def pickup(self):
# создаем браузер, которым будем ходить по wmtake.ru
b = Browser()
# сщздаем анализатор, которым будем распознавать капчу
a = Analyzer('wmtake', self.symsize, self.charset)
a.load()
b.show()
log.debug('LOADING PAGE WITH WM BONUS')
b.get('http://fotocity.info/m.base/bonus.php')
while (True):
log.debug('SAVING CAPTCHA')
captcha = b.js('$("#scode-pic img")[0].src')
#b.save(captcha, '/home/polzuka/inspirado/captcha/wmtake/%02d.gif' % i)
log.debug('CAPTCHA TRANSFORMING')
try:
t = Transformer('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 2, 2)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 150,
CV_THRESH_BINARY_INV)
t.contourSplit('breaksplit', t['binarize'], 0.001)
if len(t.symbols) != self.symqty:
raise Exception
except Exception, e:
log.debug(e)
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
log.debug('LOADING PAGE WITH WM BONUS')
b.get('http://wmtake.ru/m.base/bonus.php')
continue
t.normolize('origsplit', 'breaksplit', self.symsize)
symbols = t.slice('origsplit')
log.debug('RECOGNITION CAPTCHA')
code = a.captcha(symbols)
log.debug('ANALYZE RESULT: %s' % colorize(code))
del t
print code
log.debug('FILLING FIELDS')
b.js("$('#scode').val('%s')" % code)
b.js("$('#purse').val('R%s')" % self.purse)
b.js("$('div.news_box div.bn p').click()")
b.sleep(10)
if not b.js("$('#mess-exec:visible').length"):
log.debug('FINISH')
break
log.debug('INCORRECT CAPCTHA RECOGNITION')
log.debug('LOADING PAGE WITH WM BONUS')
b.js("$('#mess-exec p').click()")
def pickup(self):
# создаем браузер, которым будем ходить по wmtake.ru
b = Browser()
# сщздаем анализатор, которым будем распознавать капчу
a = Analyzer(self.site, self.symsize, self.charset)
a.load()
b.show()
log.debug('LOADING PAGE WITH WM BONUS')
b.get('http://wmtake.ru/m.base/bonus.php')
while(True):
log.debug('SAVING CAPTCHA')
captcha = b.js('$("#scode-pic img")[0].src')
#b.save(captcha, '/home/polzuka/inspirado/captcha/wmtake/%02d.gif' % i)
log.debug('CAPTCHA TRANSFORMING')
try:
t = Transformer('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 2, 2)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 150, CV_THRESH_BINARY_INV)
t.contourSplit('breaksplit', t['binarize'], 0.001)
if len(t.symbols) != self.symqty:
raise Exception
except Exception, e:
log.debug(e)
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
log.debug('LOADING PAGE WITH WM BONUS')
b.get('http://wmtake.ru/m.base/bonus.php')
continue
t.normolize('origsplit', 'breaksplit', self.symsize)
symbols = t.slice('origsplit')
log.debug('RECOGNITION CAPTCHA')
code = a.captcha(symbols)
log.debug('ANALYZE RESULT: %s' % colorize(code))
del t
print code
log.debug('FILLING FIELDS')
b.js("$('#scode').val('%s')" % code)
b.js("$('#purse').val('R%s')" % self.purse)
b.js("$('div.news_box div.bn p').click()")
b.sleep(10)
if not b.js("$('#mess-exec:visible').length"):
log.debug('FINISH')
break
log.debug('INCORRECT CAPCTHA RECOGNITION')
log.debug('LOADING PAGE WITH WM BONUS')
b.js("$('#mess-exec p').click()")
def pickup(self):
# создаем браузер, которым будем ходить по wmtake.ru
b = Browser()
# создаем анализатор, которым будем распознавать капчу
a = Analyzer(self.site, self.symsize, self.charset)
a.load()
b.show()
while (True):
log.debug('LOADING PAGE WITH WM BONUS')
b.get('http://exchangecity.ru/?cmd=bonus')
log.debug('SAVING CAPTCHA')
captcha = 'http://exchangecity.ru/include/anti_robot.php'
#b.save(captcha, '/home/polzuka/inspirado/captcha/wmtake/%02d.gif' % i)
log.debug('CAPTCHA TRANSFORMING')
