SE Documentation

English

Русский

English

Русский

Appearance

en / smart-document-engine

Smart Document Engine. Recognition of Documents ​

The Smart Document Engine program is designed for recognition of various documents, such as personal documents, certificates, statements and so on in more than 100 languages.

State registration

The Smart Document Engine program is registered in the State Register of Computer Programs. The certificate of state registration No 2020616760. [1]

The program works in real conditions and in real time with photos and scans.

You can search, verify, and classify all documents presented in a single image as well as multipage documents.

Automatic search, selection and recognition of document details: tables and lists, banking details are also available. It is possible to check document texts by keywords and phrases.

For detailed description of the Smart Document Engine features see here.

The program is resistant to original image quality. Even if documents are folded or bent, they are automatically smoothed.

Moreover, you can configure document forms for recognition using a special application — Document Studio Designer.

The use of the latest GreenOCR® technologies and artificial intelligence developed by our company ensures high recognition confidence. In order to increase quality, we created ultra-compact, low-bit artificial intelligence models and are continuously improving them, as well as the product itself.

Due to the modular structure of the program, the size of the library is not large. It makes the library suitable for integration with variety of mobile and server systems.

Delivery Package ​

  • API in C, С++, C#, Java, Python, PHP, Javascript/webAssembly;
  • Samples in C/C++/C#/Java/PHP/Python/Javascript;
  • Frameworks: React Native, Flutter;
  • REST API for low-code.

Advantages of Smart Document Engine ​

Information Security ​

No risk of data leakage. All recognition operations are fully automatic and do not impy the human factor. Recognition is performed offline on mobile phones or in the customer's network without manual verification or any operator intervention. No network connection is required to work with our product. All processing is done in the local RAM of the devices (100% on-premise).

Recognition Quality and Speed ​

High recognition speed: from 1 second per page on a smartphone to 15 pages per second on a server. High confidence of recognition of documents, coordinates of objects, supports cropping fields and documents.

You can get a good result even at a low image resolution (from 100 DPI) and in uneven lighting, in bright light and in the dark, at any shooting angle.

Integration Possibilities ​

A wide range of operating systems is supported: Aurora OS, Android, iOS, MS Windows, Linux, Elbrus OS, RED OS, Astra Linux, Atlix OS, Alt Linux Aurora OS, х86, x86/64, ARMv7-v8-v9 (AArch32 and AArch64), MIPS (MIPS32 and MIPS64), Elbrus.

Universality of recognized documents ​

Automatic document entry and configuration of new documents for the order are supported. Seamless integration with the Smart ID Engine for recognition of identity documents. This allows you to work with almost any type of documents.

Possibility to Create Custom Templates ​

Document Studio Designer, a special application that allows you to set:

  • Recognition of forms with a specified geometry using a single template;
  • Recognition of printed, handprinted and handwritten document details;
  • Recognition of fill marks, checkboxes, barcodes;
  • Document structure;
  • Rules for formatting and verifying data.

You can create multiple templates for the same page type.

No special skills are required for working with the application: a user-friendly graphical interface makes all operations easy. A wide range of customized documents and built-in full-text OCR helps you create your own models.

Functionality of Smart Document Engine ​

1. Detection and Recognition
  • Recognition of typescript text fields;
  • Recognition of printed text independently on the font type;
  • Recognition of handprinted text;
  • Recognition of handwritten text;
  • Recognition of single line and multiline text fields;
  • Recognition of all fields of the document and part of the fields from the document specified in the input parameter;
  • Recognition of tables;
  • Recognition of fill marks and checkboxes;
  • Recognition of 2D and linear barcodes;
  • Recognition of document scans;
  • Recognition of document photes;
  • Recognition of color (RGB), single-channel (grayscale) images;
  • Recognition of documents positioned at any angle;
  • Recognition of multipage documents;
  • Identification of the document type by its image and recognition of its fields;
  • Recognition of documents without using ROM (without saving temporary and intermediate data on ROM);
  • Recognition of information without accessing resources external to the device (application servers, online services, etc.);
  • Detection and recognition of data in graphic elements (seals, stamps, logos, etc.);
  • Classification and recognition of rigid forms;
  • Classification and recognition of flexible forms.
2. Control, verification, analysis
  • Analysis of multi-page documents;
  • Analysis of any type of documents;
  • Document completion control (understanding paragraphs, analyzing key areas of the document);
  • Checking for blots, corrections, and other marks;
  • Checking the availability and Recognition of handwritten notes and signatures;
  • Checking for keywords and phrases;
  • Control of document color;
  • Analysis of attributes of text fields (serifs, italics, uniformity and monotony ratings);
  • Checking availability of graphic elements (checking the presence of necessary elements);
  • Checking availability of signatures on the document.
3. Technological features
  • Highlighting preset graphic fields of a document;
  • Highlighting non-standard graphic areas of the document;
  • Return of document zones and individual details for each input image;
  • Ability to control the number of computing threads used;
  • Support of a wide range of acceptable physical resolution of recognized documents from 100 to 600 dpi and above;
  • Using 8-bit and 4-bit deep neural network architectures for recognition of text and graphic elements.

