{"full_name":"hiroi-sora/Umi-OCR","owner":"hiroi-sora","name":"Umi-OCR","url":"https://github.com/hiroi-sora/Umi-OCR","description":"OCR software, free and offline. 开源、免费的离线OCR软件。支持截屏/批量导入图片，PDF文档识别，排除水印/页眉页脚，扫描/生成二维码。内置多国语言库。","homepage":"","topics":[],"language":"Python","license":"MIT","avatar":"https://avatars.githubusercontent.com/u/56373419?u=e038cfb9daf0355c11f125e103023f7df85d6159&v=4","created_at":"2022-03-28T04:01:46Z","pushed_at":"2025-11-20T14:58:13Z","metrics":{"stars":48554,"forks":4739,"watchers":229,"issues":881,"prs":76,"releases":31,"issues_open":null,"issues_closed":null,"prs_open":null,"prs_closed":null,"prs_merged":null,"discussions":115},"growth":{"day":{"stars":24,"forks":2,"base_day":"2026-11-09"},"week":{"stars":176,"forks":12,"base_day":"2026-11-03"},"month":{"stars":953,"forks":87,"base_day":"2026-10-11"}},"history":{"dates":["2026-09-06","2026-09-07","2026-09-08","2026-09-09","2026-09-10","2026-09-11","2026-09-12","2026-09-13","2026-09-14","2026-09-15","2026-09-16","2026-09-17","2026-09-18","2026-09-19","2026-09-20","2026-09-21","2026-09-22","2026-09-23","2026-09-24","2026-09-25","2026-09-26","2026-09-27","2026-09-28","2026-09-29","2026-09-30","2026-10-01","2026-10-02","2026-10-03","2026-10-04","2026-10-05","2026-10-06","2026-10-07","2026-10-08","2026-10-09","2026-10-10","2026-10-11","2026-10-12","2026-10-13","2026-10-14","2026-10-15","2026-10-16","2026-10-17","2026-10-18","2026-10-19","2026-10-20","2026-10-21","2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29","2026-10-30","2026-10-31","2026-11-01","2026-11-02","2026-11-03","2026-11-04","2026-11-09","2026-11-10"],"stars":[46796,46828,46880,46909,46931,46944,46972,47031,47046,47065,47090,47113,47164,47190,47220,47234,47251,47264,47303,47320,47344,47364,47384,47395,47430,47445,47464,47486,47505,47509,47535,47548,47569,47565,47577,47601,47616,47644,47688,47729,47795,47849,47872,47905,47931,47967,47989,48051,48067,48088,48105,48131,48160,48210,48232,48246,48276,48297,48378,48402,48530,48554],"forks":[4583,4585,4586,4591,4593,4594,4594,4602,4607,4607,4607,4610,4611,4616,4618,4619,4618,4620,4621,4623,4624,4626,4630,4632,4634,4635,4637,4640,4642,4644,4647,4651,4651,4651,4650,4652,4654,4657,4660,4662,4665,4672,4673,4675,4677,4680,4681,4683,4685,4688,4692,4697,4697,4700,4706,4711,4715,4714,4727,4731,4737,4739],"watchers":[226,226,226,226,226,226,227,231,231,231,231,231,233,232,232,232,232,232,232,232,232,232,232,232,232,232,232,232,232,233,235,236,236,236,236,236,236,237,238,238,238,239,241