Document split into pages

View API docs →

Split an image that contains multiple document pages stacked vertically (a long phone snapshot of a multi-page handout, a tall scan, etc.) into one image per page using horizontal-projection segmentation.

POST /api/document-split-pages·cap 4.0 MiB·2 cost unit(s)·bucket: opencv

More OCR

See all →

Document extract structured

OCR a document and pull out common structured fields (emails, phones, dates, amounts, URLs, IBANs).

Document OCR

Extract text from a document image with -. Returns plain text + optional per-word/per-line bounding boxes and confidence scores.

Document table extraction

Detect ruled-grid table cells in a document image. Returns per-cell bounding boxes grouped by row.

OCR ID card

Prepare a photo of an ID card for OCR.

OCR receipt

Prepare a photo of a paper receipt for OCR.

OCR whiteboard

Prepare a photo of a whiteboard or poster for OCR.

Prepare for OCR

Pre-process a scanned document for better OCR results.