def random_captcha_text(char_set=captcha_params.get_char_set(),
captcha_size=captcha_params.get_captcha_size()):
captcha_text = []
for i in range(captcha_size):
c = random.choice(char_set)
captcha_text.append(c)
return captcha_text
def decodeLabel2(resu):
for i in range(len(resu)):
text = ''
predict2 = []
for j in range(MAX_CAPTCHA):
char_index = get_max(predict[i,CHAR_SET_LEN*j:(j+1)*CHAR_SET_LEN])
char = captcha_params.get_char_set()[char_index]
predict2.append(char)
text = text.join(predict2)
print(text)
def random_text(characters=captcha_params.get_char_set(),
captcha_size=captcha_params.get_captcha_size()):
captcha_text = []
for i in range(captcha_size):
char = random.choice(characters)
captcha_text.append(str(char))
captchaTextNoSpace = ''.join(captcha_text)
count_char(captchaTextNoSpace)
print('no space')
print(captchaTextNoSpace)
return captchaTextNoSpace
def predict_from_img(self, img):
X_test = get_x_input_from_image(img)
predict = self.model.predict(X_test)
text = ''
for i in range(X_test.shape[0]):
predict2 = []
for j in range(MAX_CAPTCHA):
char_index = get_max(predict[i, CHAR_SET_LEN * j:(j + 1) *
CHAR_SET_LEN])
char = captcha_params.get_char_set()[char_index]
predict2.append(char)
text = text.join(predict2)
return text
def predict_from_img(self, img):
text = ''
for part in range(captcha_params.get_captcha_nb_letters() -
1): # -1 since last is a border
if (part == 0):
continue # part °0 = border , ignored
x_test = get_x_input_from_image(img, part)
predict = self.model.predict(x_test)
# print (predict)
# decodeLabel(predict)
for i in range(x_test.shape[0]):
true = []
predict2 = []
for j in range(MAX_CAPTCHA):
char_index = get_max(predict[i, CHAR_SET_LEN * j:(j + 1) *
CHAR_SET_LEN])
char = captcha_params.get_char_set()[char_index]
predict2.append(char)
text = text + text.join(predict2)
return text
from captcha.image import ImageCaptcha # pip install captcha
import numpy as np
import matplotlib.pyplot as plt # show captcha
from PIL import Image
import random
import os
import captcha_params
import string
import collections
import plotly.plotly as py
#ordered dict
#0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ
character_set = captcha_params.get_char_set()
#d={}
character_count = collections.OrderedDict()
print(" Char Set: ", character_set)
for c in character_set:
character_count[c] = 0
#0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ
#counts the occurences of all characters (in order)
print("Char Count: ", character_count)
def count_char(captcha):
for char in captcha:
character_count[char] += 1
def random_text(characters=captcha_params.get_char_set(),
captcha_size=captcha_params.get_captcha_size()):
import os
from PIL import Image
import numpy as np
import random
import captcha_params
np.random.seed(177)
MAX_CAPTCHA = captcha_params.get_captcha_size()
CHAR_SET_LEN = captcha_params.get_char_set_len()
CHAR_SET = captcha_params.get_char_set()
Y_LEN = captcha_params.get_y_len()
height = captcha_params.get_height()
width = captcha_params.get_width()
# return the index of the max_num in the array
def get_max(array):
max_num = max(array)
for i in range(len(array)):
if array[i] == max_num:
return i
def get_text(array):
text = []
max_num = max(array)
for i in range(len(array)):
model = load_model.get_model(input_shape)
model.compile(loss='categorical_crossentropy',
optimizer='adadelta',
metrics=['accuracy'])
score = model.evaluate(X_test, Y_test, verbose=0)
predict = model.predict(X_test, batch_size=batch_size, verbose=0)
# calculate the accuracy with the test data
acc = 0
expected_str = []
predicted_str = []
labels = captcha_params.get_char_set()
for i in range(X_test.shape[0]):
# Verify old captch
if i % 5 == 0 and i >= 5:
if expected_str[i - 5:i] == predicted_str[i - 5:i]:
acc += 1
if i % 5 == 0 and i >= 5 and i <= 50:
print(i, ' true: ', expected_str[i - 5:i])
print(i, ' predict: ', predicted_str[i - 5:i])
# Find out the captcha size
len_captcha = MAX_CAPTCHA
for j in reversed(range(MAX_CAPTCHA)):
char_vec_true = Y_test[i, CHAR_SET_LEN * j:(j + 1) * CHAR_SET_LEN]
char_true_idx = get_max(char_vec_true)