Giter VIP home page Giter VIP logo

alisen39 / trwebocr Goto Github PK

View Code? Open in Web Editor NEW
2.5K 2.5K 579.0 321.84 MB

开源易用的中文离线OCR,识别率媲美大厂,并且提供了易用的web页面及web的接口,方便人类日常工作使用或者其他程序来调用~

License: Apache License 2.0

Python 57.68% Dockerfile 1.71% JavaScript 4.76% HTML 2.01% Vue 33.78% Shell 0.06%
docker ocr ocr-recognition python python3 web

trwebocr's People

Contributors

alisen39 avatar dependabot[bot] avatar imba-tjd avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

trwebocr's Issues

模型训练

您好,请问可以训练自己的模型么

请问点识别的时候报错是怎么回事呢?

前端报错:Uncaught (in promise) TypeError: Cannot read property 'msg' of undefined
at Index.vue:199
后端打印:2020-05-08 13:11:15,776 - backend.webInterface.tr_run - ERROR - {"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}
ERROR:backend.webInterface.tr_run:{"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}
2020-05-08 13:11:36,915 - backend.webInterface.tr_run - ERROR - {"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}
ERROR:backend.webInterface.tr_run:{"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}
2020-05-08 13:11:45,388 - backend.webInterface.tr_run - ERROR - {"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}
ERROR:backend.webInterface.tr_run:{"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}
2020-05-08 13:13:00,152 - backend.webInterface.tr_run - ERROR - {"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}
ERROR:backend.webInterface.tr_run:{"code": 400, "msg": "\u4ea7\u751f\u4e86\u4e00\u70b9\u9519\u8bef\uff0c\u8bf7\u68c0\u67e5\u65e5\u5fd7", "err": "274"}

英文识别不好?

  • 识别出的英文单词之间没有空格
  • 识别速度似乎较慢
  • 字母识别错误,大写的“I”会识别成小写"l"(L)

Docker运行时报错

用docker镜像pull后,然后docker run 后,一直报这个错。
`2020-10-27 07:42:45,851 INFO success: trwebocr entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)

2020-10-27 07:42:48,861 INFO exited: trwebocr (exit status 1; not expected)

2020-10-27 07:42:49,864 INFO spawned: 'trwebocr' with pid 1385`

Docker run

docker run -itd -p 8089:8089 --name trwebocr trwebocr:latest 运行错误
应该是:docker run -itd -p 8089:8089 --name trwebocr mmmz/trwebocr:latest
不然会找不到库
image

合作交流

您好,我们是一家央企的人工智能公司(中译语通科技股份有限公司)主要从事大数据、智慧城市、机器翻译、知识图谱、语音识别、ocr识别等技术的研发,我是这边的技术负责人在github上看到您的开源系统很感兴趣,希望和您进一步沟通交流。

您可以加我微信:18611586751

安装失败

Traceback (most recent call last):
File "backend/main.py", line 11, in
from backend.webInterface import tr_run
File "/Users/fizz/the-world/TrWebOCR/backend/webInterface/tr_run.py", line 8, in
import tr
File "/Users/fizz/the-world/TrWebOCR/backend/tr/init.py", line 2, in
from .tr import *
File "/Users/fizz/the-world/TrWebOCR/backend/tr/tr.py", line 24, in
_libc = ctypes.cdll.LoadLibrary(os.path.join(_BASEDIR, 'libtr.so'))
File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/ctypes/init.py", line 442, in LoadLibrary
return self._dlltype(name)
File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/ctypes/init.py", line 364, in init
self._handle = _dlopen(self._name, mode)
OSError: dlopen(/Users/fizz/the-world/TrWebOCR/backend/tr/libtr.so, 6): no suitable image found. Did find:
/Users/fizz/the-world/TrWebOCR/backend/tr/libtr.so: unknown file type, first eight bytes: 0x7F 0x45 0x4C 0x46 0x02 0x01 0x01 0x03
/Users/fizz/the-world/TrWebOCR/backend/tr/libtr.so: unknown file type, first eight bytes: 0x7F 0x45 0x4C 0x46 0x02 0x01 0x01 0x03

