you-get [url]

you-get https://www.bilibili.com/video/BV1yY4y1i7Pw?t=1079.2   # 一行代码下载视频

import cv2
import numpy as np
import random
import os
保存图片的函数
def save_images(image, addr,num):

address = addr + str(num) + ".jpg"

cv2.imwrite(address, image)
读取视频
videoCapture = cv2.VideoCapture("刘畊宏，毽子操x10+蝴蝶袖+臀腿操（自用）.mp4")

success, frame = videoCapture.read()
time_ = 100

i = 0

j = 0
while success:  # 如果成功获取到视频帧

i += 1

if i % time_ == 0:

s = 0  # 名称的编号

j = j + 1

s += j

save_images(frame, "./picture/",s)  # 写入目录后再继续读取

success, frame = videoCapture.read()
videoCapture.release()  # 释放资源

pip install baidu_aip  # 安装库，一定要是baidu_aip

# 官方demo

""" 读取图片 """

def get_file_content(filePath):

with open(filePath, 'rb') as fp:

return fp.read()

image = get_file_content('example.jpg')

""" 调用人像分割 """

client.bodySeg(image);
""" 如果有可选参数 """

options = {}

options["type"] = "labelmap"
""" 带参数调用人像分割 """

client.bodySeg(image, options)

import re

import requests

import pandas as pd

import time

from tqdm import trange

视频页面点击“浏览器地址栏小锁-Cookie-bilibili.com-Cookie-SESSDATA”进行获取
SESSDATA = ""
视频页面“按F12-Console-输入document.cookie”进行获取
cookie = ""
cookie += f";SESSDATA={SESSDATA}"
headers = {

"user-agent": "",

"cookie": cookie,

}
def get_info(vid):

"""

vid：视频号id

"""
# 构造URL地址
# https://api.bilibili.com/x/web-interface/view/detail?bvid=BV1yY4y1i7Pw
url = f"https://api.bilibili.com/x/web-interface/view/detail?bvid={vid}"

response = requests.get(url, headers=headers)
response.encoding = "utf-8"
data = response.json()  # json格式数据

info = {}
info["标题"] = data["data"]["View"]["title"]
info["总弹幕数"] = data["data"]["View"]["stat"]["danmaku"]
info["视频数量"] = data["data"]["View"]["videos"]
info["cid"] = [dic["cid"] for dic in data["data"]["View"]["pages"]]

if info["视频数量"] > 1:
    info["子标题"] = [dic["part"] for dic in data["data"]["View"]["pages"]]

for k, v in info.items():
print(k + ":", v)
return info

def get_danmu(info, start, end):
# 爬取的时间范围设置
date_list = [i for i in pd.date_range(start, end).strftime("%Y-%m-%d")]
all_dms = []

for i, cid in enumerate(info["cid"]):
    dms = []
    for j in trange(len(date_list)):
        url = f"https://api.bilibili.com/x/v2/dm/web/history/seg.so?type=1&oid={cid}&date={date_list[j]}"
        response = requests.get(url, headers=headers)
        response.encoding = "utf-8"

        data = re.findall(r"[:](.*?)[@]", response.text)
        dms += [dm[1:] for dm in data]
        time.sleep(3)

if info["视频数量"] > 1:
print("cid:", cid, "弹幕数:", len(dms), "子标题:", info["子标题"][i])
    all_dms += dms
print(f"共获取弹幕{len(all_dms)}条！")
return all_dms

if name == "main":

vid = input("输入视频编号: ")
vid = "BV1yY4y1i7Pw"
info = get_info(vid)
print("info的内容：",info)

start = input("弹幕开始时间（年-月-日）: ")
end = input("弹幕结束时间（年-月-日）: ")

