自动化测试相关代码+cnn循环神经网路识别二维码，注意：训练用的二维码图片排序要按照顺序

最新推荐文章于 2023-11-10 18:38:28 发布

San Chang

最新推荐文章于 2023-11-10 18:38:28 发布

阅读量468

点赞数

本文链接：https://blog.csdn.net/qq_26657001/article/details/103663754

版权

浏览器相关代码：
识别的图片下载
神经网络的加载

import  requests
from selenium import webdriver
from bs4 import BeautifulSoup
from selenium.webdriver.common.action_chains import ActionChains
from selenium.webdriver.common.keys import Keys
from PIL import Image
import time
import torch.nn as nn
import torch
from torch.utils.data import DataLoader,Dataset
import pyautogui
from torchvision import transforms
import shutil
import os

def default_loader(path):
    return Image.open(path).convert("L")#convert("RGB")
class MyDataset(Dataset):
    def __init__(self,txt,transform=None,target_transform=None,loader=default_loader):
        super(MyDataset,self).__init__()
        fp=open(txt,"r")
        imgs=[]
        for line in fp:
            line=line.strip("\n")
            line=line.rstrip()
            words=line.split()
            #print(words[0],int(words[1]))
            #code=np.zeros((5))
            #for i in range(len(words[1])):
                #code[i]=np.int(words[1][i])
            #print(np.shape())

            imgs.append((words[0],int(words[1])))
            #print(imgs)
            self.imgs=imgs
            self.transform=transform
            self.target_transform=target_transform
            self.loader=loader

    def __getitem__(self, index):
        fn, label = self.imgs[index]
        img = self.loader(fn)
        if self.transform is not None:
            img = self.transform(img)
        return img, label

    def __len__(self):
        return len(self.imgs)
class cnn(nn.Module):
    def __init__(self):
        super(cnn,self).__init__()
        self.conv1=nn.Sequential(
            nn.Conv2d(1,16,5,1,2),
            nn.ReLU(),
            nn.MaxPool2d(2,2)
        )
        self.conv2=nn.Sequential(
            nn.Conv2d(16,32,5,1,2),
            nn.ReLU(),
            nn.MaxPool2d(2,2)
        )
        self.out=nn.Linear(32*7*7,10)#输出为n个类别，如果编码大于这个类别则会报错。

    def forward(self, x):
        x=self.conv1(x)
        x=self.conv2(x)
        x=x.view(x.size(0),-1)
        output=self.out(x)
        #output=output.view(5,output.size(0))
        return output


headers={"User-Agent":"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36","Host":"Host: www.docin.com"}
#r=requests.get(link,headers=headers)
#print(r.text)
#soup=BeautifulSoup(r.text,"lxml")
#print(soup.text)

post_data={
    "login_tel":"17839949416",
    "password":"z.135139",
    "friendid":"0",
    "groupdid":"0",
    "guid":"0",
    "forwardUrl":"",
    "from":"registerPage",
    "end_registstat":"",
    "formatflag":"1001",
    "vregister":"null",
    "register_suc_forward":"null"
}
'''
session=requests.session()
login_page=session.post(link,data=post_data,headers=headers)
print(login_page.status_code)
'''

def douding_getpic():
    link = "https://www.docin.com/app/forward?forward=register&send=oneok"
    options=webdriver.ChromeOptions()
    #options.add_argument("--disable-extensions")
    '''
    options.add_experimental_option("prefs",{"download.default_directory":r'D://chrome_d',
         "download.prompt_for_download" : False,
         "download.directory_upgrade" : True,
         "safebrowsing.enabled" : True})
    '''
    browser = webdriver.Chrome(chrome_options=options)
    browser.get(link)

    time.sleep(0.5)
    # browser.find_element_by_name("phone").send_keys("13513974699")
    browser.find_element_by_name("login_tel").send_keys("13513974699")

    browser.find_element_by_id("password").send_keys("123456789")

    # left_click = browser.find_element_by_class_name("submit-step")
    left_click = browser.find_element_by_class_name("getvalidaCodeBtn").click()

