From 29f9dc8fb3396b03d732ebcbeb1cc8f00fe13897 Mon Sep 17 00:00:00 2001 From: RaNaN Date: Wed, 25 Aug 2010 18:22:27 +0200 Subject: new dirs --- module/plugins/captcha/GigasizeCom.py | 19 ++ module/plugins/captcha/LinksaveIn.py | 147 +++++++++++++++ module/plugins/captcha/MegauploadCom.py | 14 ++ module/plugins/captcha/NetloadIn.py | 24 +++ module/plugins/captcha/ShareonlineBiz.py | 53 ++++++ module/plugins/captcha/__init__.py | 0 module/plugins/captcha/captcha.py | 313 +++++++++++++++++++++++++++++++ 7 files changed, 570 insertions(+) create mode 100644 module/plugins/captcha/GigasizeCom.py create mode 100644 module/plugins/captcha/LinksaveIn.py create mode 100644 module/plugins/captcha/MegauploadCom.py create mode 100644 module/plugins/captcha/NetloadIn.py create mode 100644 module/plugins/captcha/ShareonlineBiz.py create mode 100644 module/plugins/captcha/__init__.py create mode 100644 module/plugins/captcha/captcha.py (limited to 'module/plugins/captcha') diff --git a/module/plugins/captcha/GigasizeCom.py b/module/plugins/captcha/GigasizeCom.py new file mode 100644 index 000000000..d31742eb5 --- /dev/null +++ b/module/plugins/captcha/GigasizeCom.py @@ -0,0 +1,19 @@ +# -*- coding: utf-8 -*- +from captcha import OCR + +class GigasizeCom(OCR): + def __init__(self): + OCR.__init__(self) + + def get_captcha(self, image): + self.load_image(image) + self.threshold(2.8) + self.run_tesser(True, False, False, True) + return self.result_captcha + +if __name__ == '__main__': + ocr = GigasizeCom() + import urllib + urllib.urlretrieve('http://www.gigasize.com/randomImage.php', "gigasize_tmp.jpg") + + print ocr.get_captcha('gigasize_tmp.jpg') diff --git a/module/plugins/captcha/LinksaveIn.py b/module/plugins/captcha/LinksaveIn.py new file mode 100644 index 000000000..3ad7b265a --- /dev/null +++ b/module/plugins/captcha/LinksaveIn.py @@ -0,0 +1,147 @@ +from captcha import OCR +import Image +from os import sep +from os.path import dirname +from os.path import abspath +from glob import glob + + +class LinksaveIn(OCR): + __name__ = "LinksaveIn" + def __init__(self): + OCR.__init__(self) + self.data_dir = dirname(abspath(__file__)) + sep + "LinksaveIn" + sep + + def load_image(self, image): + im = Image.open(image) + frame_nr = 0 + + lut = im.resize((256, 1)) + lut.putdata(range(256)) + lut = list(lut.convert("RGB").getdata()) + + new = Image.new("RGB", im.size) + npix = new.load() + while True: + try: + im.seek(frame_nr) + except EOFError: + break + frame = im.copy() + pix = frame.load() + for x in range(frame.size[0]): + for y in range(frame.size[1]): + if lut[pix[x, y]] != (0,0,0): + npix[x, y] = lut[pix[x, y]] + frame_nr += 1 + new.save(self.data_dir+"unblacked.png") + self.image = new.copy() + self.pixels = self.image.load() + self.result_captcha = '' + + def get_bg(self): + stat = {} + cstat = {} + img = self.image.convert("P") + for bgpath in glob(self.data_dir+"bg/*.gif"): + stat[bgpath] = 0 + bg = Image.open(bgpath) + + bglut = bg.resize((256, 1)) + bglut.putdata(range(256)) + bglut = list(bglut.convert("RGB").getdata()) + + lut = img.resize((256, 1)) + lut.putdata(range(256)) + lut = list(lut.convert("RGB").getdata()) + + bgpix = bg.load() + pix = img.load() + for x in range(bg.size[0]): + for y in range(bg.size[1]): + rgb_bg = bglut[bgpix[x, y]] + rgb_c = lut[pix[x, y]] + try: + cstat[rgb_c] += 1 + except: + cstat[rgb_c] = 1 + if rgb_bg == rgb_c: + stat[bgpath] += 1 + max_p = 0 + bg = "" + for bgpath, value in stat.items(): + if max_p < value: + bg = bgpath + max_p = value + return bg + + def substract_bg(self, bgpath): + bg = Image.open(bgpath) + img = self.image.convert("P") + + bglut = bg.resize((256, 1)) + bglut.putdata(range(256)) + bglut = list(bglut.convert("RGB").getdata()) + + lut = img.resize((256, 1)) + lut.putdata(range(256)) + lut = list(lut.convert("RGB").getdata()) + + bgpix = bg.load() + pix = img.load() + orgpix = self.image.load() + for x in range(bg.size[0]): + for y in range(bg.size[1]): + rgb_bg = bglut[bgpix[x, y]] + rgb_c = lut[pix[x, y]] + if rgb_c == rgb_bg: + orgpix[x, y] = (255,255,255) + + def eval_black_white(self): + new = Image.new("RGB", (140, 75)) + pix = new.load() + orgpix = self.image.load() + thresh = 4 + for x in range(new.size[0]): + for y in range(new.size[1]): + rgb = orgpix[x, y] + r, g, b = rgb + pix[x, y] = (255,255,255) + if r > max(b, g)+thresh: + pix[x, y] = (0,0,0) + if g < min(r, b): + pix[x, y] = (0,0,0) + if g > max(r, b)+thresh: + pix[x, y] = (0,0,0) + if b > max(r, g)+thresh: + pix[x, y] = (0,0,0) + self.image = new + self.pixels = self.image.load() + + def get_captcha(self, image): + self.load_image(image) + bg = self.get_bg() + self.substract_bg(bg) + self.eval_black_white() + self.to_greyscale() + self.image.save(self.data_dir+"cleaned_pass1.png") + self.clean(4) + self.clean(4) + self.image.save(self.data_dir+"cleaned_pass2.png") + letters = self.split_captcha_letters() + final = "" + for n, letter in enumerate(letters): + self.image = letter + self.image.save(ocr.data_dir+"letter%d.png" % n) + self.run_tesser(True, True, False, False) + final += self.result_captcha + + return final + +if __name__ == '__main__': + import urllib + ocr = LinksaveIn() + testurl = "http://linksave.in/captcha/cap.php?hsh=2229185&code=ZzHdhl3UffV3lXTH5U4b7nShXj%2Bwma1vyoNBcbc6lcc%3D" + urllib.urlretrieve(testurl, ocr.data_dir+"captcha.gif") + + print ocr.get_captcha(ocr.data_dir+'captcha.gif') diff --git a/module/plugins/captcha/MegauploadCom.py b/module/plugins/captcha/MegauploadCom.py new file mode 100644 index 000000000..469ee4094 --- /dev/null +++ b/module/plugins/captcha/MegauploadCom.py @@ -0,0 +1,14 @@ +from captcha import OCR + +class MegauploadCom(OCR): + __name__ = "MegauploadCom" + def __init__(self): + OCR.__init__(self) + + def get_captcha(self, image): + self.load_image(image) + self.run_tesser(True, True, False, True) + return self.result_captcha + +if __name__ == '__main__': + ocr = MegauploadCom() diff --git a/module/plugins/captcha/NetloadIn.py b/module/plugins/captcha/NetloadIn.py new file mode 100644 index 000000000..7f2e6a8d1 --- /dev/null +++ b/module/plugins/captcha/NetloadIn.py @@ -0,0 +1,24 @@ +from captcha import OCR + +class NetloadIn(OCR): + __name__ = "NetloadIn" + def __init__(self): + OCR.__init__(self) + + def get_captcha(self, image): + self.load_image(image) + self.to_greyscale() + self.clean(3) + self.clean(3) + self.run_tesser(True, True, False, False) + + self.result_captcha = self.result_captcha.replace(" ", "")[:4] # cut to 4 numbers + + return self.result_captcha + +if __name__ == '__main__': + import urllib + ocr = NetloadIn() + urllib.urlretrieve("http://netload.in/share/includes/captcha.php", "captcha.png") + + print ocr.get_captcha('captcha.png') diff --git a/module/plugins/captcha/ShareonlineBiz.py b/module/plugins/captcha/ShareonlineBiz.py new file mode 100644 index 000000000..b07fb9b0f --- /dev/null +++ b/module/plugins/captcha/ShareonlineBiz.py @@ -0,0 +1,53 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +# +#Copyright (C) 2009 kingzero, RaNaN +# +#This program is free software; you can redistribute it and/or modify +#it under the terms of the GNU General Public License as published by +#the Free Software Foundation; either version 3 of the License, +#or (at your option) any later version. +# +#This program is distributed in the hope that it will be useful, +#but WITHOUT ANY WARRANTY; without even the implied warranty of +#MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +#See the GNU General Public License for more details. +# +#You should have received a copy of the GNU General Public License +# along with this program; if not, see . +# +### +from captcha import OCR + +class ShareonlineBiz(OCR): + __name__ = "ShareonlineBiz" + + def __init__(self): + OCR.