Gray = R*0.299 + G*0.587 + B*0.114 

# -*- coding: utf-8 -*-
from PIL import Image

codeLib = '''@B%8&WM#*oahkbdpqwmZO0QLCJUYXzcvunxrjft/\|()1{}[]"^`'. '''#生成字符画所需的字符集
count = len(codeLib)

def transform1(image_file):
  image_file = image_file.convert("L")#转换为黑白图片，参数"L"表示黑白模式
  codePic = ''
  for h in range(0,image_file.size[1]): #size属性表示图片的分辨率，'0'为横向大小，'1'为纵向
    for w in range(0,image_file.size[0]):
      gray = image_file.getpixel((w,h)) #返回指定位置的像素，如果所打开的图像是多层次的图片，那这个方法就返回一个元组
      codePic = codePic + codeLib[int(((count-1)*gray)/256)]#建立灰度与字符集的映射
    codePic = codePic+'\r\n'
  return codePic

def transform2(image_file):
  codePic = ''
  for h in range(0,image_file.size[1]):
    for w in range(0,image_file.size[0]):
      g,r,b = image_file.getpixel((w,h))
      gray = int(r* 0.299+g* 0.587+b* 0.114)
      codePic = codePic + codeLib[int(((count-1)*gray)/256)]
    codePic = codePic+'\r\n'
  return codePic


fp = open(u'暴走.jpg','rb')
image_file = Image.open(fp)
image_file=image_file.resize((int(image_file.size[0]*0.75), int(image_file.size[1]*0.5)))#调整图片大小
print u'Info:',image_file.size[0],' ',image_file.size[1],' ',count

tmp = open('tmp.txt','w')
tmp.write(transform1(image_file))
tmp.close()