http://robertgawron.blogspot.hk/2010/11/almost-all-sites-use-images-with-text.html

图片的识别主要有，去色，减噪，去线，分割，二值化，提取特征码

这里比较方便的是使用tesseract

1，准备库

apt-get install python-imaging

pip install Pillow

apt-get install tesseract-ocr

apt-get -f iinstall

wget https://python-tesseract.googlecode.com/files/python-tesseract_0.8-1.5_i386.deb

dpkg -i python-tesseract_0.8-1.5_i386.deb

2，图片处理

这里用到Pillow代替PIL，import改成

from PIL import Image,ImageFilter

用法 XXX.py XX.jpg 100

import sys

import Image

import ImageFilter

def prepare_image(img):

"""Transform image to greyscale and blur it"""

img = img.filter(ImageFilter.SMOOTH_MORE)

img = img.filter(ImageFilter.SMOOTH_MORE)

if 'L' != img.mode:

img = img.convert('L')

return img

def remove_noise(img, pass_factor):

for column in range(img.size[0]):

for line in range(img.size[1]):

value = remove_noise_by_pixel(img, column, line, pass_factor)

img.putpixel((column, line), value)

return img

def remove_noise_by_pixel(img, column, line, pass_factor):

if img.getpixel((column, line)) < pass_factor:

return (0)

return (255)

if __name__=="__main__":

input_image = sys.argv[1]

output_image = 'out_' + input_image

pass_factor = int(sys.argv[2])

img = Image.open(input_image)

img = prepare_image(img)

img = remove_noise(img, pass_factor)

img.save(output_image)

3，识别

这里用到开源OCR库tesseract

开源的软件免费，但相对的说明文件就不足，只有自己查代码

因为验证码一般只有一行，这个

api.SetPageSegMode(tesseract.PSM_AUTO)

改成

api.SetPageSegMode(7)

import tesseract

api = tesseract.TessBaseAPI()

api.Init(".","eng",tesseract.OEM_DEFAULT)

api.SetVariable("tessedit_char_whitelist", "0123456789abcdefghijklmnopqrstuvwxyz")

api.SetPageSegMode(tesseract.PSM_AUTO)

mImgFile = "eurotext.jpg"

mBuffer=open(mImgFile,"rb").read()

result = tesseract.ProcessPagesBuffer(mBuffer,len(mBuffer),api)

print "result(ProcessPagesBuffer)=",result