Detection
Find signature candidates in the document, before reference comparison.
Explore detection →BETASIGNATURE DETECTION & VERIFICATION
Detect handwritten signatures, return their positions, compare against authorised references and run agreed validations across your scanned documents. ChequeDB Signatures is now in beta.
Forms · contracts · mandates · delivery notes · cheques
THE FOUR QUESTIONS
Two drawn signature candidates appear on this fictional agreement. Their positions do not identify the signers.
Synthetic drawings and example workflow. No upload, model inference or live validation is performed here.
ONE DOCUMENT. FOUR CONNECTED CAPABILITIES.
Find signature candidates in the document, before reference comparison.
Explore detection →Connect each signature region to its page and coordinate frame.
Explore positioning →Compare the visible writing with authorised reference samples.
Explore verification →Apply agreed document checks and make the next review step clear.
Explore validation →FOUR CONNECTED CAPABILITIES
Start with the question your team needs answered. Detection and positioning can run without reference samples. Add verification when you need to compare the writing with a known signer's authorised samples.
Find regions that appear to contain handwritten signatures. Use this first step to investigate whether an expected field has been completed.
See where a candidate occurs on the page. Keep the region connected to its source so it can be highlighted, cropped and inspected in context.
Compare the candidate with the authorised reference samples supplied for the intended signer. Consider visible differences alongside reference variation and capture quality.
Run the agreed checks for your document workflow. Decide which conditions need a clearer scan, an available reference or a reviewer before processing continues.
BEYOND CHEQUES
A signature on an agreement and a signature on a cheque serve different operations. The beta covers scanned documents generally; the expected fields, reference source and next action should fit each document type.
Define the applicant, account holder or authorised-signatory fields your process requires. Include optional fields and multi-signer layouts in the evaluation.
Keep signature regions attached to the relevant page and field. Review customer and countersigner requirements separately from how many marks were found.
Check the required signature areas before the next operational step. Agree when presence is enough and when reference comparison is needed.
Include the drawer's signature and relevant endorsement areas in your scope. Combine signature evidence with your existing cheque and transaction checks.
THE CASES THAT MATTER
Clean scans show the happy path. Faint ink, clipped strokes, nearby handwriting and stamps show where a workflow needs help. Include those cases from the start, with the outcomes your reviewers expect.
A region may be visible but too obscured for comparison. An empty detection result can mean a missing signature, a missed candidate or an unusable image. Keep those possibilities separate when deciding whether to recapture or review the document.
ILLUSTRATIVE DOCUMENT WORKFLOW
Consider a fictional account mandate requiring one account-holder signature. These examples explain a review process; they are not measured beta results or API status names.
Detection, comparison and validation answer different questions. Keep their outcomes distinct and assign ownership of the final document decision within your operation. Visual similarity provides comparison evidence; it does not establish that a named person signed this document.
| Situation | Question to resolve | Example next action |
|---|---|---|
| No candidate in the required field | Is the field blank, or is the scan unusable? | Inspect the page or request a clearer copy. |
| Candidate visible; no authorised reference | Does this process need presence only or comparison? | Complete the presence check or obtain the right reference. |
| Candidate and reference available | What does the comparison support under the agreed policy? | Apply the comparison review criteria. |
| Candidate partly hidden by a stamp | Is enough writing visible for a useful comparison? | Review the full page or recapture. |
PLAN THE BETA
Choose representative document layouts, identify required signature fields and describe the capture conditions your team sees. If verification is in scope, explain how authorised samples are obtained and associated with the intended signer.
Agree the signature data contract, validation responsibilities and acceptance criteria before deciding what to automate. Measure candidate detection, region placement, reference comparison and review workload separately, using evaluation documents held apart from calibration data.
QUESTIONS & ANSWERS
Yes. ChequeDB Signatures is launching in beta for scanned documents generally, including forms, agreements, mandates, delivery notes, claims documents and cheques. Evaluate the specific layouts and capture conditions in your workflow.
No. Detection and positioning identify candidate regions in the submitted document image. Verification adds comparison with authorised samples for the intended signer.
Detection asks whether a region appears to contain a handwritten signature. Verification compares that writing with supplied references. Validation then applies the agreed document checks and handling rules.
It provides image-based comparison evidence. The reference process, image quality and operational review still matter. A scan does not record the original signing dynamics or independently prove who signed.
Request a beta evaluation through the ChequeDB team. Bring representative document types, your reference-signature process if comparison is required, and the decisions your workflow needs to make.
CHEQUEDB SIGNATURES · BETA
Discuss your document types, signature references and validation workflow with the ChequeDB team.