Remote Check Certification via Image Capture and OCR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current check certification methods are limited to being performed only at a bank, restricting the ability to certify checks from other locations such as personal computers, mobile phones, or ATMs.
Innovation Solution
A system and method that allows check certification via personal computers, mobile devices, or ATMs, where an image of the check is captured, and using optical character recognition to verify account details and funds availability, issuing a certification number and password for remote verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If check certification is performed only at a bank, then security and control are maintained, but accessibility and convenience are reduced
Solution Approach 1:
The patent uses image capture devices to create digital copies of physical checks. The system captures images of the check front and back, extracts relevant data through OCR and image processing, and uses these digital representations for certification verification. This eliminates the need for physical check handling while maintaining certification integrity.
Solution Approach 2:
The patent replaces traditional mechanical bank-based certification processes with automated image processing and optical character recognition systems. The system uses computer vision algorithms, pattern recognition, and automated fund verification to substitute the manual bank teller process, enabling remote certification while maintaining security controls.
2Ease of operation
If remote check certification is enabled, then convenience is improved, but verification accuracy and security may be compromised
Solution Approach 1:
The patent introduces multiple intermediary layers between the user and the certification decision. These include image capture devices that standardize input quality, OCR systems that accurately extract data, pattern recognition algorithms that verify check authenticity, and automated fund verification systems that confirm account status. Each intermediary layer enhances verification accuracy while enabling remote operation.
Solution Approach 2:
The system implements comprehensive feedback mechanisms including image quality assessment that prompts users to recapture blurry images, OCR confidence scoring that requests manual verification when uncertain, pattern matching feedback that identifies potential fraud, and real-time fund availability verification. This feedback loop ensures high verification accuracy while maintaining user-friendly remote access.
3Productivity
If automated image processing is used, then processing speed is improved, but system complexity increases
Solution Approach 1:
The patent divides the check certification process into distinct segmented modules: image capture, image preprocessing, OCR data extraction, pattern recognition for authenticity verification, fund availability checking, and certification decision-making. Each module is independently optimized and can be processed in parallel, improving overall speed while managing complexity through modular architecture.
Solution Approach 2:
The system performs preliminary actions including image quality validation before processing, pre-extraction of check data using OCR, pre-verification of account status, and pre-computation of certification decisions. By preparing and validating data beforehand, the system reduces actual processing time while maintaining comprehensive verification through structured preliminary processing steps.
Data Source
AI summary
A system and method of certifying a check which involves a captured image of the check. An example method includes receiving by a first computer a request from a second computer to certify a check from a payer to a payee, the check serving to withdraw funds from an account of the payer, receiving by the first computer an image file containing an image of the check from the second computer, obtaining an account number and an amount from the image file by the first computer, determining using the account number whether funds equal to the amount are in the account, and when funds equal to the amount are in the account, putting a hold on the funds equal to the amount and issuing a certification number and a password by the first computer.