,241,241,242,243,243,242,242,236,235,230,229,229,229,229,229,229,229,229,229],"issues_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null],"prs_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null]},"boards":{},"same_owner":[],"issues":[{"number":958,"title":"我在ARM64的linux尝试部署UMIocr时遇到各种问题，用docker打镜像也不行","comments":40,"created_at":"2025-08-13T06:58:00Z"},{"number":254,"title":"【测试】Umi-OCR 支持数学公式识别啦","comments":32,"created_at":"2023-11-30T14:50:46Z"},{"number":447,"title":"【FAQ】 常见问题 / Frequently Asked Questions","comments":20,"created_at":"2024-03-28T13:40:29Z"},{"number":861,"title":"亲爱的作者，我有个问题。通过http调用长时间无反应，查看日志显示报错信息","comments":12,"created_at":"2025-04-23T02:45:33Z"},{"number":617,"title":"HTTP报错","comments":12,"created_at":"2024-08-01T08:56:16Z"},{"number":905,"title":"是不是在酝酿跟进升级到PaddleOCR v3了？精度提升13%好期待","comments":11,"created_at":"2025-05-31T03:39:57Z"},{"number":833,"title":"特定目录下不能通过cmd启动软件","comments":11,"created_at":"2025-03-31T08:25:31Z"},{"number":577,"title":"大神，我最近使用下咱们的软件，整体很不错，不过有的文字会识别错误，","comments":11,"created_at":"2024-07-02T12:40:19Z"}],"discussions":[{"number":1093,"title":"我重置了Umi-OCR，可以参考：[Umi-OCR](https://github.com/Dominic485649/Umi-OCR)","comments":0,"category":"Show and tell","created_at":"2026-10-13T16:48:58Z"},{"number":1090,"title":"项目被仿冒 翻冒网站umi-ocr.com.cn 实际银狐病毒","comments":0,"category":"General","created_at":"2026-10-09T07:40:35Z"},{"number":1086,"title":"UMI-OCR 一直将\"冗\"识别为\"亢\"","comments":0,"category":"Q&A","created_at":"2026-09-23T01:23:11Z"},{"number":1084,"title":"[插件分享] Umi-OCR OCR 专为离线设计插件：PP-OCRv6、HyperLPR3、ddddocr","comments":0,"category":"Show and tell","created_at":"2026-09-08T14:59:40Z"},{"number":1078,"title":"启动后看不到窗口？可能是窗口跑到屏幕外面了","comments":0,"category":"General","created_at":"2026-07-22T09:35:10Z"},{"number":1076,"title":"OCR输出PDF清晰度","comments":0,"category":"General","created_at":"2026-07-18T13:26:43Z"},{"number":1075,"title":"请问能否对台版竖排书的识别做增强补丁","comments":0,"category":"General","created_at":"2026-07-14T06:20:11Z"},{"number":1074,"title":"Proposal: Optional multilingual Windows installer","comments":0,"category":"General","created_at":"2026-07-07T08:06:17Z"}],"latest_release":{"name":"Umi-OCR v2.1.5","tag":"v2.1.5","published_at":"2025-03-25T16:31:42Z","body":"[English](https://github-com.translate.goog/hiroi-sora/Umi-OCR/releases/tag/v2.1.5?_x_tr_sl=zh-CN&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp)\r\n\r\n> Umi-OCR 三周年撒花~~\r\n\r\n## v2.1.5 主要更新：\r\n\r\n- 新增：日志机制。在命令行中启动 Umi-OCR 可查看实时日志。