    '''
    try:
        ActionChains(browser).click(left_click)
        print('success')
    except Exception as e:
        print('fail')
    '''
    path = "D:/BaiduNetdiskDownload/1.jpg"
    #browser.get_screenshot_as_file("D:/BaiduNetdiskDownload/1.jpg")
    # browser.switch_to.frame(0)
    # browser.switch_to_frame("")
    # browser.switch_to.frame(0)
    # browser.switch_to_active_element
    # im=browser.find_element_by_xpath("//")
    # browser.switch_to_frame(browser.find_element_by_xpath("//iframe"))
    # browser.switch_to_frame(browser.find_element_by_xpath("//div[@id='getPicCode']/iframe"))
    time.sleep(0.5)
    pic=browser.find_element_by_xpath("//div[@class='dialogCont']//img")
    # browser.find_element_by_id("getPicCode")
    # print(im)
    action = ActionChains(browser).move_to_element(pic)
    time.sleep(0.5)
    for i in range(10):
        action.click(pic)
        time.sleep(1)
        action.context_click(pic)
        action.perform()
        time.sleep(2)
        pyautogui.typewrite(["v"])
        time.sleep(2)
        num=i+21
        pyautogui.typewrite([str(int(num/10)),str(num%10),".","j","p","g"])
        pyautogui.sleep(2)
        pyautogui.typewrite(["enter"])
        time.sleep(2)
        pyautogui.typewrite(["enter"])
    #time.sleep(1)
    #
    #Image.open()

def xiaomi():
    link = "https://cn.account.xiaomi.com/pass/register"
    browser = webdriver.Chrome()
    browser.get(link)
    time.sleep(2)
    browser.find_element_by_name("phone").send_keys("17839949416")
    time.sleep(1)
    left_click = browser.find_element_by_class_name("submit-step").click()

#xiaomi()
#douding_getpic()

#def pic_rec()

def restore_net(path,net):
    predict_data=MyDataset(txt="D:/viladate_pic/rec/code.txt",transform=transforms.Compose([transforms.Resize((28,28),Image.BICUBIC),transforms.ToTensor()]))
    predict_data_loder=DataLoader(predict_data,batch_size=1,shuffle=False)
    # restore entire net1 to net2
    net = torch.load(path+net)
    for step,(p_x,p_y) in enumerate(predict_data_loder):
        out=net(p_x)
    out=torch.max(out,1)[1]
    return out

def rec_pic():
    path_org="D:/viladate_pic/"
    net_file="net.pkl"
    code=[]
    for i in range(5):
        path=path_org+str(i+1)+"/"
        out=restore_net(path,net_file)
        code.append(out.numpy())
    result=[]
    for i in range(len(code)):
        result.append(int(code[i]))
    result=str(result)
    print(result)

def main():
    link = "https://www.docin.com/app/forward?forward=register&send=oneok"
    options = webdriver.ChromeOptions()
    browser = webdriver.Chrome(chrome_options=options)
    browser.get(link)

    time.sleep(0.5)
    browser.find_element_by_name("login_tel").send_keys("13513974699")
    browser.find_element_by_id("password").send_keys("123456789")
    browser.find_element_by_class_name("getvalidaCodeBtn").click()
    time.sleep(1)
    pic = browser.find_element_by_xpath("//div[@class='dialogCont']//img")
    action = ActionChains(browser).move_to_element(pic)
    time.sleep(0.5)
    action.context_click(pic)
    action.perform()
    time.sleep(2)
    pyautogui.typewrite(["v"])
    time.sleep(2)
    pyautogui.typewrite(["enter"])
    time.sleep(2)
    if (os.path.exists("D:/viladate_pic/rec/getctime.jpg")):
        os.remove("D:/viladate_pic/rec/getctime.jpg")
        shutil.move("D:/Documents/Downloads/getctime.jpg","D:/viladate_pic/rec/")
    else:
        shutil.move("D:/Documents/Downloads/getctime.jpg", "D:/viladate_pic/rec/")
    rec_pic()
#main()
douding_getpic()