t = Transformer('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 2, 2)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 200, CV_THRESH_BINARY_INV)
t.contourSplit('breaksplit', t['binarize'], 0.001)
if len(t.symbols) != self.symqty:
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
continue
t.normolize('origsplit', 'breaksplit', self.symsize)
symbols = t.slice('origsplit')
log.debug('RECOGNITION CAPTCHA')
code = a.captcha(symbols)
log.debug('ANALYZE RESULT: %s' % colorize(code))
del t
print code
log.debug('FILLING FIELDS')
b.js("$('input[name = img]').val('%s')" % code)
b.js("$('input[name = WALLET_BONUS]').val('R%s')" % self.purse)
b.js("$('input[name = get_bonus]').click()")
b.sleep(1)
if not b.js(
"$('font.textbl:contains(Вы получили бонус в размере)').length"
):
log.debug('FINISH')
break
log.debug('INCORRECT CAPCTHA RECOGNITION')
self.quit()
def pickup(self):
# создаем браузер, которым будем ходить по wmtake.ru
b = Browser()
# создаем анализатор, которым будем распознавать капчу
a = Analyzer(self.site, self.symsize, self.charset)
a.load()
b.show()
while(True):
log.debug('LOADING PAGE WITH WM BONUS')
b.get('http://exchangecity.ru/?cmd=bonus')
log.debug('SAVING CAPTCHA')
captcha = 'http://exchangecity.ru/include/anti_robot.php'
#b.save(captcha, '/home/polzuka/inspirado/captcha/wmtake/%02d.gif' % i)
log.debug('CAPTCHA TRANSFORMING')
t = Transformer('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 2, 2)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 200, CV_THRESH_BINARY_INV)
t.contourSplit('breaksplit', t['binarize'], 0.001)
if len(t.symbols) != self.symqty:
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
continue
t.normolize('origsplit', 'breaksplit', self.symsize)
symbols = t.slice('origsplit')
log.debug('RECOGNITION CAPTCHA')
code = a.captcha(symbols)
log.debug('ANALYZE RESULT: %s' % colorize(code))
del t
print code
log.debug('FILLING FIELDS')
b.js("$('input[name = img]').val('%s')" % code)
b.js("$('input[name = WALLET_BONUS]').val('R%s')" % self.purse)
b.js("$('input[name = get_bonus]').click()")
b.sleep(1)
if not b.js("$('font.textbl:contains(Вы получили бонус в размере)').length"):
log.debug('FINISH')
break
log.debug('INCORRECT CAPCTHA RECOGNITION')
self.quit()
def train(self):
log.info('prepare(): trainpath: %s, netpath: %s' % colorize(
(self.trainpath, self.netpath)))
self.ann.train_on_file(self.trainpath, self.maxEpochs,
self.epochsBetweenReports, self.desiredError)
self.ann.save(self.netpath)
def prepare(self):
log.info('prepare(): trainpath: %s, symbolpath: %s' % colorize(
(self.trainpath, self.symbolpath)))
trainfile = open(self.trainpath, 'w')
filter = re.compile(SYMBOL_FILTER)
symbols = []
for name in os.listdir(self.symbolpath):
if filter.match(name):
symbols.append(name)
trainfile.write('%d %d %d\n' % (len(symbols), self.input, self.output))
for name in symbols:
img = cvLoadImage(SYMBOLS_PATH + self.site + '/' + name,
CV_LOAD_IMAGE_GRAYSCALE)
data = self.getdata(img)
trainfile.write('%s\n' % ' '.join(map(str, data)))
c = name[0]
n = self.charset.index(c)
trainfile.write('-1 ' * n + '1' + ' -1' * (self.output - n - 1) +
'\n')
trainfile.close()
def __init__(self, site, picsize, charset):
self.site = site
self.trainpath = TRAINS_PATH + self.site + '.trn'
self.netpath = NETS_PATH + self.site + '.ann'
self.symbolpath = SYMBOLS_PATH + self.site
self.charset = charset
self.input = picsize[0] * picsize[1]
self.output = len(charset)
self.hidden = self.input / 3
self.layers = 3
self.desiredError = 0.00006
self.maxEpochs = 50000
self.epochsBetweenReports = 1000