指定级别以上（默认为ERROR）的日志被保存到 `Umi-OCR/UmiOCR-data/logs` 目录中，保存级别可以在全局设置标签页中更改。\r\n- 新增：大部分标签页能手动切换左右/上下双栏模式。 (#789)\r\n- 新增：Esc键隐藏主窗口。 (#652)\r\n- 新增：调整二维码生成相关参数后，自动刷新二维码生成。 (#690)\r\n- 新增：命令行指令 `--reload` ，用于重新加载配置文件。 👉 [文档](https://github.com/hiroi-sora/Umi-OCR/blob/main/docs/README_CLI.md#软件操控指令)\r\n- 修复：文档识别提取PDF自带的文本内容时，未考虑页面旋转的影响。 (#785)\r\n- 修复：文档识别生成单层PDF时，未写入原PDF自带的文本内容。\r\n- 修复：OCR结果展示列表的一些显示Bug和鼠标划选Bug。\r\n- 修复：调整标签页顺序或删除标签页后，未及时保存顺序信息。\r\n- 修复：HTTP接口 `/api/doc/download` 参数 `ignore_blank` 的错误。\r\n- 修复：Linux版本截图时，系统任务栏推移顶层窗口，导致截图位置偏移。 (#778)\r\n- 修复：Linux版本截图后，主窗口的位置与操作前不一致。\r\n- 优化：图片/文档的异步加载机制。现在可以流畅地加载含有数万个子文件的文件夹，且能预览加载进度。 (#710)\r\n- Windows 版本更新第三方依赖库：`PyMuPDF 1.24.11` ,`fontTools 4.56.0` ,`Pillow 10.4.0` ,`psutil 10.4.0` ,`pynput 1.8.0` ,`zxing-cpp 2.3.0`\r\n- 新增UI语言：俄语 `Русский` ，译者：[Вячеслав Анатольевич Малышев](https://hosted.weblate.org/user/1969)、[Muhammadyusuf Kurbonov](https://hosted.weblate.org/user/muhammadyusuf.kurbonov2002)。泰米尔语 `தமிழ்` ，译者：[தமிழ்நேரம்](https://hosted.weblate.org/user/TamilNeram/)。\r\n\r\n<img width=\"80%\" alt=\"未标题-1\" src=\"https://github.com/user-attachments/assets/38fa1df1-c783-4592-8415-4eaa7b3fc1d8\" />\r\n\r\n---\r\n\r\n## Windows 下载\r\n\r\n不同版本仅OCR引擎插件不同，其它功能完全一致，附带多国语言识别库。`.7z.exe`为自解压包，可以用压缩软件打开，也可以在没有安装压缩软件的电脑上直接双击解压。\r\n\r\n### Paddle 引擎插件版\r\n\r\n（性能好，速度快，占用率高，适合高配机器。**不兼容奔腾、赛扬、凌动CPU**）\r\n\r\n**重要提示：如果执行OCR时报错 `0xc0000142`、`[Error] OCR init fail`，大概率是CPU不兼容Paddle，请换用Rapid版本。**\r\n\r\n**Umi-OCR_Paddle_v2.1.5.7z.exe** （128MB）\r\nSHA256: `abe00248d6e68dda75029739751ad7258f0ecc14d720b19240a844f5341859b2`\r\n\r\n### Rapid 引擎插件版\r\n\r\n（速度稍慢，内存占用低，适合低配机器，兼容性好）\r\n\r\n**Umi-OCR_Rapid_v2.1.5.7z.exe** （98.5MB）\r\nSHA256: `659c55896c32a5e019dc7bde1713d0e5c73186a2c653bed84c4480fa1795b722`\r\n\r\n更多下载途径： [蓝奏云](https://hiroi-sora.lanzoul.com/s/umi-ocr) | [SourceForge](https://sourceforge.net/projects/umi-ocr/files/v2.1.5/) | [Scoop](https://scoop.sh/#/apps?q=umi-ocr)\r\n\r\n## Linux 下载\r\n\r\n**Umi-OCR_Linux_Paddle_2.1.5.tar.xz** （264.5MB）\r\nSHA256: `7ff32e5dc818d67171820545100433c86e7e71155f9bad0851fad2ced9588f9b`\r\n\r\n首次运行请在命令行中执行 `./umi-ocr.sh` ，随后可在全局设置中创建桌面/开始菜单快捷方式，便于日常使用。\r\n\r\n## Docker 部署\r\n\r\n- 请按文档进行：https://github.com/hiroi-sora/Umi-OCR_runtime_linux/blob/main/README-docker.md\r\n\r\n[![Download Umi-OCR](https://a.fsdn.com/con/app/sf-download-button)](https://sourceforge.net/projects/umi-ocr/files/v2.1.5/)"},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-10T09:35:25Z","history_start":"2026-09-06"}