EFFECTIVE PROTECTION AGAINST FRAUD IN THE DIGITAL WORLD

Document Integrity Check is an image fraud detection solution to automatically check whether documents received from customers contain authentic information or not.

If yes, scroll down.

Sum&Substance utilizes AI based technology with unique proprietary modular architecture and asynchronously operating algorithms to reduce the number of checks through third-party databases.

icons-4-1

Fast

Automatic verification of a document takes less than a second. This makes it easier and faster the process of working with digital documents fast and enjoyable. Only fraudsters suffer.

icons-4-2

Efficient

Remove on average 15% of bad contractors at the stage of uploading documents. Increase the level of confidence in the digital documents.

icons-4-3

Easy and convenient

Easy and convenient integration allows you to improve processing of digital documents without changing the business processes of your company.

HOW IT WORKS

scheme-idtool

Data

All photos (JPEG, PDF, PNG) received from companies are automatically checked for digital authenticity using our API or manually via a web service.

Technology

The biggest part of technical information about photos is hidden from an ordinary user. Our technology allows not only to view, but also to automatically analyze these “invisible” data, comparing them to the reference database consisting of millions of photos and thousands of different photographic devices.

Result

Once analysis is done, each document gets a scoring rating of the level of trust. Photos with low trust level immediately receive a negative decision. Photos with high or medium trust level go through, according to the internal business processes of your company.

50 free tests available. All features included.

KEY FEATURES

There are four major components that make up Document Integrity Check’s system – a unique technology that helps Sum&Substance be useful and keep improving.

platform-modules-blue

Digital Footprint

In order to validate the image authenticity or estimate the risk of an intended tampering (as opposed to cosmetic changes like cropping, resizing or rotation) we utilize an extensive database of cameras and software signatures.

A signature includes not only file metadata, but also compression parameters, vendor – or software – specific tags or sections, etc. – that allows to determine the source of an image or a modification software with very high accuracy.

ind-analisys-s

Pattern Matching

We collected a huge amount of template documents that people use to scam someone or just hide their real identity. All incoming documents are checked against our constantly updated database of these fraudulent patterns, greatly improving the efficiency of the entire system. As a result it greatly improves the efficiency of the whole system.

digital-footprint

The method utilizes additional tools and algorithms to analyze the pixel structure of an image and to identify abnormal areas with possible traces of retouching and editing.

Stamp

The method allows to determine the copied and pasted areas within the perimeter of an image.

stamp-docus

JPEG Ghosts

In order to determine edited areas the algorithm analyzes the image for the presence of areas with different levels of compression. The result comes as a mask or a probability map.

ghosts-docus-3
GHOSTS-docus-4

BAG (Block Artifact Grid)

The algorithm is based on the characteristics of the JPEG compression algorithm and allows in some cases to detect artifacts by analyzing the image for abnormalities in the block artificial grid (thickening of the disappearance, shift, etc.).

BAG-docus

Group Analysis

The method allows to compare multiple documents within one group. Group conformity of the parameters being checked: image creation time, camera model, shooting location, digital footprints of the photos. This makes more accurate conclusions about the authenticity of the photos, on the basis of their membership in one of the criteria, the same lighting
conditions at one time and place of the capture.

group_analysis
+ Digital Footprint

Digital Footprint

In order to validate the image authenticity or estimate the risk of an intended tampering (as opposed to cosmetic changes like cropping, resizing or rotation) we utilize an extensive database of cameras and software signatures.

A signature includes not only file metadata, but also compression parameters, vendor – or software – specific tags or sections, etc. – that allows to determine the source of an image or a modification software with very high accuracy.

ind-analisys-s
+ Pattern Matching

Pattern Matching

We collected a huge amount of template documents that people use to scam someone or just hide their real identity. All incoming documents are checked against our constantly updated database of these fraudulent patterns, greatly improving the efficiency of the entire system. As a result it greatly improves the efficiency of the whole system.

digital-footprint
+ Editing Areas Search

The method utilizes additional tools and algorithms to analyze the pixel structure of an image and to identify abnormal areas with possible traces of retouching and editing.

Stamp

The method allows to determine the copied and pasted areas within the perimeter of an image.

stamp-docus

JPEG Ghosts

In order to determine edited areas the algorithm analyzes the image for the presence of areas with different levels of compression. The result comes as a mask or a probability map.

ghosts-docus-3
GHOSTS-docus-4

BAG (Block Artifact Grid)

The algorithm is based on the characteristics of the JPEG compression algorithm and allows in some cases to detect artifacts by analyzing the image for abnormalities in the block artificial grid (thickening of the disappearance, shift, etc.).

BAG-docus
+ Group Analysis

Group Analysis

The method allows to compare multiple documents within one group. Group conformity of the parameters being checked: image creation time, camera model, shooting location, digital footprints of the photos. This makes more accurate conclusions about the authenticity of the photos, on the basis of their membership in one of the criteria, the same lighting
conditions at one time and place of the capture.

group_analysis

HAVE ANY QUESTIONS?

Connect with us – we welcome your questions and feedback.