__init__(self) + + def get_captcha(self, image): + self.load_image(image) + self.to_greyscale() + self.image = self.image.resize((160, 50)) + self.pixels = self.image.load() + self.threshold(1.85) + #self.eval_black_white(240) + #self.derotate_by_average() + + letters = self.split_captcha_letters() + + final = "" + for letter in letters: + self.image = letter + self.run_tesser(True, True, False, False) + final += self.result_captcha + + return final + + #tesseract at 60% + +if __name__ == '__main__': + import urllib + ocr = ShareonlineBiz() + urllib.urlretrieve("http://www.share-online.biz/captcha.php", "captcha.jpeg") + print ocr.get_captcha('captcha.jpeg') diff --git a/module/plugins/captcha/__init__.py b/module/plugins/captcha/__init__.py new file mode 100644 index 000000000..e69de29bb diff --git a/module/plugins/captcha/captcha.py b/module/plugins/captcha/captcha.py new file mode 100644 index 000000000..d8c2aa38d --- /dev/null +++ b/module/plugins/captcha/captcha.py @@ -0,0 +1,313 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +# +#Copyright (C) 2009 kingzero, RaNaN +# +#This program is free software; you can redistribute it and/or modify +#it under the terms of the GNU General Public License as published by +#the Free Software Foundation; either version 3 of the License, +#or (at your option) any later version. +# +#This program is distributed in the hope that it will be useful, +#but WITHOUT ANY WARRANTY; without even the implied warranty of +#MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +#See the GNU General Public License for more details. +# +#You should have received a copy of the GNU General Public License +# along with this program; if not, see . +# +### +from __future__ import with_statement +import os +from os.path import join +from os.path import abspath +import logging +import subprocess +#import tempfile +import threading + +import Image +import TiffImagePlugin +import PngImagePlugin +import GifImagePlugin +import JpegImagePlugin + +from module.web.ServerThread import Output + +class OCR(object): + + __name__ = "OCR" + + def __init__(self): + self.logger = logging.getLogger("log") + + def load_image(self, image): + self.image = Image.open(image) + self.pixels = self.image.load() + self.result_captcha = '' + + def unload(self): + """delete all tmp images""" + pass + + def threshold(self, value): + self.image = self.image.point(lambda a: a * value + 10) + + def run(self, command, inputdata=None): + """Run a command""" + + popen = subprocess.Popen(command, bufsize = -1, stdout=subprocess.PIPE, stderr=subprocess.PIPE) + popen.wait() + + def run_tesser(self, subset=False, digits=True, lowercase=True, uppercase=True): + #self.logger.debug("create tmp tif") + + + #tmp = tempfile.NamedTemporaryFile(suffix=".tif") + tmp = open(join("tmp", "tmpTif_%s.tif" % self.__name__), "wb") + tmp.close() + #self.logger.debug("create tmp txt") + #tmpTxt = tempfile.NamedTemporaryFile(suffix=".txt") + tmpTxt = open(join("tmp", "tmpTxt_%s.txt" % self.__name__), "wb") + tmpTxt.close() + + self.logger.debug("save tiff") + self.image.save(tmp.name, 'TIFF') + + if os.name == "nt": + tessparams = [join(pypath,"tesseract","tesseract.exe")] + else: + tessparams = ['tesseract'] + + tessparams.extend( [abspath(tmp.name), abspath(tmpTxt.name).replace(".txt", "")] ) + + if subset and (digits or lowercase or uppercase): + #self.logger.debug("create temp subset config") + #tmpSub = tempfile.NamedTemporaryFile(suffix=".subset") + tmpSub = open(join("tmp", "tmpSub_%s.subset" % self.__name__), "wb") + tmpSub.write("tessedit_char_whitelist ") + if digits: + tmpSub.write("0123456789") + if lowercase: + tmpSub.write("abcdefghijklmnopqrstuvwxyz") + if uppercase: + tmpSub.write("ABCDEFGHIJKLMNOPQRSTUVWXYZ") + tmpSub.write("\n") + tessparams.append("nobatch") + tessparams.append(abspath(tmpSub.name)) + tmpSub.close() + + self.logger.debug("run tesseract") + self.run(tessparams) + self.logger.debug("read txt") + + with open(tmpTxt.name, 'r') as f: + self.result_captcha = f.read().replace("\n", "") + + self.logger.debug(self.result_captcha) + + try: + os.remove(tmp.name) + os.remove(tmpTxt.name) + if subset and (digits or lowercase or