介绍Python的抖音快手字符舞

import argparseimport osimport cv2import subprocessfrom cv2 import VideoWriter_fourccfrom PIL import Image, ImageFont, ImageDraw# 命令行输入参数处理# aparser = argparse.ArgumentParser()# aparser.add_argument('file')# aparser.add_argument('-o','--output')# aparser.add_argument('-f','--fps',type = float,default= 24)#帧# aparser.add_argument('-s','--save',type = bool, nargs='?',default= False,const= True)# 是否保留Cache文件，默认不保存classVideo2CodeVideo:

    def __init__(self):

        self.config_dict = {

            # 原视频文件

            "input_file":"video/test.mp4",

            # 中间文件存放目录

            "cache_dir":"cache",

            # 是否保留过程文件。True--保留，False--不保留

            "save_cache_flag": False,

            # 使用使用的字符集

            "ascii_char_list": list("01B%8&WM#*oahkbdpqwmZO0QLCJUYXzcvunxrjft/\|()1{}[]?-_+~<>i!lI;:oa+>!:+. "),

        }

 

    # 第一步从函数，将像素转换为字符

    # 调用栈：video_2_txt_jpg -> txt_2_image -> rgb_2_char

    def rgb_2_char(self, r, g, b, alpha=256):

        ifalpha == 0:

            return''

        length = len(self.config_dict["ascii_char_list"])

        gray = int(0.2126 * r + 0.7152 * g + 0.0722 * b)

        unit = (256.0 + 1) / length       returnself.config_dict["ascii_char_list"][int(gray / unit)]

 

    # 第一步从函数，将txt转换为图片

    # 调用栈：video_2_txt_jpg -> txt_2_image -> rgb_2_char

    def txt_2_image(self, file_name):

        im = Image.open(file_name).convert('RGB')

        # gif拆分后的图像，需要转换，否则报错，由于gif分割后保存的是索引颜色

        raw_width = im.width

        raw_height = im.height

        width = int(raw_width / 6)

        height = int(raw_height / 15)

        im = im.resize((width, height), Image.NEAREST)

 

        txt =""

        colors = []

        fori in range(height):

            forj in range(width):

                pixel = im.getpixel((j, i))

                colors.append((pixel[0], pixel[1], pixel[2]))

                if(len(pixel) == 4):

                    txt += self.rgb_2_char(pixel[0], pixel[1], pixel[2], pixel[3])

                else:

                    txt += self.rgb_2_char(pixel[0], pixel[1], pixel[2])

            txt +='\n'

            colors.append((255, 255, 255))

 

        im_txt = Image.new("RGB", (raw_width, raw_height), (255, 255, 255))

        dr = ImageDraw.Draw(im_txt)

        # font = ImageFont.truetype(os.path.join("fonts","汉仪楷体简.ttf"),18)

        font = ImageFont.load_default().font

        x = y = 0

        # 获取字体的宽高

        font_w, font_h = font.getsize(txt[1])

        font_h *= 1.37  # 调整后更佳

        # ImageDraw为每个ascii码进行上色

        fori in range(len(txt)):

            if(txt[i] =='\n'):

                x += font_h

                y = -font_w            # self, xy, text, fill = None, font = None, anchor = None,

            # *args, ** kwargs

            dr.text((y, x), txt[i], fill=colors[i])

            # dr.text((y, x), txt[i], font=font, fill=colors[i])

            y += font_w

 

        name = file_name        #print(name +' changed')

        im_txt.save(name)

 

 

    # 第一步，将原视频转成字符图片

    # 调用栈：video_2_txt_jpg -> txt_2_image -> rgb_2_char

    def video_2_txt_jpg(self, file_name):

        vc = cv2.VideoCapture(file_name)

        c = 1

        ifvc.isOpened():

            r, frame = vc.read()

            ifnot os.path.exists(self.config_dict["cache_dir"]):

                os.mkdir(self.config_dict["cache_dir"])

            os.chdir(self.config_dict["cache_dir"])

        else:

            r = False

        whiler:

            cv2.imwrite(str(c) +'.jpg', frame)

            self.txt_2_image(str(c) +'.jpg')  # 同时转换为ascii图

            r, frame = vc.read()

            c += 1

        os.chdir('..')

        returnvc    # 第二步，将字符图片合成新视频

    def txt_jpg_2_video(self, outfile_name, fps):

        fourcc = VideoWriter_fourcc(*"MJPG")

 

        images = os.listdir(self.config_dict["cache_dir"])

        im = Image.open(self.config_dict["cache_dir"] +'/'+ images[0])

        vw = cv2.VideoWriter(outfile_name +'.avi', fourcc, fps, im.size)

 

        os.chdir(self.config_dict["cache_dir"])

        forimage in range(len(images)):

            # Image.open(str(image)+'.jpg').convert("RGB").save(str(image)+'.jpg')

            frame = cv2.imread(str(image + 1) +'.jpg')

            vw.write(frame)

            #print(str(image + 1) +'.jpg'+' finished')

        os.chdir('..')

        vw.release()

 

    # 第三步，从原视频中提取出背景音乐

    def video_extract_mp3(self, file_name):

        outfile_name = file_name.split('.')[0] +'.mp3'

        subprocess.call('ffmpeg -i '+ file_name +' -f mp3 -y '+ outfile_name, shell=True)

 

    # 第四步，将背景音乐添加到新视频中

    def video_add_mp3(self, file_name, mp3_file):

        outfile_name = file_name.split('.')[0] +'-txt.mp4'

        subprocess.call('ffmpeg -i '+ file_name +' -i '+ mp3_file +' -strict -2 -f mp4 -y '+ outfile_name, shell=True)