对于"-"识别的不是很好

在实际测试中,金额前有“-” 或是文字中有“-” 的,有时候会识别不出来。修改缩放比例后,有改进,但是还其他图片还是有没识别出来的情况,
请问 这个如何提升?
谢谢

请问哪个版本使用的Tr1.5呢

由于是无网络环境的运行,新版需要验证Tr,无法使用,想问哪个版本是用的旧版Tr,可以无网络直接运行呢

405错误

import requests
import cv2
import base64
import json
url3 = 'http://192.168.1.102:8089/api/tr-run'
image = cv2.imread('a.png')
image = base64.b64encode(image)
data = {'img': image.decode('utf-8'), 'compress': 0}
json_mod = json.dumps(data)
res = requests.post(url=url3, data=json_mod)
print(res)
请问下这段代码为什么会报405错误?

为什么老是自动killed?

我测试了下,服务启动,用着用着就自动杀掉了
centos 8 下启动的docker

每使用一次,内存好像没释放

python调用多进程锁死

后台用supervisor启动了4个进程,python用进程池4个进程开启tr.run函数无法返回,改成1个进程可以正常返回。请问是啥问题?谢谢。

docker镜像如何用GPU加速识别?

想做一个pdf批量转word的接口,但是识别速度有点慢,尤其是在目标多的图片中表现得更慢。求告知GPU加速的方法,感激不尽

树莓派运行报错,请问这个怎样解决??ModuleNotFoundError: No module named 'cv2'

root@raspberrypi:~/TrWebOCR-master# python backend/main.py
Traceback (most recent call last):
File "backend/main.py", line 10, in
from backend.webInterface import tr_run
File "/root/TrWebOCR-master/backend/webInterface/tr_run.py", line 8, in
import tr
File "/root/TrWebOCR-master/backend/tr/init.py", line 2, in
from .tr import *
File "/root/TrWebOCR-master/backend/tr/tr.py", line 6, in
import cv2
ModuleNotFoundError: No module named 'cv2'

Win10下面不能运行,是因为缺失 libtr.dll吗?

Traceback (most recent call last):
File "backend/main.py", line 10, in
from backend.webInterface import tr_run
File "c:\Projects\TrWebOCR\TrWebOCR-master\backend\webInterface\tr_run.py", line 8, in
import tr
File "c:\Projects\TrWebOCR\TrWebOCR-master\backend\tr_init_.py", line 2, in
from .tr import *
File "c:\Projects\TrWebOCR\TrWebOCR-master\backend\tr\tr.py", line 22, in
_libc = ctypes.cdll.LoadLibrary(os.path.join(BASEDIR, 'libtr.dll'))
File "C:\ProgramData\Miniconda3\lib\ctypes_init
.py", line 442, in LoadLibrary
return self.dlltype(name)
File "C:\ProgramData\Miniconda3\lib\ctypes_init
.py", line 364, in init
self._handle = _dlopen(self._name, mode)
OSError: [WinError 126] The specified module could not be found

issue while restoring the un-minified version.

I understand that webpack is used in this project to minify the vuejs part of this project,
but I can't get it the original version back.
I used the shuji CLI to restore it but it was very distorted and didn't work, so any help with that ?
thanks.

如何设置输出过滤

现在的实现把一些confidence =0.2的都输出来了,全是错的,能调一下阈值就好了

一个证件照要识别14秒,请问原因是什么?

4核4G内存的,而且仅限办公室自己用,似乎很慢,请问有什么改进的办法吗?而且手机端识别更慢,用Post,img,base64我等了将近1分钟,是不是我哪里设置不对?比如我想提高速度,降低精度,该怎么做?谢谢!

可否给出前端的源码

蛮不错的前端页面,所以想复用下,
发现前端代码是编译后的,
可否给出前端的源码,
多谢。

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.