Yandex Vision OCR
A computer vision service for text recognition and extraction.
Text recognition
Yandex Vision OCR recognizes text in an image and outputs it along with automatic punctuation. The service supports and automatically identifies 48 languages.
Extracting text from documents
Extract standard fields and recognize text in templates and documents, e.g., passports, driver’s licenses, vehicle registration certificates, and license plates.
Handwriting recognition
With support for Russian and English, as well as combinations of handwritten and printed texts.
Table recognition
The service scans the table structure and outputs text in row and column coordinates.
Implement your projects using Yandex Vision OCR
Automate your workflow
Automate business processes involving scanning and recognizing text in standard document types and add them to your document workflow system. The service recognizes text and analyzes which part of a document it should be placed in.
Search for text in scanned document archives
If you need to find specific information in an archive of scanned documents only saved as images, Yandex Vision OCR helps you recognize text and makes it available for indexing and searching.
Automatically translate text from images
Find and immediately translate text in images to other languages by using Yandex Vision OCR and Yandex Translate in tandem. To embed this feature in your apps, use the REST or gRPC API that both services provide.
FAQ
What types of analysis does Yandex Vision OCR support?
What types of analysis does Yandex Vision OCR support?
The service supports the following features: optical character recognition (OCR), document recognition and license plate number recognition.
What file formats does the service support? What limitations are there?
What file formats does the service support? What limitations are there?
Which document templates does the service support?
Which document templates does the service support?
