API SuiteDocument OCR (ML)
Passport Back OCR
OCR the endorsement or back page region of a passport from an image. Server-side multipart POST with Bearer auth.
API reference
Try itLoading playground…
Loading…
AuthorizationBearer <token>
JWT Bearer token authentication. Obtain a token from the KwikID dashboard.
In: header
filestring
flags?string
Optional JSON string. Keys: CARD_ORIENT (0/1) corrects orientation; CARD_CROP (0/1) runs document crop before OCR.
unique_id?string
Optional correlation id for logs.
Response Body
curl -X POST "https://__mock__/ocr/passport/back" \ -F file="string"{
"address": "string",
"father_name": "string",
"mother_name": "string",
"old_passport_date_of_issue": "string",
"old_passport_number": "string",
"old_passport_place_of_issue": "string",
"passport_number": "string",
"spouse_name": "string"
}{
"detail": {
"<location>": {
"<field_name>": [
"string"
]
}
},
"message": "string"
}{
"detail": {},
"message": "string"
}Overview
Call POST /ocr/passport/back with Authorization: Bearer <token> and file. Optional flags and unique_id. Use when your process requires endorsements or secondary stamps captured from the back.
Key features
- Complements front OCR for full document digitization.
- Same multipart contract as other OCR routes.
Implementation
Step 1: Call from your backend
Upload back page image as file.
Error handling
| HTTP status | When |
|---|---|
| 400 | Unreadable image. |
| 401 | Invalid token. |
Benefits
- Supports journeys that require both passport sides.