Image-Matching DLP for Sensitive Data Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional data-loss-prevention (DLP) systems face challenges in accurately detecting sensitive information transmission due to the resource-intensive and inaccurate nature of optical character recognition (OCR) techniques, especially when dealing with diverse data transmission methods.

Innovation Solution

The system treats all documents as images and uses image-processing techniques to extract features for comparison with a gallery of protected files, employing image-matching methods to identify and block attempts to transmit sensitive information via data-distribution channels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If OCR techniques are used to examine outgoing documents, then sensitive information can be detected, but the process becomes resource intensive and inaccurate

Engineering Contradiction:
Improvedetection accuracyVSAvoidresource consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent replaces OCR (optical character recognition) with image-matching techniques. Instead of converting images to text and analyzing text content, the system directly compares document images using image-matching algorithms. This substitution eliminates the resource-intensive OCR process while maintaining detection capability, as image-matching operates directly on visual patterns without requiring text conversion.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If OCR techniques are used to examine outgoing documents, then sensitive information can be detected, but the detection becomes inaccurate

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetection reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The system replaces OCR with image-matching techniques that directly compare visual patterns. This substitution improves reliability because image-matching algorithms can detect sensitive information based on visual characteristics without the errors introduced by text recognition, especially for handwritten or complex formatted documents where OCR struggles.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If traditional DLP systems examine documents using OCR, then policy enforcement is attempted, but the process is slow and resource intensive

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidcomputational resources
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent substitutes OCR-based text analysis with image-matching-based visual comparison. This replacement significantly reduces computational resource requirements because image-matching algorithms can quickly compare visual patterns without the complex text recognition and processing steps required by OCR, thereby improving processing efficiency while lowering energy consumption.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

4Use of energy by moving object

If image-matching techniques are used instead of OCR, then resource consumption is reduced, but the system must handle diverse document formats

Engineering Contradiction:
Improveresource consumptionVSAvoidformat compatibility
Core Design Contradiction:
Use of energy by moving objectVSAdaptability or versatility

Solution Approach 1:

The system implements a universal image-matching approach that can handle diverse document formats (PDF, images, Word documents, etc.) by converting them all to image format for comparison. This multi-functional capability allows the same image-matching algorithm to process any document type uniformly, reducing resource consumption while maintaining broad format compatibility through a standardized image-based processing pipeline.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3192001B1Systems and methods for detecting attempts to transmit sensitive information via data-distribution channels
Publication Date: 2021.01.13 CA TECH INC
  • EP3192001B1 patent drawingFigure 1
  • EP3192001B1 patent drawingFigure 2
  • EP3192001B1 patent drawingFigure 3

AI summary

The disclosed computer-implemented method for detecting attempts to transmit sensitive information via data-distribution channels may include (1) identifying an attempt to transmit a file through a data-distribution channel, (2) comparing, using an image-matching technique, the file with at least one known sensitive file that is both stored in an image format and protected by a data-loss-prevention policy, (3) determining, based on the results of the image-matching technique, that the file violates the data-loss-prevention policy, and (4) performing a security action in response to determining that the file violates the data-lossprevention policy. Various other methods, systems, and computer-readable media are also disclosed.