A crosscheck verifies that the data that is listed in multiple data sources (Visual Inspection Zone, MRZ, barcode, magnetic stripe, or contactless chip) is the same. If a data source is unable to be read, the test will not include that source. This process includes a method for allowing common misreads and transliterations, such as an accented or unaccented version of the same character. A document can fail based on the collective frequency of failed crosschecks, which would not cause a document to fail as individual tests. Crosschecks are used to detect significantly diverse types of data, with the intention of ignoring common misreads and transliterations, such as an accented and unaccented version of the same character. Some crosschecks are dependent on knowledge of the IDs of the fields they are testing. For example, they cannot be used to compare a field on the front of a document to a field on the back.

Universal crosschecks perform more extensive testing. They look up the fields to crosscheck by prefix and name at runtime and use the inherent reliability of those data sources to determine how important a mismatch is. The following individual crosschecks may be performed on a document:

In addition, the Document Crosscheck Aggregation test compares the results of multiple crosschecks and can fail a document based on the collective frequency of failed crosschecks that would not in itself fail a document.

The following table describes some of the authentication tests. This is not a comprehensive list and not every test is executed on every type of document.

Note: these names are intended to be descriptive and do not represent actual authentication test names.

*Text, photos, and other features on a document may not be visible under near-infrared, depending on the type of ink used. The presence or absence of these features may indicate that the document has either been tampered with or is a counterfeit.

The Document Library is an extensive compilation of document-specific metadata (image and layout characteristics) and authentication tests. The library now includes over 7,000 document types and is frequently updated with additional document types to ensure broad and reliable coverage.

For users of compatible systems, the Document Library is regularly updated and available for download and installation as a standalone package. Updates ensure that the library remains current, enhancing document recognition and authentication capabilities.

The Document Library's operational performance is enhanced through a data collection program that gathers anonymous metrics. These metrics help diagnose operational performance and improve the recognition and authentication of documents supported by the library.

Data Privacy and Security:

The only document-related information collected is the document type (e.g., "USA ePassport Series 2006"). This information is critical for correlating performance issues with specific document types, enabling targeted improvements to the Document Library to enhance performance, recognition, and authentication reliability.

PALIDIN performs several tests on a document to determine its authenticity. Each document type in the Document Library contains a defined set of individual authentication tests that is relevant for that document type only. A typical document will be subjected to 10–50 individual authentication tests.

The failure of a single authentication test will not necessarily result in overall document authentication failure. The sensitivity of the individual tests may be reduced, resulting in fewer false rejections (authentic documents being flagged as suspect). In most cases, fraudulent documents will tend to fail several authentication tests.

If a document could not be classified, it could indicate a problem such as a blurred image, bad position on the reader, or other issues associated with the device used to scan the document. In some cases, it might mean that PALIDIN was unable to find a supported document type to match.

A process that inspects a document in a controlled environment and performs extensive tests on the document’s physical layout and security features. It then reports the results, which reflect the differences between what is known about a given document type and an actual document of that type being examined. Despite this advanced authentication technology, however, the quality of the result is still dependent on the interpretation of the test results by the operator or application.

An optical, machine-readable representation of data that typically describes the object that carries the barcode. PALIDIN supports the PDF-417 stacked barcode used for RealID-compliant driver’s licenses and state-issued identification cards that meet the American Association of Motor Vehicle Administrators (AAMVA) standard.

The currently standardized biometrics used for document identification are facial recognition, fingerprint recognition, and iris recognition. The International Civil Aviation Organization (ICAO) defines the biometric file formats and communication protocols used for passports. Note that only the digital image (usually in JPEG or JPEG2000 format) of each biometric feature is stored in the chip. The comparison of biometric features is performed outside the passport chip by electronic border control systems (e-borders).

Also known as an e-passport, ePassport, or a digital passport, a combined paper and electronic passport that contains biometric information which can be used to authenticate an individual’s identity. See contactless chip.

The scanning or reading of an image that is used to authenticate a document. PALIDIN requires at least one image to be captured using a white light source of the front side of the document. We strongly recommend capturing both sides of a two-sided document for optimal verification results.

Classification is the identification of a document type based on its presentation and orientation. The presentation indicates the front or back side of a document. The orientation indicates whether the top edge of the document is at the top of the captured image.

A microprocessor chip and antenna (for power to both the chip and for communication) that is embedded in either the front or back cover, or on the center page of the passport. Public Key Infrastructure (PKI) is used to authenticate the data stored electronically in the passport chip, making it expensive and difficult to forge when all security mechanisms are fully and correctly implemented.

A crosscheck verifies that the data that is listed in multiple data sources (Visual Inspection Zone, MRZ, barcode, magnetic stripe, or contactless chip) is the same. If a data source is unable to be read, the test will not include that source. This process includes a method for allowing common misreads and transliterations, such as an accented or unaccented version of the same character. A document can fail based on the collective frequency of failed crosschecks, which would not cause a document to fail as individual tests.

PALIDIN supports generic and non-generic document types. The generic type is an ICAO identification card, the non-generic type is an ID1 non-ICAO identification card.

Extraction is the process of accessing and obtaining information that is read from text fields, image fields (such as a photograph) or contained in a 1D- or 2D-barcode, a magnetic stripe, or contactless chip.

An area of a document from which PALIDIN extracts (accesses, reads, and obtains) biographic and data contained in OCR-readable text, a 1D barcode, 2D barcode, magnetic stripe, or contactless chip.

Generic types are those which PALIDIN can partially but not fully identify. For example, PALIDIN can identify an ICAO Passport and run certain tests on it without knowing the exact type.

The International Civil Aviation Organization (ICAO) is a specialized agency of the United Nations which defines standards for international travel documents, international passports, and IDs, to ensure document standardization for the layout and location of biographic data and images.

The issuer is the government jurisdiction (country, state, or province) that issued the document.

PALIDIN supports the following light source types: visible (white), near-infrared, and ultraviolet.

A type of document manufacturing technique that uses special laser technology to create minute holes right through a document’s substrate (front and back). The holes are arranged to form an image that becomes clearly visible to the naked eye when the document is held up to the light. It cannot be copied or mechanically reproduced.

PALIDIN supports 12 languages for localized strings in the Document Library for standard fields, standard authentications, document issuers, and document classes. Note that localization is not the same as translation.

The machine-readable zone (MRZ) is an area of a document, such as a passport and some IDs, that contains encoded information, including the type of passport, passport ID, issuing country, name, nationality, and expiration date. It is typically located at the bottom of the identity page of a passport.

Some documents (such as passports) contain information in two or more languages, typically one that uses a Latin-based alphabet and another that uses a non-Latin alphabet. PALIDIN returns the Latin version as the normal field and the secondary language version as the native field.

Near-infrared (NIR) is the region closest in wavelength to the radiation detectable by the human eye. NIR is sometimes called reflected infrared.

Non-image data includes binary data such as magnetic stripes and machine-readable zone (MRZ) barcodes, or contactless chip information.

Optical character recognition (OCR) is a common method of digitizing printed texts so that they can be electronically edited, searched, and stored as machine-encoded text.

Public Key Infrastructure (PKI) is used to authenticate the data stored electronically in the passport chip, making it expensive and difficult to forge when all security mechanisms are fully and correctly implemented.

The series is a label that identifies the version of a document when the format or design changed. It may or may not correspond to the date of issue.

The document size represents the physical size of a document; for example, ID1 is a credit-card sized document such as a driver's license and ID3 is a passport.

The Visual Inspection Zone (VIZ) is a field that is scanned and read using optical character recognition (OCR).