Developers: | Cognitive Technologies (Cognitive technologies) |
Last Release Date: | 2013/11/21 |
Technology: | Cybersecurity - Biometric identification |
Content |
2013: Cognitive Passport 3.0
On November 21, 2013 the Cognitive Technologies company announced a release of the new version of Cognitive Passport 3.0.
In this version the quality of recognition of new Russian passports of A6 format grew by 20%. It in many respects became possible thanks to implementation of new technologies of intellectual search, identification and removal of noises in the document. The textural background, lines, holograms and other elements on the color image which are quite often imposed on useful text information belong to noises and interfere with its recognition by traditional means.
The possibility of recognition of the machine-readable zone (MChZ) which appeared in the Russian passports of new generation, as well as in similar documents in many countries of the world is available. A main objective of introduction of zone MChZ is simplification of the procedure of reading and considerable reduction of time of reading of basic data. MChZ is also an additional measure of protection of the form of the passport from counterfeits. At the same time it performs function of correction of results of optical recognition of personal data of the passport and minimization of number of possible recognition errors. MChZ consists of two lines 44 characters long everyone.
According to the requirements and standards shown to machine-readable documents, MChZ contains the main information which is available in this document: a surname, a name, a middle name, date of birth (number, month, year) of the owner of the passport, a sex of the owner of the passport, a series and the passport number, the subdivision code which issued the passport, date of issue of the passport and also designation of document type, the state which issued the passport, nationality of the owner of the passport.
MChZ also contains target figures (checksums of values of certain fields of personal data) which allow to distinguish it with high quality. The font and also line and intersymbol distance of MChZ are tightly regulated. However, in practice at issue of passports production errors of several types (the most rough of them see rice 1, 2) quite often meet when only one of the MChZ elements is printed, or a part of a line oversteps the bounds of printing.
Only one part of MChZ of a line is printed
MChZ a line is printed not completely
The new version successfully copes with such shortcomings. The general quality of recognition of passports with MChZ grew approximately by 10%.
- At the existing high quality of recognition of the data which are contained in MChZ to a line increase in this indicator by 10% is very serious result, - Vladimir Arlazarov, the CEO of Cognitive Technologies noted.
Stability of system operation increased. Expansion of the Cognitive Passport API interface with preserving of backward compatibility will give to developers the chance - not to change proprietary applications under the new version.
Olga Uskova, the president of Cognitive Technologies emphasized: "In the field of the systems of recognition of difficult documents, as well as in other knowledge-intensive directions, very quickly there is "aging" of versions of software products. Therefore we continuously invest in it R&D and we try to propose to our users the best solution".
Version 3.0 is available to buyers since November 21, 2013. Users of the previous versions of Cognitive Passport can carry out upgrade free of charge.
2012: Cognitive Passport API 2.5
For 2012 more than 950 licenses of the Cognitive Passport system (dynamics of sales of this product was about 30%) were sold, and the general growth of sales volume in this direction was about 34% (against 23% in 2011).
The Cognitive Technologies company announced in September, 2012 a release of the new version of Cognitive Passport API - the program interface providing third-party applications with all systems capabilities of Cognitive Passport: scanning, identification and recognition of identity documents.
In the new version corresponding to number 2.5 the accuracy of recognition grew by 40%. It in many respects became possible thanks to implementation of new technologies of intellectual search and selection in the document of lines and holograms on the color image. Version 2.5 allows to identify in the document of a noise - lines and the holograms which are quite often imposed on useful text information and to delete them. The solution of a problem of selection of lines on the color image has the increased complexity, however developers were forced to go this way. As alternative and simpler option of image processing after binarization (on the b/w image) is integrated to considerable losses of useful information that as a result would affect result.
The most important result of developments became that they did not lead system performance to any losses.
It should be noted that so far the impossibility of department of lines and holograms from the text on the color image was the key problem interfering increase in accuracy of recognition of identity documents. According to the President of Cognitive Technologies Olga Uskova, "For the first time in Russia the serious scientific problem allowing to bring the systems of recognition of documents to the new quality level is solved".
The possibilities of intellectual search of persons and holograms implemented in version 2.5 in the document are also additional signs for ensuring identification of the document and therefore, allow to avoid practically errors at simultaneous scanning of several documents (passports, driver's licenses, a passport photocopy).
Emergence of the user interface allowing clients "with all conveniences" became other advantage of the new version to hold testing of Cognitive Passport API.
Let's remind that Cognitive Passport API is built in directly a corporate information system or the application and allows to implement the automated input of documents.
The version went on sale on September 5, 2012.
See Also
- National Biometric Platform (NBP)
- Technologies of biometric identification
- Biometrics and border control
- Sonda RTI
- IDensic
- VisionLabs
- ABBYY PassportReader
- Vocord FaceControl
- Biometric identification
- Gemalto Coesys Border and Visa Management