完善

Author: lc6464

.gitignore

@@ -127,3 +127,9 @@ dmypy.json
 
 # Pyre type checker
 .pyre/
+
+# VSCode
+.vscode/
+
+# 下载的图片
+image/

README.md

@@ -1,2 +1,6 @@
-# identification-codes-py
-由 Python 编写的超级简单的不怎么靠谱的验证码识别系统
+# Python 验证码识别
+
+> 由 Python 编写的超级简单的不怎么靠谱的验证码识别系统
+
+- 使用 PIL 处理验证码图片
+- 使用 ORC 方式识别验证码

program.py

@@ -0,0 +1,76 @@
+from PIL import Image
+import pytesseract, os
+from urllib.request import urlopen, Request
+
+headers = {
+	'Referer': 'http://chaxun.heyuanedu.cn:88/',
+	'Accept': 'text/html,image/webp,image/png,image/jpeg,*/*;q=0.8',
+	'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.135 Safari/537.36 Edg/84.0.522.63',
+}
+
+class GetCode:
+	def __init__(self, path, url=None, headers=headers, origin='chaxun.heyuanedu.cn:88', threshold=185): # 初始化对象实例
+		self.path = path
+		self.url = url
+		self.headers = headers
+		self.origin = origin
+		self.threshold = threshold
+
+	def getReq(self, url=None, headers=None, origin='chaxun.heyuanedu.cn:88'): # 获取 Request 对象
+		self.url = url if url != None else self.url
+		self.headers = headers if headers != None else self.headers
+		self.origin = origin if origin != None else self.origin
+		return Request(self.url, None, self.headers, self.origin)
+
+	def setting(self, path=None, url=None, headers=None, origin=None, threshold=None): # 设置实例的各类变量
+		self.path = path if path != None else self.path
+		self.url = url if url != None else self.url
+		self.headers = headers if headers != None else self.headers
+		self.origin = origin if origin != None else self.origin
+		self.threshold = threshold if threshold != None else self.threshold
+
+	def getSession(self, url=None, headers=None, origin=None): # 获取对应系统的 Session ，注意：请勿重复调用！否则将可能导致不可预知的问题！
+		self.url = url if url != None else self.url
+		self.headers = headers if headers != None else self.headers
+		self.origin = origin if origin != None else self.origin
+		self.headers['Cookie'] = urlopen(self.getReq()).info()['Set-Cookie'].split(';')[0]
+		return self.headers
+
+	def download(self, url=None, path=None, headers=None, origin=None): # 下载验证码图片
+		self.url = url if url != None else self.url
+		self.path = path if path != None else self.path
+		self.headers = headers if headers != None else self.headers
+		self.origin = origin if origin != None else self.origin
+		if not os.path.exists(self.path[:self.path.rfind(os.sep)]): # 防止目录不存在出错
+			os.mkdir(self.path[:self.path.rfind(os.sep)])
+		with open(self.path, 'wb') as f:
+			size = f.write(urlopen(self.getReq()).read())
+		return size
+
+	def identify(self, path=None, threshold=None): # 识别验证码
+		self.path = path if path != None else self.path
+		self.threshold = threshold if threshold != None else self.threshold
+		image = Image.open(self.path)
+		pixdata = image.load()
+		w, h = image.size
+		for y in range(h): # 处理多余色彩
+			for x in range(w):
+				if pixdata[x, y][0] < self.threshold and pixdata[x, y][1] < self.threshold and pixdata[x, y][2] < self.threshold:
+					pixdata[x, y] = (0, 0, 0)
+				else:
+					pixdata[x, y] = (255, 255, 255)
+		for y in range(1,h-1): # 处理干扰线
+			for x in range(1,w-1):
+				count = 0
+				if pixdata[x,y-1][0] > 245:
+					count = count + 1
+				if pixdata[x,y+1][0] > 245:
+					count = count + 1
+				if pixdata[x-1,y][0] > 245:
+					count = count + 1
+				if pixdata[x+1,y][0] > 245:
+					count = count + 1
+				if count > 2:
+					pixdata[x,y] = (255,255,255)
+		image.save(self.path[:self.path.rfind(os.sep)+1] + 'code-Pretreatment.png') # 保存预处理图片（非必须，生产环境建议删除）
+		return pytesseract.image_to_string(image, 'eng')

test.py

@@ -0,0 +1,38 @@
+print('preparing...')
+
+from program import GetCode, headers
+from bs4 import BeautifulSoup
+from urllib.request import Request, urlopen
+from urllib.parse import urlencode
+import re, os
+
+ele0 = ''
+p = re.compile(r'(\d)[^0-9]*(\d)[^0-9]*(\d)[^0-9]*(\d)') # 此系统验证码正则
+
+getCode = GetCode('image%scode.png'%os.sep, 'http://chaxun.heyuanedu.cn:88/validatecode.php?act=getimg') # 实例化对象
+getCode.getSession() # 获取 Session
+
+userName = input('姓名：')
+identity = input('准考证号：')
+
+while ele0 == '': # 识别通过正则但错误循环
+	result2 = None
+	while result2 == None: # 识别无法通过正则循环
+		#time.sleep(0.2)
+		print('Downloading...')
+		getCode.download()
+		print('Identifying...')
+		result = getCode.identify()
+		result2 = ('%s%s%s%s'%p.search(result).groups() if p.search(result) != None else None)
+		result0 = (result2 if result2 != None else result) if result != '' else 'failure!'
+		print('Result: ' + result0)
+
+	req = Request('http://chaxun.heyuanedu.cn:88/search.php', urlencode({ 'userName': userName, 'identity': identity, 'code': result0 }).encode(), headers, 'chaxun.heyuanedu.cn:88')
+	res = urlopen(req) # 获取结果
+	html = res.read().decode()
+	#print(res.headers)
+	#print(html)
+	soup = BeautifulSoup(html,'lxml') # 解析 html
+	ele = soup.select('#printGrade')
+	ele0 = str(ele)[1:-1]
+	print(ele0 if ele0 != '' else 'Error!')