神经网络的初始化和训练

import cv2
import torch
import torch.nn as nn
import numpy as np
from torch.utils.data import DataLoader,Dataset
from torch.autograd import Variable
from torchvision import transforms
import PIL.Image as Image
#from PIL import

def default_loader(path):
    return Image.open(path).convert("L")#convert("RGB")
class MyDataset(Dataset):
    def __init__(self,txt,transform=None,target_transform=None,loader=default_loader):
        super(MyDataset,self).__init__()
        fp=open(txt,"r")
        imgs=[]
        for line in fp:
            line=line.strip("\n")
            line=line.rstrip()
            words=line.split()
            #print(words[0],int(words[1]))
            #code=np.zeros((5))
            #for i in range(len(words[1])):
                #code[i]=np.int(words[1][i])
            #print(np.shape())

            imgs.append((words[0],int(words[1])))
            #print(imgs)
            self.imgs=imgs
            self.transform=transform
            self.target_transform=target_transform
            self.loader=loader

    def __getitem__(self, index):
        fn, label = self.imgs[index]
        img = self.loader(fn)
        if self.transform is not None:
            img = self.transform(img)
        return img, label

    def __len__(self):
        return len(self.imgs)


class cnn(nn.Module):
    def __init__(self):
        super(cnn,self).__init__()
        self.conv1=nn.Sequential(
            nn.Conv2d(1,16,5,1,2),
            nn.ReLU(),
            nn.MaxPool2d(2,2)
        )
        self.conv2=nn.Sequential(
            nn.Conv2d(16,32,5,1,2),
            nn.ReLU(),
            nn.MaxPool2d(2,2)
        )
        self.out=nn.Linear(32*25*25,10)                                                        #输出为n个类别，如果编码大于这个类别则会报错。

    def forward(self, x):
        x=self.conv1(x)
        x=self.conv2(x)
        x=x.view(x.size(0),-1)
        output=self.out(x)
        return output                                                                               #输出为n个类别，如果编码大于这个类别则会报错。

def train(path,file):
    train_data = MyDataset(txt=path+file, transform=transforms.Compose([
        transforms.Resize((100, 100), Image.BICUBIC),
        # transforms.Normalize([0.5],[0.5]),
        # transforms.CenterCrop(224),
        transforms.ToTensor()]))
    train_loader = DataLoader(train_data, batch_size=1, shuffle=False)

    for i, (batch_x, batch_y) in enumerate(train_loader):
        print(i, batch_x, batch_y.size())

    cnn1=cnn()
    optimizer = torch.optim.Adam(cnn1.parameters(), lr=0.00001)  # optimize all cnn parameters
    loss_func = nn.CrossEntropyLoss()  # the target label is not one-hotted

    for epoch in range(50):
        for step,(b_x,b_y) in enumerate(train_loader): # 分配 batch data, normalize x when iterate train_loader
            b_x=Variable(b_x)
            b_y=Variable(b_y)
            output = cnn1(b_x)  # cnn output
            #b_y=b_y.long()
            #print(output)
            #print(b_y)
            loss = loss_func(output,b_y)  # cross entropy loss
            optimizer.zero_grad()  # clear gradients for this training step
            loss.backward()  # backpropagation, compute gradients
            optimizer.step()
    torch.save(cnn1, path+'net.pkl')


def restore_net(path,net):
    predict_data=MyDataset(txt="D:/viladate_pic/test/code.txt",transform=transforms.Compose([transforms.Resize((100,100),Image.BICUBIC),transforms.ToTensor()]))
    predict_data_loder=DataLoader(predict_data,batch_size=1,shuffle=False)
    # restore entire net1 to net2
    final_out=[]
    net = torch.load(path+net)
    for step,(p_x,p_y) in enumerate(predict_data_loder):
        out=net(p_x)
        #print(out)
        out=torch.max(out,1)[1]
        final_out.append(out.numpy())
    return final_out

def rec_pic():
    path_org="D:/viladate_pic/"
    net_file="net.pkl"
    code=[]
    for i in range(5):
        path=path_org+str(i+1)+"/"
        out=restore_net(path,net_file)
        code.append(out)
    print(code)
    result=[]
    for i in range(len(code[0])):
        code_re=[]
        for j in range(len(code)):
            code_re.append(int(code[j][i]))
        result.append(code_re)
    result=str(result)
    print(result)

def main_train():
    path_org="D:/viladate_pic/"
    file="code.txt"
    for i in range(5):
        path=path_org+str(i+1)+"/"
        train(path,file)

main_train()
rec_pic()