人工智能项目资料-基于机器学习的登录验证码爆破工具.zip

import numpy as np import sys import tensorflow as tf import cv2 import os import random import time # 验证码中的字符 number = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9'] alphabet = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z'] ALPHABET = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z'] class Model(object): def __init__(self,MAX_CAPTCHA,Is_alphabet,IMAGE_HEIGHT,IMAGE_WIDTH,model_path, train_image_path=None,valid_image_path=None): self.model_path = model_path self.train_path = train_image_path self.valid_path = valid_image_path self.MAX_CAPTCHA=MAX_CAPTCHA self.char_set=self.get_char_set(Is_alphabet) self.IMAGE_WIDTH=IMAGE_WIDTH self.IMAGE_HEIGHT=IMAGE_HEIGHT self.CHAR_SET_LEN=len(self.char_set) self.X = tf.placeholder(tf.float32, [None, self.IMAGE_HEIGHT * self.IMAGE_WIDTH]) self.Y = tf.placeholder(tf.float32, [None, self.CHAR_SET_LEN]) self.keep_prob = tf.placeholder(tf.float32) # dropout self.image_filename_list, self.total = self.get_image_file_name(self.train_path) self.image_filename_list_valid, self.valid_total = self.get_image_file_name(self.valid_path) self.output = self.crack_captcha_cnn() #根据是否包含英文字母获取字符集. def get_char_set(self,Is_alphabet): char_set=[] if(Is_alphabet==0): char_set=number else: char_set=number+alphabet+ALPHABET return char_set def get_image_file_name(self,imgFilePath): fileName = [] total = 0 for filePath in os.listdir(imgFilePath): captcha_name = filePath.split('/')[-1] fileName.append(captcha_name) total += 1 random.seed(time.time()) random.shuffle(fileName) return fileName, total # 构建网络，定义ｃｎｎ网络 def crack_captcha_cnn(self,w_alpha=0.01, b_alpha=0.1): x = tf.reshape(self.X, shape=[-1, self.IMAGE_HEIGHT, self.IMAGE_WIDTH, 1]) w_c1 = tf.Variable(w_alpha * tf.random_normal([3, 3, 1, 32])) w_c1 = tf.Variable(w_alpha * tf.random_normal([3, 3, 1, 32])) b_c1 = tf.Variable(b_alpha * tf.random_normal([32])) conv1 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(x, w_c1, strides=[1, 1, 1, 1], padding='SAME'), b_c1)) conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME') conv1 = tf.nn.dropout(conv1, self.keep_prob) w_c2 = tf.Variable(w_alpha * tf.random_normal([3, 3, 32, 64])) b_c2 = tf.Variable(b_alpha * tf.random_normal([64])) conv2 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(conv1, w_c2, strides=[1, 1, 1, 1], padding='SAME'), b_c2)) conv2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME') conv2 = tf.nn.dropout(conv2, self.keep_prob) w_c3 = tf.Variable(w_alpha * tf.random_normal([3, 3, 64, 64])) b_c3 = tf.Variable(b_alpha * tf.random_normal([64])) conv3 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(conv2, w_c3, strides=[1, 1, 1, 1], padding='SAME'), b_c3)) conv3 = tf.nn.max_pool(conv3, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME') conv3 = tf.nn.dropout(conv3, self.keep_prob) w_d = tf.Variable(w_alpha * tf.random_normal([5 * 15 * 64, 1024])) b_d = tf.Variable(b_alpha * tf.random_normal([1024])) dense = tf.reshape(conv3, [-1, w_d.get_shape().as_list()[0]]) dense = tf.nn.relu(tf.add(tf.matmul(dense, w_d), b_d)) dense = tf.nn.dropout(dense, self.keep_prob) w_out = tf.Variable(w_alpha * tf.random_normal([1024, self.CHAR_SET_LEN])) b_out = tf.Variable(b_alpha * tf.random_normal([self.CHAR_SET_LEN])) out = tf.add(tf.matmul(dense, w_out), b_out) return out def predict_builder(self): self.sess = tf.Session() saver = tf.train.Saver() saver.restore(self.sess, tf.train.latest_checkpoint(self.model_path)) predict = tf.reshape(self.output, [-1, 1, self.CHAR_SET_LEN]) self.max_idx_p = tf.argmax(predict, 2) pass def predict_image_num(self, predict_image_path): #print("predict_image_path"+predict_image_path) result="" if not os.path.isfile(predict_image_path): print('file %s is not exist' % predict_image_path) return else: result = self.get_result(self.sess, predict_image_path, self.max_idx_p) return result return def predict_image(self, predict_image_path): with tf.Session() as sess: saver = tf.train.Saver() saver.restore(sess, tf.train.latest_checkpoint(self.model_path)) predict = tf.reshape(self.output, [-1, 1, self.CHAR_SET_LEN]) max_idx_p = tf.argmax(predict, 2) if not os.path.isdir(predict_image_path): print("test_path is not dir") return while True: image_name = input("input image name:") image_path = os.path.join(predict_image_path, image_name) image_path = image_path.replace('\\', '/') print("file path: %s" % image_path) if not os.path.isfile(image_path): print('file %s is not exist' % image_path) else: result = self.get_result(sess, image_path, max_idx_p) print(result) def get_result(self, sess, predict_image_path, max_idx_p): captcha_image = self.get_image(predict_image_path) result = "" for i in range(self.MAX_CAPTCHA): batch_x_predict = captcha_image[i] text_list = sess.run(max_idx_p, feed_dict={self.X: [batch_x_predict], self.keep_prob: 1}) text = text_list[0].tolist() vector = np.zeros(self.CHAR_SET_LEN) j = 0 for n in text: vector[j * self.CHAR_SET_LEN + n] = 1 j += 1 predict_text = self.vec2text(vector) result = result + predict_text return result def predict_folder(self,predict_image_folder_path): predict_image_list=[] true_list=[] result_list=[] for filePath in os.listdir(predict_image_folder_path): true_list.append(filePath.split(".")[0]) predict_image_list.append(os.path.join(predict_image_folder_path,filePath)) saver = tf.train.Saver() predict = tf.reshape(self.output, [-1, 1, self.CHAR_SET_LEN]) max_idx_p = tf.argmax(predict, 2) with tf.Session() as sess: for image_path in predict_image_list: result="" captcha_image=self.get_image(image_path) for i in range(self.MAX_CAPTCHA): saver.restore(sess, tf.train.latest_checkpoint(self.model_path)) batch_x_predict=captcha_image[i] text_list = sess.run(max_idx_p, feed_dict={self.X: [batch_x_predict], self.keep_prob: 1}) text = text_list[0].tolist() vector = np.zeros(self.CHAR_SET_LEN) j = 0 for n in text: vector[j * self.CHAR_SET_LEN + n] = 1 j += 1 predict_text= self.vec2text(vector) result=result+predict_text result_list.append(result) return true_list,result_list def gen_image_for_test(self,image_Path): img = cv2.imread(image_Path,0) img = cv2.resize(img, (self.IMAGE_WIDTH*self.MAX_CAPTCHA,self.IMAGE_HEIGHT)) img = np.float32(img) return img def get_image(self,im