danmu = get_danmu(info, start, end)
with open("danmu3.txt", "w", encoding="utf-8") as f:
    for dm in danmu:
        f.write(dm + &#34;\n&#34;)</code></pre></div></div><figure class=""></figure><h3 id="4v61l" name="%E5%90%88%E6%88%90%E8%AF%8D%E4%BA%91%E5%9B%BE">合成词云图</h3><p>弹幕的分词是自己的方法和收集的一份常用的停用词表：</p><p>1、分词使用的jieba分词。关于jieba分词的使用入门，参考：https://github.com/fxsjy/jieba</p><p>快速安装jieba：</p><div class="rno-markdown-code"><div class="rno-markdown-code-toolbar"><div class="rno-markdown-code-toolbar-info"><div class="rno-markdown-code-toolbar-item is-type"><span class="is-m-hidden">代码语言：</span>javascript</div></div><div class="rno-markdown-code-toolbar-opt"><div class="rno-markdown-code-toolbar-copy"><i class="icon-copy"></i><span class="is-m-hidden">复制</span></div></div></div><div class="developer-code-block"><pre class="prism-token token line-numbers language-javascript"><code class="language-javascript" style="margin-left:0">pip install jieba</code></pre></div></div><div class="rno-markdown-code"><div class="rno-markdown-code-toolbar"><div class="rno-markdown-code-toolbar-info"><div class="rno-markdown-code-toolbar-item is-type"><span class="is-m-hidden">代码语言：</span>javascript</div></div><div class="rno-markdown-code-toolbar-opt"><div class="rno-markdown-code-toolbar-copy"><i class="icon-copy"></i><span class="is-m-hidden">复制</span></div></div></div><div class="developer-code-block"><pre class="prism-token token line-numbers language-javascript"><code class="language-javascript" style="margin-left:0">import pandas as pd

import numpy as np

import jieba
from wordcloud import WordCloud

from tkinter import _flatten

import matplotlib.pyplot as plt

%matplotlib inline
import collections

import re

import os

from PIL import Image
df = pd.DataFrame()
获取了3个和刘教练相关的视频弹幕
txt_list = ["danmu.txt", "danmu1.txt", "danmu2.txt"]
for txt in txt_list:

df1 = pd.read_table(txt, header=None, on_bad_lines='skip')

df1.columns = ["information"]  # 重命名

df1.drop_duplicates("information",inplace=True)
df = pd.concat([df, df1])

df.head()

stopwords = [i.strip() for i in open(r'/Users/peter/Desktop/spider/nlp_stopwords.txt').readlines()]

def data_cut(sentence):

# [\u4e00-\u9fa5] 表示只需要弹幕的中文内容

cut_list = jieba.lcut(''.join(re.findall('[\u4e00-\u9fa5]', sentence)), cut_all = False)
# 循环整个cut_list
for i in range(len(cut_list)-1, -1, -1):
    if cut_list[i] in stopwords:   # 如果元素在停用词表中则删除该信息
        del cut_list[i]
return cut_list

result = list(map(lambda x: data_cut(x), information_list))

useful_result = [j for i in result for j in i]  # 双层列表的转换

from pyecharts.globals import CurrentConfig, OnlineHostType

from pyecharts import options as opts  # 配置项

from pyecharts.charts import WordCloud  # 词云图

from pyecharts.commons.utils import JsCode

from pyecharts.globals import ThemeType,SymbolType

rec_words = [tuple(z) for z in zip(information["word"].tolist(), information["number"].tolist())]

c = (

WordCloud()

.add("", rec_words[:80], word_size_range=[20, 100], shape=SymbolType.DIAMOND)

.set_global_opts(title_opts=opts.TitleOpts(title="B站词云图"))