self.ann = libfann.neural_net()
self.ann.create_sparse_array(self.layers,
(self.input, self.hidden, self.output))
self.ann.set_activation_function_hidden(
libfann.SIGMOID_SYMMETRIC_STEPWISE)
self.ann.set_activation_function_output(
libfann.SIGMOID_SYMMETRIC_STEPWISE)
log.info(
'init(): site: %s, size: %sx%s, charset: %s, input: %s, hidden: %s, output: %s'
% colorize((site, picsize[0], picsize[1], charset, self.input,
self.hidden, self.output)))
def captcha(self, symbols):
result = ''
for smb in symbols:
data = self.getdata(smb)
result += self.symbol(data)
log.info('captcha(): result: %s' % colorize(result))
return result
def collect(self):
b = Browser()
for i in xrange(200):
log.info('LOAD PAGE WITH CAPTCHA')
b.get('http://sms-webmoney.ru/')
captcha = 'http://sms-webmoney.ru/img.php'
b.save(captcha, CAPTCHA_PATH + self.site + '/%02d.png' % i)
t = Transformer()
t.load('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 3, 3)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 200, CV_THRESH_BINARY_INV)
radius = 2
kernel = cvCreateStructuringElementEx(radius * 2 + 1, radius * 2 + 1, radius, radius, CV_SHAPE_ELLIPSE)
t.morphology('morphology', t['binarize'], 0, 1, kernel)
t.contourSplit('breaksplit', t['morphology'], 0.01)
if len(t.symbols) != self.symqty:
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
continue
t.normolize('origsplit', 'breaksplit', self.symsize)
t.savesymbols('origsplit', SYMBOLS_PATH + self.site, '%02d' % i)
del t
def collect(self):
b = Browser()
for i in xrange(100):
log.info('LOAD PAGE WITH CAPTCHA')
b.get('http://exchangecity.ru/?cmd=bonus')
captcha = 'http://exchangecity.ru/include/anti_robot.php'
b.save(captcha, CAPTCHA_PATH + Implem.name + '/%02d.png' % i)
t = Transformer()
t.load('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 2, 2)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 200, CV_THRESH_BINARY_INV)
'''
radius = 3
kernel = cvCreateStructuringElementEx(radius * 2 + 1, radius * 2 + 1, radius, radius, CV_SHAPE_ELLIPSE)
t.morphology('morphology', t['binarize'], 1, 1, kernel)
'''
t.contourSplit('breaksplit', t['binarize'], 0.001)
if len(t.symbols) != self.symqty:
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
continue
t.normolize('origsplit', 'breaksplit', Implem.size)
t.savesymbols('origsplit', SYMBOLS_PATH + Implem.name, '%02d' % i)
del t
def slice(self, srckey):
log.debug('slice(): srckey: %s, len: %s' % colorize(
(srckey, len(self.symbols))))
res = []
for smb in self.symbols:
res.append(smb[srckey])
return res
def captcha(self, symbols):
result = ''
for smb in symbols:
data = self.getdata(smb)
result += self.symbol(data)
log.info('captcha(): result: %s' % colorize(result))
return result
def collect(self):
b = Browser()
for i in xrange(200):
log.info('LOAD PAGE WITH CAPTCHA')
b.get('http://sms-webmoney.ru/')
captcha = 'http://sms-webmoney.ru/img.php'
b.save(captcha, CAPTCHA_PATH + self.site + '/%02d.png' % i)
t = Transformer()
t.load('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 3, 3)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 200, CV_THRESH_BINARY_INV)
radius = 2
kernel = cvCreateStructuringElementEx(radius * 2 + 1,
radius * 2 + 1, radius,
radius, CV_SHAPE_ELLIPSE)
t.morphology('morphology', t['binarize'], 0, 1, kernel)
t.contourSplit('breaksplit', t['morphology'], 0.01)
if len(t.symbols) != self.symqty:
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
continue
t.normolize('origsplit', 'breaksplit', self.symsize)
t.savesymbols('origsplit', SYMBOLS_PATH + self.site, '%02d' % i)
