add lab-tesseract

lab-tesseract/.gitignore

@@ -0,0 +1 @@
+/clear_image

lab-tesseract/@res/网址.txt

@@ -0,0 +1,7 @@
+https://yunnan.chinatax.gov.cn/col/col8041/index.html
+https://app.yunnan.chinatax.gov.cn/static/pages/ynds/fpcx/fpcx3.html?gdslxdm=1
+
+153002209100
+03839955
+915303005577687468
+dfefb

lab-tesseract/README.md

@@ -0,0 +1,10 @@
+研究使用 `tesseract` 和相关技术进行验证码识别
+
+需求资源：[@res](/%40res/)
+
+参考文章：https://blog.csdn.net/u011457798/article/details/84063963
+
+- [√] 使用官方的数据模型进行识别
+- [×] 优化和提升图像降噪处理技术
+- [×] 使用自己训练的模型进行识别
+- [×] 使用Python、Java等语言的绑定开发接口应用，以及脱离安装 `tesseract` 程序安装

lab-tesseract/clear.py

@@ -0,0 +1,41 @@
+import os
+from PIL import Image
+
+def clear_save_image(image_path):
+	image = Image.open(image_path)
+	image = clear_image(image)
+	image = image.convert('L') #灰度处理
+	file_name = os.path.splitext(os.path.basename(image_path))[0]
+	save_path = os.path.join(os.path.dirname(image_path), '../clear_image', file_name + '.jpg')
+	image.save(save_path) #图片转换成tif保存到clear_image文件夹中
+
+def clear_image(image):
+	image = image.convert('RGB')
+	width = image.size[0]
+	height = image.size[1]
+	noise_color = get_noise_color(image)
+	for x in range(width):
+		for y in  range(height):
+			#清除边框和干扰色
+			if (x == 0 or y == 0 or x == width - 1 or y == height - 1 
+				or image.getpixel((x, y)) == noise_color):
+				image.putpixel((x, y), (255, 255, 255))
+			#背景调整为白色	
+			if (image.getpixel((x, y))[0] > 180 and image.getpixel((x, y))[1] > 180 and image.getpixel((x, y))[2] > 180):
+				image.putpixel((x, y), (255, 255, 255))
+	return image
+
+def get_noise_color(image):
+	for y in range(1, image.size[1] - 1):
+		# 获取第2列非白的颜色
+		(r, g, b) = image.getpixel((2, y))
+		if r < 255 and g < 255 and b < 255:
+			return (r, g, b)
+
+if __name__ == '__main__':
+	os.makedirs('./clear_image/', exist_ok=True)
+	for file_path in os.listdir('image'):
+		file_path = os.path.join('image', file_path)
+		if os.path.isfile(file_path):
+			clear_save_image(file_path)
+	print("clear end.")

lab-tesseract/download.py

@@ -0,0 +1,31 @@
+#!/usr/bin/env python3
+# -*- coding:utf-8 -*-
+
+import requests  # http客户端
+import os  # 创建文件夹
+import time
+from PIL import Image
+
+os.makedirs('./image/', exist_ok=True)
+IMAGE_URL = "https://app.yunnan.chinatax.gov.cn/app/base/captcha.do?time=" + (str(round(time.time() * 1000)))
+print(IMAGE_URL)
+
+
+head = {
+    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.69 Safari/537.36"
+}
+
+def request_download(num):
+    r = requests.get(IMAGE_URL, headers=head)
+    with open('./image/%s.png' % str(num), 'wb') as f:
+        f.write(r.content)
+
+
+try:
+    for i in range(100):
+        request_download((i+1))
+        print('download to ./image/%s.png' % str((i+1)))
+        #im = Image.open('./image/img.png')
+        #im.show()
+except:
+    print('download error!')