图像到 Markdown 转换器 — 使用 OCR 将屏幕截图转为 Markdown

Text is read line by line and kept in reading order. When words line up in columns, the converter rebuilds them as a Markdown pipe table with a header row, which renders as a real table in GitHub, Notion and Obsidian. Headings, bullet points and code come out as plain text lines, so add # heading marks, - list markers or backtick fences in the editable output before you copy or download it.

Yes, for the text and any tables. Upload the rendered README, blog post or docs page and you get its words in reading order, with tables rebuilt as pipe tables. Headings, lists and code formatting need to be marked again by hand, so it suits rescuing your own content from a page whose source you no longer have rather than producing an exact copy of the original file.

Tables use GitHub-Flavored Markdown (GFM) pipe-table syntax, a header row followed by a |-|-| separator line, which VS Code, Notion, Obsidian, Bear, Logseq and the GitHub web UI all render. Everything outside a table is plain text, so the rest of the file is valid CommonMark too.

Yes. The Markdown is plain text that you can paste directly into any editor. Notion turns pasted pipe tables into table blocks; Obsidian and VS Code render them as-is. Use the Copy button for clipboard paste, or download the .md file to drag into your vault or repository; both include any corrections you typed.

Code in a screenshot comes back as plain text lines, without backticks. Wrap snippets in triple backticks yourself in the editable output, add a language hint (```js, ```python) for syntax highlighting, and proofread brackets, quotes and indentation before you copy the result.

根据我们的安全政策，上传的内容将被删除——除非您明确共享结果，这会将其保存在公共链接上，任何拥有该结果的人都可以打开长达 30 天——并且永远不会重复用于培训。 Markdown 输出没有水印，没有归属注释，也没有跟踪像素 - 它是您完全拥有的纯文本。这使得该工具对于客户文档、技术写作和内部知识库导入来说是安全的。

使用“设置”中的引擎下拉菜单——“默认”能很好地处理大多数截图和照片，而引擎 1 或引擎 2 在风格化字体或异常表格边框上能得到更准确的结果。切换后点击“重新运行提取”以重新生成 Markdown 输出。

可以——语言选择器让你选择图片中包含的文字（默认英语），这可提高对带重音字符、非拉丁字母和特定语区标点的识别率。列表中只显示当前所选引擎支持的语言，因此切换引擎可能会把不受支持的选择重置回英语。

Yes — the result appears in an editable textarea, so you can fix a misread word, correct a table cell, or add heading and list marks right there. Copy Markdown and the .md download both use your edited version.

Set the Language picker to the page's language, try Engine 2 first since documentation often uses monospaced or stylised fonts, then add heading marks and code fences and proofread the tables in the editable output textarea before copying. Text comes back line by line, so that markup step is what turns the read into a usable README.

是的 - Pixoate 支持批处理和批量处理。切换到批处理模式，在 Premium 上添加最多 60 个图像，在 Pro 上添加最多 200 个图像，设置一次选项，在下载单个 ZIP 之前，每个图像都会使用相同的设置进行处理。批量处理是一项高级功能；输出使用与单一模式相同的质量和设置。

是的 - 通过批量处理，您只需配置一次设置，它们就会应用到批次中的每个项目 - Premium 上最多 60 个图像，Pro 上最多 200 个图像。无需对每个项目重复设置，临时上传和生成的文件将被安全处理并自动删除。

几乎总是三件事之一。文件超过 20 MB 免费上传限制 - Premium 需要 50 MB，Pro 需要 120 MB。该工具不接受该格式，因此请检查上传面板上显示的列表。或者该文件实际上并不是其名称所示的格式：重命名文件不会更改其中的内容，这就是看起来有效的扫描或照片被拒绝的原因。

准确度比任何设置都更能跟踪源图像。直接拍摄或扫描而不是倾斜，获得均匀的光线，没有阴影落在页面上，并用文档填充框架。在运行之前设置语言 - OCR 猜测错误的字母表是导致输出看起来像废话的常见原因。在依赖结果之前，请务必先阅读结果。

Image to Markdown supports free standard use within the published processing allowances. Free downloads are clean — no watermark. No signup is required for standard tool use. Photo effects and AI edits have a web-size limit on Free, with the original-resolution result available through Premium or one credit. Crop, resize, rotate, compress and convert stay full size. Premium supports batches of up to 60 files, with the applicable plan limits shown before processing.

否 - Image to Markdown 在桌面和手机上的浏览器中运行。在您看到它是否满足您的需要之前，无需安装任何东西，无需试用，也无需购买许可证。在您实际遇到的文件而不是样本上尝试一下。