del t
def collect(self):
b = Browser()
for i in xrange(100):
log.info('LOAD PAGE WITH CAPTCHA')
b.get('http://exchangecity.ru/?cmd=bonus')
captcha = 'http://exchangecity.ru/include/anti_robot.php'
b.save(captcha, CAPTCHA_PATH + Implem.name + '/%02d.png' % i)
t = Transformer()
t.load('orig', b.image(captcha))
t.resizeby('resize', t['orig'], 2, 2)
t.grayscale('grayscale', t['resize'], 2)
t.binarize('binarize', t['grayscale'], 200, CV_THRESH_BINARY_INV)
'''
radius = 3
kernel = cvCreateStructuringElementEx(radius * 2 + 1, radius * 2 + 1, radius, radius, CV_SHAPE_ELLIPSE)
t.morphology('morphology', t['binarize'], 1, 1, kernel)
'''
t.contourSplit('breaksplit', t['binarize'], 0.001)
if len(t.symbols) != self.symqty:
log.debug(colorize('INCORRECT SYMBOL NUMBER', RED))
continue
t.normolize('origsplit', 'breaksplit', Implem.size)
t.savesymbols('origsplit', SYMBOLS_PATH + Implem.name, '%02d' % i)
del t
def normolize(self, dstkey, srckey, normsize):
log.debug('normolize(): dstkey: %s, srckey: %s, size: %sx%s' % colorize((dstkey, srckey, normsize[0], normsize[1])))
for smb in self.symbols:
img = smb.transforms[srckey]
size = cvGetSize(img)
imgX = size.width
imgY = size.height
smb.resizeto(dstkey, img, normsize[0], normsize[1])
def save(self, url, path):
log.info('save(): %s to %s' % (colorize((url, path))))
file = QFile(path)
cached = self.cache.data(QUrl(url)).readAll()
if file.open(QIODevice.WriteOnly):
file.write(cached)
file.close()
return path
def normolize(self, dstkey, srckey, normsize):
log.debug('normolize(): dstkey: %s, srckey: %s, size: %sx%s' %
colorize((dstkey, srckey, normsize[0], normsize[1])))
for smb in self.symbols:
img = smb.transforms[srckey]
size = cvGetSize(img)
imgX = size.width
imgY = size.height
smb.resizeto(dstkey, img, normsize[0], normsize[1])
def savesymbols(self, srckey, smbdir, ser):
log.debug('saveSymbols(): %s' % colorize(srckey))
savedir = os.getcwd()
os.chdir(smbdir)
i = 0
for smb in self.symbols:
img = smb.transforms[srckey]
self.save(img, '_%d_%s.png' % (i, ser))
i += 1
os.chdir(savedir)
def getdata(self, img):
log.debug('getdata(): size: %sx%s' % colorize((img.width, img.height)))
data = []
for x in range(img.width):
for y in range(img.height):
data.append(1 if img[y, x] > 0 else 0)
'''
for i in xrange(img.height):
print data[i * img.width : (i + 1) * img.width]
print
'''
return data
def savesymbols(self, srckey, smbdir, ser):
log.debug('saveSymbols(): %s' % colorize(srckey))
savedir = os.getcwd()
os.chdir(smbdir)
i = 0
for smb in self.symbols:
img = smb.transforms[srckey]
self.save(img, '_%d_%s.png' % (i, ser))
i += 1
os.chdir(savedir)
def getdata(self, img):
log.debug('getdata(): size: %sx%s' % colorize((img.width, img.height)))
data = []
for x in range(img.width):
for y in range(img.height):
data.append(1 if img[y, x] > 0 else 0)
'''
for i in xrange(img.height):
print data[i * img.width : (i + 1) * img.width]
print
'''
return data
def js(self, script):
# выполняем js
log.debug('js(): evalute %s' % colorize(script))
result = self.mainFrame().evaluateJavaScript(script).toString()
log.debug('js(): result %s' % colorize(result))
# подождем немножко, вдруг прилетит сигнал о начале загрузки
if self.waiter.wait(self.loadStarted, 30):
# началась загрузка, ждем ее окончания за разумное время
log.debug('js(): loading start')
if self.waiter.wait(self.loadFinished, self.timeoutsec * 1000):
log.debug('js(): loading fifnish')
else:
log.debug('js(): loading timeout')
# подгружаем jQuery, если указано
if self.autojq:
self.jquerify()
return result
