Automated Image Analysis for Account Migration

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional account migration processes are time-consuming and prone to errors, leading to issues like lost information, late transactions, and costly corrections due to insufficient funds, confusion between source and destination accounts, and technical issues.

Innovation Solution

The method involves analyzing images of documents, such as bills, to identify transaction information using optical character recognition and machine learning, and scheduling transactions between accounts to migrate recurring payments from a first account to a second account, thereby automating the account migration process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual account migration processes are used, then flexibility and control are maintained, but time consumption and error rates increase

Engineering Contradiction:

Improveaccuracy of account migrationVSAvoidtime for account migration

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical processes (handwritten checks, physical document handling, manual data entry) with automated optical scanning and image recognition systems. The scanner captures images of checks and documents, automatically extracts data through pattern recognition, and electronically processes transactions, eliminating the need for manual handling while improving both speed and accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service account migration by automatically scanning, recognizing, and processing transaction data without requiring manual intervention. The image recognition system autonomously extracts account information, routing numbers, and transaction details from captured images, and the system automatically schedules and executes migrations, reducing both time and error rates.

Inventive Principle:

Principle #25Self-service

2Productivity

If automated image analysis is implemented, then processing speed and accuracy improve, but system complexity increases

Engineering Contradiction:

Improvetransaction processing speedVSAvoidcomplexity of image analysis system

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by pre-scanning documents and creating image archives before actual processing. The image recognition system is pre-trained with sample documents and patterns, allowing it to quickly recognize and extract data during actual transactions. This preliminary preparation reduces processing time during live operations while managing complexity through advance setup.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary image recognition layer between physical documents and the account migration system. This intermediary automatically converts physical check images into structured data, serving as a bridge that simplifies the overall system architecture. The intermediary handles the complexity of image analysis internally, presenting clean, processed data to the migration system without requiring direct complex interactions.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If traditional manual entry methods are used, then system simplicity is maintained, but error rates and correction costs increase

Engineering Contradiction:

Improveaccuracy of transaction dataVSAvoidease of implementing migration system

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The system creates accurate digital copies of physical documents through high-resolution scanning. These digital copies serve as precise replicas that can be analyzed without handling the original physical documents. The copying process captures all necessary data (account numbers, routing numbers, amounts) in a format that can be automatically processed, eliminating transcription errors while maintaining data accuracy.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces manual data entry mechanics with automated optical recognition mechanics. Instead of human operators manually typing data from physical documents, the system uses scanners and image recognition algorithms to automatically capture and process information. This substitution eliminates human error in data entry while reducing the complexity of manual handling procedures.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11574493B1Image data extraction for transaction management

Publication Date: 2023.02.07 UNITED SERVICES AUTOMOBILE ASSOCIATION (USAA)

AI summary

Techniques are described for migrating information from a first account to a second account, based on analyzed image(s) of document(s). Image(s) of a document may be generated using an image capture device of a smartphone or other portable computing device. The image(s) may be analyzed, through pattern recognition analysis or barcode scanning, to extract the information from the image(s). The information may then be employed to schedule a transaction, such as payment of a bill described in the information. In some instances, the extracted information may be used as part of an account migration process, in which transactions are migrated from a first account to a second account.