从照片和截图中提取文字——支持 JPG、PNG、WebP、BMP、TIFF 和 HEIC，OCR 全程在你的浏览器中运行。
点击选择文件 或将文件拖放到此处
JPG, PNG, WebP, BMP, TIFF, HEIC
您的文件从未离开您的设备。
请选择图片中文字的语言。语言数据只需下载一次，浏览器会自动缓存。
提高识别准确率的技巧：
处理中...
拖入照片、截图或扫描件，可一次多张。全部在本地读取，不会上传。
选择图片中文字的语言，然后点击“提取文字”，识别完全在浏览器中进行。
复制文字、下载为 .txt，或保存为保留原图并添加文字层的可搜索 PDF。
JPG, PNG, WebP, BMP, TIFF, AVIF and HEIC — photos from an iPhone work directly. Images are converted locally before recognition, so nothing is uploaded.
Clean, straight, well-lit photos of printed text usually come out almost perfect. Handwriting, heavy shadows and very small fonts are harder — crop and brighten the image first for a better result.
No. Recognition runs inside your browser with a WebAssembly OCR engine. The only thing downloaded is the language data for the language you choose, and your browser caches it.
Yes — use “Download searchable PDF”. You get a PDF that keeps the original image and adds an invisible text layer, so you can search and select the text afterwards.