)
c.render_notebook()

wordcloud.WordCloud(

font_path=None,  # 字体路径，英文不用设置路径，中文需要，否则无法正确显示图形

width=400, # 默认宽度

height=200, # 默认高度

margin=2, # 边缘

ranks_only=None,

prefer_horizontal=0.9,

mask=None, # 背景图形，如果想根据图片绘制，则需要设置

scale=1,

color_func=None,

max_words=200, # 最多显示的词汇量

min_font_size=4, # 最小字号

stopwords=None, # 停用词设置，修正词云图时需要设置

random_state=None,

background_color='black', # 背景颜色设置，可以为具体颜色,比如white或者16进制数值

max_font_size=None, # 最大字号

font_step=1,

mode='RGB',

relative_scaling='auto',

regexp=None,

collocations=True,

colormap='viridis', # matplotlib 色图，可更改名称进而更改整体风格

normalize_plurals=True,

contour_width=0,

contour_color='black',

repeat=False

word_counts = collections.Counter(useful_result)  # 筛选后统计词频

path = './wordcloud/'  # 新建：存放词云图的路径

img_files = os.listdir('./mask')
遍历mask目录下的全部文件
for num in range(1, len(img_files) + 1):

img = r'./mask/mask_{}.png'.format(num)    # 原图片路径

mask_ = 255 - np.array(Image.open(img))  # 获取蒙版图片

# 绘制词云

plt.figure(figsize=(8, 5), dpi=200)
my_cloud = WordCloud(
    background_color='black',  # 背景颜色
    mask=mask_,      # 自定义蒙版
    mode='RGBA',
    max_words=500,
    # 地址路径要改成自己的ttf文件路径
    font_path=r'/Users/peter/Desktop/spider/SimHei.ttf'
).generate_from_frequencies(word_counts)

# 显示词云图
plt.imshow(my_cloud)
# 词云图中无坐标轴
plt.axis('off')
wordcloud_name = path + 'wordcloud_{}.png'.format(num)
my_cloud.to_file(wordcloud_name)    # 保存词云图片</code></pre></div></div><p>对应生成的词云图效果：</p><p></p><p></p><h3 id="cefis" name="%E5%90%88%E6%88%90%E8%AF%8D%E4%BA%91%E8%A7%86%E9%A2%91">合成词云视频</h3><p>基于上面的835张词云图来生成视频：</p><p></p><div class="rno-markdown-code"><div class="rno-markdown-code-toolbar"><div class="rno-markdown-code-toolbar-info"><div class="rno-markdown-code-toolbar-item is-type"><span class="is-m-hidden">代码语言：</span>javascript</div></div><div class="rno-markdown-code-toolbar-opt"><div class="rno-markdown-code-toolbar-copy"><i class="icon-copy"></i><span class="is-m-hidden">复制</span></div></div></div><div class="developer-code-block"><pre class="prism-token token line-numbers language-javascript"><code class="language-javascript" style="margin-left:0">import cv2

import os
输出视频的保存路径
video_dir = 'jianshen.mp4'
帧率，控制视频快慢
fps = 5
图片尺寸
img_size = (1920, 1080)
fourcc = cv2.VideoWriter_fourcc('M', 'P', '4', 'V')  # opencv3.0 mp4会有警告但可以播放

videoWriter = cv2.VideoWriter(video_dir, fourcc, fps, img_size)

img_files = os.listdir('./wordcloud')
for i in range(1, 836):

img_path = './wordcloud//wordcloud_{}.png'.format(i)

frame = cv2.imread(img_path)

frame = cv2.resize(frame, img_size)   # 生成视频   图片尺寸和设定尺寸相同

videoWriter.write(frame)      # 写进视频里

print(f'=按照视频顺序第{i}张图片合进视频=')
释放资源
videoWriter.release()

pip install moviepy

import moviepy.editor as mpy

读取词云视频：上个步骤生成的视频
my_clip = mpy.VideoFileClip('jianshen.mp4')
截取背景音乐 指定时间范围
本草纲目的MP3自己下载，放到同一个目录下
audio_background = mpy.AudioFileClip('本草纲目.mp3').subclip(28,60)

audio_background.write_audiofile('本草纲目1.mp3')
插入音频
final_clip = my_clip.set_audio(audio_background)
最终视频
final_clip.write_videofile('LGH.mp4')