
    stream.close()
    p.terminate()
    wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
    wf.setnchannels(CHANNELS)
    wf.setsampwidth(p.get_sample_size(FORMAT))
    wf.setframerate(RATE)
    wf.writeframes(b''.join(frames))

    def get_file_content():
        with open(WAVE_OUTPUT_FILENAME, 'rb') as fp:
            return fp.read()

    result = client.asr(get_file_content(), 'wav', 8000, {'dev_pid': 1537, })
    result = json.loads(str(result).replace("'", '"'))
    return result["result"][0]

將識別到人臉後的代碼修改成如下

if matches[best_match_index]:
    speak("識別到人臉，開始詢問暗號，請回答接下來我說的問題")
    time.sleep(1)
    speak("天王蓋地虎")

    flag = False
    for times in range(0, 3):
        content = record()
        if "小雞燉蘑菇" in content:
            speak("暗號通過")
            flag = True
            break
        else:
            speak("暗號不通過，再試一次")
    if flag:
        print("解鎖")
    else:
        print("發送郵件並將壞人人臉圖片上傳！")
    error = 1 / 0
    name = known_face_names[best_match_index]

運行看看效果，回答電腦 小雞燉蘑菇 ，電腦回答暗號通過。這樣功能就基本上完成了。