def morphology(self, key, src, method, iterations=1, kernel=None):
log.debug('morphology(): %s' % colorize(key))
tmp = cvCreateImage(cvGetSize(src), src.depth, src.nChannels)
if not kernel:
kernel = cvCreateStructuringElementEx(3, 3, 1, 1, CV_SHAPE_ELLIPSE)
if method == 0:
cvErode(src, tmp, kernel, iterations)
self.transforms[key] = tmp
elif method == 1:
cvDilate(src, tmp, kernel, iterations)
self.transforms[key] = tmp
elif 2 <= method <= 6:
res = cvCreateImage(cvGetSize(src), src.depth, src.nChannels)
cvMorphologyEx(src, res, tmp, kernel, method, iterations)
self.transforms[key] = res
def morphology(self, key, src, method, iterations=1, kernel=None):
log.debug('morphology(): %s' % colorize(key))
tmp = cvCreateImage(cvGetSize(src), src.depth, src.nChannels)
if not kernel:
kernel = cvCreateStructuringElementEx(3, 3, 1, 1, CV_SHAPE_ELLIPSE)
if method == 0:
cvErode(src, tmp, kernel, iterations)
self.transforms[key] = tmp
elif method == 1:
cvDilate(src, tmp, kernel, iterations)
self.transforms[key] = tmp
elif 2 <= method <= 6:
res = cvCreateImage(cvGetSize(src), src.depth, src.nChannels)
cvMorphologyEx(src, res, tmp, kernel, method, iterations)
self.transforms[key] = res
def __init__(self, autojq=True, timeoutsec=120):
QWebPage.__init__(self)
self.autojq = autojq
self.timeoutsec = timeoutsec
self.loop = QEventLoop()
self.waiter = Waiter()
# фиксенные кеш и менеджер
self.cache = Cache()
self.manager = Manager()
self.cache.setCacheDirectory(CACHE_PATH)
self.manager.setCache(self.cache)
self.setNetworkAccessManager(self.manager)
log.info('init(): autojq: %s, timeoutsec: %s, cache: %s' % colorize((autojq, timeoutsec, CACHE_PATH)))
def __init__(self, site, picsize, charset):
self.site = site
self.trainpath = TRAINS_PATH + self.site + '.trn'
self.netpath = NETS_PATH + self.site + '.ann'
self.symbolpath = SYMBOLS_PATH + self.site
self.charset = charset
self.input = picsize[0] * picsize[1]
self.output = len(charset)
self.hidden = self.input / 3
self.layers = 3
self.desiredError = 0.00006
self.maxEpochs = 50000
self.epochsBetweenReports = 1000
self.ann = libfann.neural_net()
self.ann.create_sparse_array(self.layers, (self.input, self.hidden, self.output))
self.ann.set_activation_function_hidden(libfann.SIGMOID_SYMMETRIC_STEPWISE)
self.ann.set_activation_function_output(libfann.SIGMOID_SYMMETRIC_STEPWISE)
log.info('init(): site: %s, size: %sx%s, charset: %s, input: %s, hidden: %s, output: %s' %
colorize((site, picsize[0], picsize[1], charset, self.input, self.hidden, self.output)))
def prepare(self):
log.info('prepare(): trainpath: %s, symbolpath: %s' % colorize((self.trainpath, self.symbolpath)))
trainfile = open(self.trainpath, 'w')
filter = re.compile(SYMBOL_FILTER)
symbols = []
for name in os.listdir(self.symbolpath):
if filter.match(name):
symbols.append(name)
trainfile.write('%d %d %d\n' % (len(symbols), self.input, self.output))
for name in symbols:
img = cvLoadImage(SYMBOLS_PATH + self.site + '/' + name, CV_LOAD_IMAGE_GRAYSCALE)
data = self.getdata(img)
trainfile.write('%s\n' % ' '.join(map(str, data)))
c = name[0]
n = self.charset.index(c)
trainfile.write('-1 ' * n + '1' + ' -1' * (self.output - n - 1) + '\n')
trainfile.close()
def resizeby(self, key, src, scaleX, scaleY, method=1):
log.debug('resizeby(): %s' % colorize(key))
res = cvCreateImage((src.width * scaleX, src.height * scaleY), src.depth, src.nChannels)
cvResize(src, res, method)
self.transforms[key] = res
def resizeto(self, key, src, resultX, resultY, method=1):
log.debug('resizeto(): %s' % colorize(key))
