Extract table data from images into Excel.
Scan, chat with, and transform your documents. Convert, compress, merge, split and more — free to use.
Three workflows a generic PDF converter does not cover.
Extract table data from images into Excel.
Turn statement transactions into structured Excel data.
Pull structured fields out of a PDF.
Compress, convert, edit and organise files in the formats people use most.
Extraction, checking and question-answering for when converting a file is not enough.
A photograph of a table is just pixels. Doclium reads the words out of it with OCR, keeps the position of every word, and rebuilds the grid from where those words sit on the page — then writes it into Excel with the numbers as real numbers. Position is the part that matters. Reading the words alone gives you a list; a table is the words plus their columns. Doclium finds columns the way a reader does, from the vertical bands of whitespace running down the image, which is why it works on tables whose columns are separated by spacing rather than ruled lines. It will not invent a table that is not there. A page of prose comes back with a clear message instead of an imaginary one-column spreadsheet, because a wrong table that looks right is worse than no table at all.
Yes — that is the case it is built for. Columns come from the whitespace between them, not from ruled lines, so a table separated only by spacing still converts.
Real numbers, with the decimals the source showed, so SUM and AVERAGE work straight away. Anything that is an identifier rather than a quantity — an account number with a leading zero, a reference code — stays text so it is not altered.
Yes. JPG, PNG and HEIC all work, and so does a screenshot or a picture taken on a phone — the file type matters far less than whether the table is legible. A PDF goes through PDF to Excel instead, which reads the text layer directly when the file has one.
Yes. Multiple photos are read together, which is how a table running over several pages or several phone shots is handled.
The text is as accurate as OCR is on your image, and that is the honest limit: a flat, well-lit, in-focus photo reads well; a skewed, shadowed or low-resolution one loses characters, and no converter can recover what the image does not show. Check the numbers against the original before relying on them.
Handwriting, very tight column spacing, and cells merged across columns. Merged headers in particular extract once per column they span, so a header may repeat.
No tool matches that.
Questions, problems, or a document Doclium handled badly — tell us and a person will read it.
Review
Edit
Pages
Convert
Optimize
Protect
Scan, chat with, and transform your documents.
Popular document tasks
Recent Documents
AI Actions
Edit & Convert
Organize
Everything you can do to a document, in one place.
Most used
Doclium Ready
AI
Extract
Convert from PDF
Convert to PDF
Organize
Optimize
Edit
Security
Sign
No tool matches that.
Account, storage, and appearance.
Recent Files