Document Assembly Checks for OCR-Based Fraud Verification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing document verification systems struggle to efficiently and accurately authenticate documents, particularly in the face of sophisticated fraudulent attempts, such as those using photo editing software, leading to potential identity theft and security breaches.

Innovation Solution

A document evaluator system that utilizes optical character recognition (OCR) and object detection to generate a document assembly object, which includes bounding boxes and derived checks, enabling rapid validation of documents by analyzing features like microprint and ghost images, and adjusting for false positives/negatives through machine learning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional document verification methods are used, then the verification process is simple to implement, but the accuracy of detecting sophisticated fraudulent documents is insufficient

Engineering Contradiction:

Improvedocument verification accuracyVSAvoidverification system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The verification system segments the document analysis into multiple independent checks: OCR text extraction, bounding box detection, derived checks (microprint, ghost image, barcode validation), and machine learning-based authentication. Each segment focuses on specific document features, improving overall verification accuracy while maintaining modular implementation

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by pre-generating a document assembly object containing all expected bounding boxes, text regions, and derived checks before actual verification. This pre-computed reference framework enables faster and more accurate comparison during the actual verification process, reducing computational complexity in real-time operation

Inventive Principle:

Principle #10Preliminary action

2Reliability

If multiple derived checks are generated to improve fraud detection, then the detection accuracy increases, but the processing time and computational resources increase

Engineering Contradiction:

Improvefraud detection reliabilityVSAvoiddocument processing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system implements a hierarchical verification approach where critical derived checks (barcode validation, ghost image detection, microprint analysis) are performed first on all documents. Less critical checks are performed selectively based on initial results or document risk assessment, reducing average processing time while maintaining high detection reliability for sophisticated fraud

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system replaces manual or simple automated verification with machine learning models that can perform multiple derived checks simultaneously. The ML-based object detection and authentication algorithms process multiple document features in parallel, reducing overall processing time while improving fraud detection reliability compared to sequential traditional methods

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If the system analyzes multiple features like microprint and ghost images, then the ability to detect sophisticated fraud improves, but the complexity of implementing and maintaining the system increases

Engineering Contradiction:

Improvedocument feature analysis capabilityVSAvoidsystem implementation complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system employs a universal document assembly object structure that can accommodate multiple document types and their specific features (microprint, ghost images, barcodes, text fields). This unified framework allows the same verification infrastructure to handle diverse document features through configurable derived checks, reducing implementation complexity compared to separate systems for each feature

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The document assembly object serves as an intermediary layer between the complex feature analysis algorithms and the verification decision logic. It pre-organizes all extracted features (bounding boxes, text, microprint regions, ghost image locations) into a standardized format, simplifying the implementation and maintenance of the verification system by decoupling feature extraction from feature validation

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12499703B2Generating a document assembly object and derived checks

Publication Date: 2025.12.16 JUMIO CORP

AI summary

The disclosure includes a system and method for obtaining a document specification in an electronic format, wherein the document specification is associated with a first document, and describes features present in valid instances of the first document; determining a set of labels describing the first document from the document specification; obtaining one or more digital images of at least one valid instance of the first document from the document specification; obtaining information describing a set of bounding boxes resulting from application, to the one or more images of the least one valid instance of the first document, of one or more of optical character recognition and object detection; generating a set of derived checks based on the set of bounding boxes; and generating a document assembly object describing valid instances of the document and the set of derived checks usable to determine validity of a document under test.