通过pytesseract识别验证码

准备工作：

1.使用国内源安装库pytesseract
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple pytesseract
2.安装tesseract
自己下载安装包吧

from PIL import Image
import pytesseract
import os

def get_all_path():
    start = r"E:\hhh\soft\中国知网验证码"
    lis = os.listdir(start)
    path = [start+os.sep+name for name in lis]
    return path
    
def get_ma(path):
    image = Image.open(path)
    text = pytesseract.image_to_string(image)
    return text
    
def open_code(): 
    num = 0
    all_path = get_all_path()
    for path in all_path:
        right = path.split("\\")[-1].replace(".jpg", "")
        ma = get_ma(path)
        if(right == ma):
            print("正确为：{},机器识别为：{},结果正确".format(right, ma))
            num += 1
        else:
            print("正确为：{},机器识别为：{},结果错误".format(right, ma)) 
    print("最终成功率为%.4f" % (num/len(all_path)))
    
open_code()

设置阈值，转灰度图像，二值化

from PIL import Image
import pytesseract
import os

def get_all_path():
    start = r"E:\hhh\soft\中国知网验证码"
    lis = os.listdir(start)
    path = [start+os.sep+name for name in lis]
    return path

def set_threshold():
    threshold = 127  #阈值
    table=[]
    for i in range(256):
        if i<threshold:
            table.append(0)
        else:
            table.append(1)
    return table
        
def get_ma2(path):
    image = Image.open(path)
    image = image.convert("L")    # 转换为灰度图像
    table = set_threshold()
    newImage = image.point(table,'1')   #二值化
    text = pytesseract.image_to_string(newImage)
    return text

def open_code2(): 
    num = 0
    all_path = get_all_path()
    for path in all_path:
        right = path.split("\\")[-1].replace(".jpg", "")
        ma = get_ma2(path)
        if(right == ma):
            print("正确为：{},机器识别为：{},结果正确".format(right, ma))
            num += 1
        else:
            print("正确为：{},机器识别为：{},结果错误".format(right, ma)) 
    print("最终成功率为%.4f" % (num/len(all_path)))
    
open_code2()