uppercase): + os.remove(tmpSub.name) + except: + pass + + def get_captcha(self): + raise NotImplementedError + + def to_greyscale(self): + if self.image.mode != 'L': + self.image = self.image.convert('L') + + self.pixels = self.image.load() + + def eval_black_white(self, limit): + self.pixels = self.image.load() + w, h = self.image.size + for x in xrange(w): + for y in xrange(h): + if self.pixels[x, y] > limit: + self.pixels[x, y] = 255 + else: + self.pixels[x, y] = 0 + + def clean(self, allowed): + pixels = self.pixels + + w, h = self.image.size + + for x in xrange(w): + for y in xrange(h): + if pixels[x, y] == 255: continue + # no point in processing white pixels since we only want to remove black pixel + count = 0 + + try: + if pixels[x-1, y-1] != 255: count += 1 + if pixels[x-1, y] != 255: count += 1 + if pixels[x-1, y + 1] != 255: count += 1 + if pixels[x, y + 1] != 255: count += 1 + if pixels[x + 1, y + 1] != 255: count += 1 + if pixels[x + 1, y] != 255: count += 1 + if pixels[x + 1, y-1] != 255: count += 1 + if pixels[x, y-1] != 255: count += 1 + except: + pass + + # not enough neighbors are dark pixels so mark this pixel + # to be changed to white + if count < allowed: + pixels[x, y] = 1 + + # second pass: this time set all 1's to 255 (white) + for x in xrange(w): + for y in xrange(h): + if pixels[x, y] == 1: pixels[x, y] = 255 + + self.pixels = pixels + + def derotate_by_average(self): + """rotate by checking each angle and guess most suitable""" + + w, h = self.image.size + pixels = self.pixels + + for x in xrange(w): + for y in xrange(h): + if pixels[x, y] == 0: + pixels[x, y] = 155 + + highest = {} + counts = {} + + for angle in range(-45, 45): + + tmpimage = self.image.rotate(angle) + + pixels = tmpimage.load() + + w, h = self.image.size + + for x in xrange(w): + for y in xrange(h): + if pixels[x, y] == 0: + pixels[x, y] = 255 + + + count = {} + + for x in xrange(w): + count[x] = 0 + for y in xrange(h): + if pixels[x, y] == 155: + count[x] += 1 + + sum = 0 + cnt = 0 + + for x in count.values(): + if x != 0: + sum += x + cnt += 1 + + avg = sum / cnt + counts[angle] = cnt + highest[angle] = 0 + for x in count.values(): + if x > highest[angle]: + highest[angle] = x + + highest[angle] = highest[angle] - avg + + hkey = 0 + hvalue = 0 + + for key, value in highest.iteritems(): + if value > hvalue: + hkey = key + hvalue = value + + self.image = self.image.rotate(hkey) + pixels = self.image.load() + + for x in xrange(w): + for y in xrange(h): + if pixels[x, y] == 0: + pixels[x, y] = 255 + + if pixels[x, y] == 155: + pixels[x, y] = 0 + + self.pixels = pixels + + def split_captcha_letters(self): + captcha = self.image + started = False + letters = [] + width, height = captcha.size + bottomY, topY = 0, height + pixels = captcha.load() + + for x in xrange(width): + black_pixel_in_col = False + for y in xrange(height): + if pixels[x, y] != 255: + if started == False: + started = True + firstX = x + lastX = x + + if y > bottomY: bottomY = y + if y < topY: topY = y + if x > lastX: lastX = x + + black_pixel_in_col = True + + if black_pixel_in_col == False and started == True: + rect = (firstX, topY, lastX, bottomY) + new_captcha = captcha.crop(rect) + + w, h = new_captcha.size + if w > 5 and h > 5: + letters.append(new_captcha) + + started = False + bottomY, topY = 0, height + + return letters + + def correct(self, values, var=None): + + if var: + result = var + else: + result = self.result_captcha + + for key, item in values.iteritems(): + + if key.__class__ == str: + result = result.replace(key, item) + else: + for expr in key: + result = result.replace(expr, item) + + if var: + return result + else: + self.result_captcha = result + + +if __name__ == '__main__': + ocr = OCR() + ocr.load_image("B.jpg") + ocr.to_greyscale() + ocr.eval_black_white(140) + ocr.derotate_by_avergage() + ocr.run_gocr() + print "GOCR", ocr.result_captcha + ocr.run_tesser() + print "Tesseract", ocr.result_captcha + ocr.image.save("derotated.jpg") + -- cgit v1.2.3