Supported File Formats ​

The following file formats are supported:

Pass an image of the special class se.common.image to the system for recognition. You can create it using the following image formats:

  • jpeg, png;
  • tiff (✔️TIFF_LZW, ✔️TIFF_PACKBITS,✔️TIFF_CCITT);
  • base64 (above mentioned formats);
  • file buffer with a preliminary indication of the color scheme, width\height\number of channels.

The maximum allowed image size by default is 15000x15000px. You can change the maximum image size.

Handling a HEIC file

A HEIC file in the mobile SDK are handled similarly to other image formats. The HEIC is read using system tools.

In the server SDK, open the HEIC format using external tools and convert it either to one of the formats we support, or transfer the raw pixels directly as an RGB buffer (this is recommended).

Recognized Documents ​

Smart Document Engine supports recognition of the following document types:

  • Certificate of income of an individual (2-Person income tax);
  • Certificate 1C;
  • Certificate of rights transfer;
  • Act of reconciliation;
  • Bank warranty;
  • Balance sheet;
  • Certificate of entry into the Unified State Register of Individual Entrepreneurs;
  • Certificate of entry into the Unified State Register of Legal Entities;
  • TIN of the legal entity;
  • Collection order;
  • Corrected invoice;
  • Invoice TORG-12;
  • Invoice TORG-13;
  • Payment order;
  • Payment request;
  • Payment warranty;
  • Reference KS-3;
  • Certificate of registration (deregistration) of an individual as a taxpayer of professional income tax;
  • Invoice 1C;
  • Invoice;
  • Consignment note;
  • Universal correction document;
  • Universal transfer document.

The full list of the supported documents is contained in a separate file.

Languages Supported by Smart Document Engine ​

  • Cyrillic
    (Russian, Abkhaz, Belarusian, Bulgarian, Kazakh, Kyrgyz, Mongolian, Ossetian, Serbian, Tajik, Uzbek, Ukrainian, Tatar).

  • Latin
    (English, Azerbaijani, Aymara, Albanian, Afrikaans, Berber, Bosnian, Hungarian, Wole, Haitian Creole, Danish, Zulu, Indonesian, Irish, Icelandic, Spanish, Italian, Cantonese, Castilian, Catalan, Quechua, Comorian, Xhosa, Latvian, Lithuanian, Luxembourgish, Macedonian, Malay, Maltese, Mandarin, Maori, Moldovan, Nauruan, Ndau, Ndebele, German, Dutch, Norwegian, Palau, Polish, Portuguese, Rwanda, Romanian, Rundi, Swazi, Northern Soto, Seychelles Creole, Sesoto, Slovak, Slovenian, Somali, Swahili, Tamazite, Tetum, Tok Pisin, Tonga, Tswana, Tsonga, Turkish, Turkmen, Fijian Hindi, Fijian, Filipino, Finnish, French, Croatian, Ceva, Czech, Chibarwe, Chichewa, Shangani, Swedish, Shona, Estonian, South Ndebele).

  • Others
    Chinese (Classical, 20,000 characters), Japanese (Hiragana, Katakana and Kanji), Korean (Hangul), Sinhalese, Greek, Georgian, Armenian, Hebrew, Arabic, Farsi, Urdu, Tamil, Thai, Bengali.

  1. Registered in the State Register of Computer Programs. The certificate of state registration No 2020616760. Included in the Unified Register of Russian programs for electronic computers and databases (entry in the register No. 9617 dated 15.03.2021 was made on the basis of the order of the Ministry of Digital Development, Communications and Mass Communications of the Russian Federation dated 15.03.2021 No.151) https://reestr.digital.gov.ru/search/?q=2020616760. ↩︎

Copyright © 2025 SmartEngines