Label
Keep the complete printed prompt.
Convert scanned forms to JSON with field-by-field review
Form OCR identifies visible labels, entered values, and selection marks, then returns them as editable key-value data. Blank fields and checkboxes are meaningful, so compare every extracted state with the source before automation.
Your form fields are ready to review
N/Alocal review signal
0 pages0 words0 characters0 ms
Confirm every label-value relationship and selection state. A visible signature mark is not identity verification.
Keep each answer attached to its label
Forms are spatial. The tool proposes key-value relationships and selection states, while the reviewer resolves ambiguous blanks, repeated labels, and marks that touch borders.
- Confirm every value is paired with the correct label.
- Distinguish blank, not applicable, unchecked, and unreadable states.
- Do not treat a visible signature mark as identity verification.
Preserve the relationship between every form label and answer
Use this tool for forms where labels and answers must stay paired in a structured export. Spatial proximity suggests a relationship, but repeated sections, side-by-side columns, and handwritten entries can make the pairing ambiguous.
Value
Preserve the entered wording without interpretation.
State
Separate blank, unchecked, selected, and unreadable.
Context
Add the section name when labels repeat.
Treat blank fields and selection marks as data
An empty box is not automatically false, and a visible mark is not automatically a valid choice. The result page keeps extracted states editable before JSON export.
Keep labels and answers together
Crop around complete field groups instead of isolating values from their printed labels.
Preserve empty boxes
Aggressive cleanup can erase faint borders or turn dust into a selected mark.
Name repeated fields clearly
Add section context when the same label appears for several people or items.
Minimize personal data
Remove unrelated sensitive fields before uploading when the task does not require them.
Review section boundaries
A value beside two similar labels may belong to the section above or below, so confirm its surrounding heading before export.
Use form OCR for structure, not identity or consent verification
Apply the structured review method in A Field-First OCR Workflow for Receipts, Invoices, and Tables.
For form structure and sensitive-data handling, continue with the form OCR privacy guides.
Best for
Surveys, applications, intake forms, checklists, and repeated label-value layouts.
Not designed for
Authenticating identities, validating consent, or interpreting legal and medical meaning.
Review first
Blank fields, checkbox states, repeated labels, signatures, and handwritten entries.
Form extraction questions
Can the tool tell whether a checkbox is selected?
It can propose a visible state, but faint ticks, crosses, and printed boxes need source review. Use uncertainty to prioritize review.
Does signature detection verify a person?
No. A visible mark can be reported as present, but authorship and legal validity are outside OCR. Read the recognition boundaries.