res = cvCreateImage((resultX, resultY), src.depth, src.nChannels)
cvResize(src, res, method)
self.transforms[key] = res
def grayscale(self, key, src, flags=0):
log.debug('garyscale(): %s' % colorize(key))
res = cvCreateImage(cvGetSize(src), src.depth, 1)
cvConvertImage(src, res, flags)
self.transforms[key] = res
def binarize(self, key, src, threshold, method):
log.debug('binarize(): %s' % colorize(key))
res = cvCreateImage(cvGetSize(src), IPL_DEPTH_8U, 1)
cvThreshold(src, res, threshold, 255, method)
self.transforms[key] = res
def clone(self, key, src):
log.debug('clone(): %s' % colorize(key))
self.transforms[key] = cvCloneImage(src)
def resizeby(self, key, src, scaleX, scaleY, method=1):
log.debug('resizeby(): %s' % colorize(key))
res = cvCreateImage((src.width * scaleX, src.height * scaleY),
src.depth, src.nChannels)
cvResize(src, res, method)
self.transforms[key] = res
def grayscale(self, key, src, flags=0):
log.debug('garyscale(): %s' % colorize(key))
res = cvCreateImage(cvGetSize(src), src.depth, 1)
cvConvertImage(src, res, flags)
self.transforms[key] = res
def train(self):
log.info('prepare(): trainpath: %s, netpath: %s' % colorize((self.trainpath, self.netpath)))
self.ann.train_on_file(self.trainpath, self.maxEpochs, self.epochsBetweenReports, self.desiredError)
self.ann.save(self.netpath)
def load(self):
self.ann.create_from_file(self.netpath)
log.info('load(): netpath: %s' % colorize(self.netpath))
def slice(self, srckey):
log.debug('slice(): srckey: %s, len: %s' % colorize((srckey, len(self.symbols))))
res = []
for smb in self.symbols:
res.append(smb[srckey])
return res
def load(self, key, src):
self.transforms[key] = QIm2Ipl(src)
log.debug('load(): %s' % colorize(key))
def save(self, src, path):
log.debug('save(): %s' % colorize(path))
cvSaveImage(path, src)
def open(self, key, path):
log.debug('open(): %s from %s' % (colorize(key, path)))
self.transforms[key] = cvLoadImage(path)
def resizeto(self, key, src, resultX, resultY, method=1):
log.debug('resizeto(): %s' % colorize(key))
res = cvCreateImage((resultX, resultY), src.depth, src.nChannels)
cvResize(src, res, method)
self.transforms[key] = res
def get(self, url, headers=None, pull=None):
log.info('get(): url: %s' % colorize(url))
self.js('window.location = "%s"' % url) # идем на url
def breakSplit(self, key, src, threshold=0):
log.debug('breakSplit(): %s' % colorize(key))
storage = cvCreateMemStorage(0)
res = cvCloneImage(src)
# алгоритм, не поддающийся документированию и сопровождению
# тем не менее
leps = [] # переходы
prev = 0 # предполагаем, что слева пустое пространство (ничего страшного, если это не так)
# для каждого столбца считаем количество белых пикселей
for y in xrange(src.cols):
qty = 0.0
for x in xrange(src.rows):
qty += 1 if src[x][y] > 0 else 0
# если процент белых пикселей в столбце превышает пороговое значение, считаем что там символ
curr = 1 if qty / src.rows > threshold else 0
# фиксируем переходы
if prev != curr:
leps.append(y)
prev = curr
# корректный последний переход
if len(leps) % 2 != 0:
leps.append(src.cols - 1)
boundPairs = [] # границы
for i in xrange(0, len(leps), 2):
boundPairs.append((leps[i], leps[i + 1]))
# отображение рамок
for bounds in boundPairs:
p1 = (bounds[0], 0)
p2 = (bounds[1], src.rows - 1)
cvRectangle(res, p1, p2, gray)
self.transforms[key] = res
# разбиение на символы
for bounds in boundPairs:
rect = (bounds[0], 0, bounds[1] - bounds[0], src.rows)
roi = cvGetSubRect(src, rect)
tmp1 = cvCreateImage(cvGetSize(roi), IPL_DEPTH_8U, 1)
