Read text in images
Sort the job · flatten and clean · extract text · model only when needed
그림 속 글자 읽기
사진 속 글자를 읽어 냅니다. 글자가 든 PDF 는 더 정확한 도구로 보내 주고, 모형은 켤 때만 받습니다
읽기 좋게 다듬은 그림
읽어 낸 글
사진을 고르세요
한국어·일본어는 아직 못 읽습니다(모형이 없습니다). 인쇄된 라틴 문자는 읽습니다. 모형은 켤 때만 받고, 사진은 브라우저를 벗어나지 않습니다.
Reads text out of images — but sorts the job first: a PDF that already has text only needs extracting, a skewed photo needs flattening, and only the rest needs a model.
How to use it
- Upload an image or a PDF
- Read which kind of job it is (extract / flatten / read)
- If it is skewed or shaded, flatten and clean it first
- Pull the text out and copy it
Common questions
- Why sort the job instead of just reading?
- "Get the text out of this picture" is really three jobs. Push all three down one path and all three get worse — a PDF that already has text needs no reading, and a skewed photo has to be flattened first.
- What about PDFs that already contain text?
- That is extraction, not reading, so pdf2text does it. This tool does not rebuild the same thing.
- Does the reading model always run?
- No. Only for the kind that truly needs reading (scanned PDFs and photos), and only for people who turned it on.
Other tools
Stays on your device (downloads data) What you enter is handled inside your browser. Only the data or model it needs is downloaded once.
Goes to — cdn.jsdelivr.net (모형 파일)
This tool page states where what you enter goes — see the badge above. All tools · KarmoLab · Who made this · Built with AI. View source