-
大佬,这个代码只能筛选 截图或者粘贴图片的结果. 如果 我本身复制的结果是 文字(21156一二三四),然后想通过 粘贴图片的功能 筛选出21156.请问是修改哪一段的代码
_Originally posted by @kyon777 in https://github.com/hiroi-sora/Umi-OCR/issues/403#issuecomment-1987396098_
…
-
请问可以加入scoop库,通过scoop管理吗
-
Qt plugins 目录导入失败!
Qt plugins directory import failed!
DLL load failed while importing shiboken2:找不到指定的模块
-
如图。识别到的文字保存不进kk.txt
-
启动截图后截图识别任务异常[Error] MissionOCR: APl object is None.
-
hiroi-sora/RapidOCR-json/issues/16 的伴生issue。
PDF样本已上传:
[心血管-中药新药临床研究指导原则.pdf](https://github.com/hiroi-sora/Umi-OCR/files/14472277/default.pdf)
在执行完显示引擎后,在执行批量识别文档的时候偶然发现此文档识别速度极快,并且没有弹出引擎开启的黑框。
…
-
首先,非常感谢大佬的辛勤付出,才让我们有了这么好用的网盘工具;我个人看了一圈,什么nextcloud,seafile等等,虽然各有特色,但最轻量,功能恰到好处,集成blog,部署简单的,唯此一家!!有种终于等到你的感觉。
我从使用场景上提两个小小的需求,希望大佬有空余时间的时候,能考虑下实现。
1.上传功能,增加基于服务端的本地路径扫描
场景:在部署之前,服务器上已经存在大量文件,部署之…
-
### Expected behavior 预期的功能
大大好,现在有忽略识别区域的功能,能不能加一个自己选定识别区域的功能。
如果也能导入自己的规则那就更好了。就是可以有一个配置文件,里面写满了各个识别区域的参数。识别区域通常是四个参数决定的,比如左上的xy坐标还有识别区域的长和宽。可能这个功能有些小众,我也不需要这个功能太自动化,识别区域的脚本我会自己写,谢谢大大。
### Approx…
-
作者您好,本人是文科专业,书籍大多是扫描版本图书,写论文时看扫描图书最头疼的就是无法直接检索,能不能开发该软件,可以批量制作双层PDF,谢谢您!
-
[Error] OCR init fail. Argd: {'enable_mkldnn': True, 'cpu_threads': 4, 'config_path': 'models/config_chinese.txt', 'cls': False, 'use_angle_cls': False, 'limit_side_len': 960}
OCR init fail.
![WeCha…