tmp2 = cvCreateImage(cvGetSize(roi), IPL_DEPTH_8U, 1)
cvCopy(roi, tmp1)
cvCopy(roi, tmp2)
num, contours = cvFindContours(tmp1, storage, sizeof_CvContour, CV_RETR_EXTERNAL, CV_CHAIN_APPROX_SIMPLE, (0, 0))
if num == 0:
raise TransformError
# выбираем самый крупный контур
saveArea = 0
for contour in contours.hrange():
currArea = cvContourArea(contour)
if currArea > saveArea:
saveArea = currArea
saveCont = contour
# и вычленяем его
rect = cvBoundingRect(saveCont)
roi = cvGetSubRect(tmp2, rect)
smb = cvCreateImage(cvGetSize(roi), IPL_DEPTH_8U, 1)
cvCopy(roi, smb)
self.symbols.append(Transformer(key, Ipl2QIm(smb)))
def image(self, url):
log.info('image(): from %s' % colorize(url))
cached = self.cache.data(QUrl(url))
img = QImage()
img.loadFromData(cached.readAll())
return img
def load(self, key, src):
self.transforms[key] = QIm2Ipl(src)
log.debug('load(): %s' % colorize(key))
def breakSplit(self, key, src, threshold=0):
log.debug('breakSplit(): %s' % colorize(key))
storage = cvCreateMemStorage(0)
res = cvCloneImage(src)
# алгоритм, не поддающийся документированию и сопровождению
# тем не менее
leps = [] # переходы
prev = 0 # предполагаем, что слева пустое пространство (ничего страшного, если это не так)
# для каждого столбца считаем количество белых пикселей
for y in xrange(src.cols):
qty = 0.0
for x in xrange(src.rows):
qty += 1 if src[x][y] > 0 else 0
# если процент белых пикселей в столбце превышает пороговое значение, считаем что там символ
curr = 1 if qty / src.rows > threshold else 0
# фиксируем переходы
if prev != curr:
leps.append(y)
prev = curr
# корректный последний переход
if len(leps) % 2 != 0:
leps.append(src.cols - 1)
boundPairs = [] # границы
for i in xrange(0, len(leps), 2):
boundPairs.append((leps[i], leps[i + 1]))
# отображение рамок
for bounds in boundPairs:
p1 = (bounds[0], 0)
p2 = (bounds[1], src.rows - 1)
cvRectangle(res, p1, p2, gray)
self.transforms[key] = res
# разбиение на символы
for bounds in boundPairs:
rect = (bounds[0], 0, bounds[1] - bounds[0], src.rows)
roi = cvGetSubRect(src, rect)
tmp1 = cvCreateImage(cvGetSize(roi), IPL_DEPTH_8U, 1)
tmp2 = cvCreateImage(cvGetSize(roi), IPL_DEPTH_8U, 1)
cvCopy(roi, tmp1)
cvCopy(roi, tmp2)
num, contours = cvFindContours(tmp1, storage, sizeof_CvContour,
CV_RETR_EXTERNAL,
CV_CHAIN_APPROX_SIMPLE, (0, 0))
if num == 0:
raise TransformError
# выбираем самый крупный контур
saveArea = 0
for contour in contours.hrange():
currArea = cvContourArea(contour)
if currArea > saveArea:
saveArea = currArea
saveCont = contour
# и вычленяем его
rect = cvBoundingRect(saveCont)
roi = cvGetSubRect(tmp2, rect)
smb = cvCreateImage(cvGetSize(roi), IPL_DEPTH_8U, 1)
cvCopy(roi, smb)
self.symbols.append(Transformer(key, Ipl2QIm(smb)))
def jquerify(self):
log.debug('jquerify(): load from %s' % colorize(JQUERY_PATH))
jquery = open(JQUERY_PATH).read()
self.mainFrame().evaluateJavaScript(jquery)
def save(self, src, path):
log.debug('save(): %s' % colorize(path))
cvSaveImage(path, src);
def load(self):
self.ann.create_from_file(self.netpath)
log.info('load(): netpath: %s' % colorize(self.netpath))
def open(self, key, path):
log.debug('open(): %s from %s' % (colorize(key, path)))
self.transforms[key] = cvLoadImage(path)
def clone(self, key, src):
log.debug('clone(): %s' % colorize(key))
self.transforms[key] = cvCloneImage(src)
def binarize(self, key, src, threshold, method):
log.debug('binarize(): %s' % colorize(key))
res = cvCreateImage(cvGetSize(src), IPL_DEPTH_8U, 1)
cvThreshold(src, res, threshold, 255, method)
self.transforms[key] = res