smart_document_capture 0.1.0
smart_document_capture: ^0.1.0 copied to clipboard

→

Extract structured data from document, receipt and form images with per-field confidence, manual review and pluggable OCR engines.

0.1.0 #

Initial release.
OcrEngine interface, OcrResult model and FakeOcrEngine.
DocumentSchema, FieldDefinition and locators: keyword, pattern, region and header.
Per-field confidence (DefaultConfidenceModel), alternatives and confidence levels.
Validators: text, regex, number, amount, time, Jalali date, Iranian national code, bank card, Sheba and mobile number.
ReceiptSchema.persian and KeyValueSchema.
ExtractionResult with corrections and JSON output.
DocumentReviewView for manual review and correction.