中科创达实训第八天：摄像头和音视频

一、课件

二、课后习题

1、准备开发环境

安装opencv

Python2：apt install python-opencv -y

Python3：pip3 install opencv-python

安装PyAduio

Python2：

apt-get install portaudio19-dev python-all-dev apt-get install python-pyaudio

Python3：

apt-get install portaudio19-dev python3-all-dev

apt-get install python3-pyaudio

Windows： PyAudio‑0.2.11‑cp27‑cp27m‑win32.whl https://download.lfd.uci.edu/pythonlibs/w3jqiv8s/cp27/PyAudio-0.2.11-cp27-cp27m-win32.whl

PyAudio‑0.2.11‑cp37‑cp37m‑win32.whl https://download.lfd.uci.edu/pythonlibs/w3jqiv8s/PyAudio-0.2.11-cp37-cp37m-win32.whl

2、学习Python摄像头、Python音频录放音、Python视频播放例程 学习三个例程代码，运行一下，把过程录屏，保存到视频文件中

3、修改Python摄像头或者视频播放例程的代码，实现三个函数，分别把彩色图像转为灰度图、二值化和调整图片尺寸，提交Python脚本源文件

三、答案

音频

#!/usr/bin/python
# -*- coding: utf-8 -*-

import wave
import pyaudio

CHUNK = 1024

RECORD_RATE = 44100
RECORD_FORMAT = pyaudio.paInt16
RECORD_SECONDS = 10
RECORD_CHANNELS = 1


class Audio:
    def __init__(self, wave_file):
        self.wave = None
        self.audio_params = dict()
        self.audio_file = wave_file
        self.audio = pyaudio.PyAudio()

    def get_wav_file(self):
        if not self.wave:
            self.wave = wave.open(self.audio_file, 'rb')
        return self.wave

    def get_audio_params(self):
        if 0 == len(self.audio_params):
            self.get_wav_file()
            self.audio_params['rate'] = self.wave.getframerate()
            self.audio_params['channels'] = self.wave.getnchannels()
            self.audio_params['format'] = self.audio.get_format_from_width(self.wave.getsampwidth())
        print(self.audio_params)
        return self.audio_params

    def play(self):
        params = self.get_audio_params()
        stream = self.audio.open(format=params['format'], channels=params['channels'], rate=params['rate'], output=True)
        data = self.wave.readframes(CHUNK)
        while data != b'':
            stream.write(data)
            data = self.wave.readframes(CHUNK)
        stream.stop_stream()
        stream.close()
        self.audio.terminate()

    def record(self):
        stream = self.audio.open(format=RECORD_FORMAT, channels=RECORD_CHANNELS,
                                 rate=RECORD_RATE, input=True, frames_per_buffer=CHUNK)
        print("* start recording")
        frames = []
        for i in range(0, int(RECORD_RATE / CHUNK * RECORD_SECONDS)):
            data = stream.read(CHUNK)
            frames.append(data)
        print("* done recording")
        stream.stop_stream()
        stream.close()
        self.audio.terminate()
        self.save_wav_file(frames)

    def save_wav_file(self, frames):
        # 新建wav文件
        wf = wave.open('record.wav', 'wb')
        wf.setnchannels(RECORD_CHANNELS)
        wf.setsampwidth(self.audio.get_sample_size(RECORD_FORMAT))
        wf.setframerate(RECORD_RATE)
        # 音频数据写入到wav文件中
        wf.writeframes(b''.join(frames))
        wf.close()


def aikit_paly_audio():
    audio = Audio('test.wav')
    audio.play()


def aikit_record_audio():
    audio = Audio('record.wav')
    audio.record()
    audio = Audio('record.wav')
    audio.play()


if __name__ == '__main__':
    # aikit_paly_audio()
    aikit_record_audio()
    pass

视频

#!/usr/bin/python
# -*- coding: utf-8 -*-


import cv2


def to_grey(frame):
    video_gray = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
    cv2.imshow('video', video_gray)


def rgb_color(frame):
    cv2.imshow('video', frame)  # 显示


def resize(frame):
    # 获取图片的宽和高
    width, height = frame.shape[:2][::-1]
    # 将图片缩小便于显示观看
    video_resize = cv2.resize(frame,(int(width * 0.5), int(height * 0.5)), interpolation=cv2.INTER_CUBIC)
    cv2.imshow("video", video_resize)


def two_value(frame):
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    ret, binary = cv2.threshold(gray, 100, 255, cv2.THRESH_BINARY)
    cv2.imshow("video", binary)


def play_video():
    # 获得视频的格式
    video_capture = cv2.VideoCapture('喝水智能助手2.mp4')

    # 获得码率及尺寸
    fps = video_capture.get(cv2.CAP_PROP_FPS)
    size = (int(video_capture.get(cv2.CAP_PROP_FRAME_WIDTH)), int(video_capture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
    nums = video_capture.get(cv2.CAP_PROP_FRAME_COUNT)

    # 读帧
    success, frame = video_capture.read()
    while success:

        # to_grey(frame)
        # resize(frame)
        rgb_color(frame)
        # two_value(frame)

        cv2.waitKey(int(1000 / int(fps)))  # 延迟
        success, frame = video_capture.read()  # 获取下一帧

    video_capture.release()
    cv2.destroyAllWindows()


if __name__ == '__main__':
    play_video()
    pass

摄像头

#!/usr/bin/python
# -*- coding: utf-8 -*-

import cv2


def aikit_take_pictures():
    camera = cv2.VideoCapture(0)
    while True:
        ret, frame = camera.read()  # 一帧一帧读取视频
        cv2.imshow('frame', frame)  # 显示结果
        # 根据用户的键盘操作，保存照片或者停止程序
        key_code = cv2.waitKey(1)
        if key_code & 0xFF == ord('p'):  # 按p保存照片
            cv2.imwrite("photo.jpg", frame)
            print('Successful photo')
        if key_code & 0xFF == ord('q'):  # 按q停止
            break

    camera.release()
    cv2.destroyAllWindows()


if __name__ == '__main__':
    aikit_take_pictures()
    pass

四、Opencv和pyAudio安装

注意事项：

1.Pycharm在创建工程的时候要勾选

inherit global site-packages

Make available to all projectes

2.使用摄像头和麦克风的时候一定要开启系统调用权限