Check Image Data Inference Processing for OCR Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic check processing systems face errors due to unreliable algorithms for reading check text, leading to issues like processing incorrect amounts or dates, which necessitate manual verification, increasing costs and defeating the efficiency of electronic processing.
Innovation Solution
Implementing systems and methods for check image data inference processing, which group checks by account, perform optical character recognition, and apply inference processing to missing or ambiguously read dates and amounts based on other checks in the group, using modules for check amount and date inference to update data accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual verification is used to avoid reading errors, then processing accuracy is improved, but processing cost increases
Solution Approach 1:
The system performs self-verification by using inference processing to automatically detect and correct reading errors. The algorithm compares OCR results against expected patterns and account group data, enabling the system to self-correct without manual intervention while maintaining high accuracy.
Solution Approach 2:
The system implements feedback mechanisms where inference results are used to validate and correct initial OCR readings. The process continuously refines data accuracy by comparing against account group patterns and providing corrective feedback loops that eliminate errors without manual verification.
2Reliability
If manual verification is implemented to correct reading errors, then data accuracy is improved, but processing efficiency deteriorates
Solution Approach 1:
The system automatically performs error detection and correction through inference processing, eliminating the need for manual verification steps. This self-service capability maintains high data accuracy while preserving the automated processing speed and efficiency of electronic check processing.
Solution Approach 2:
The system performs inference processing in advance to identify and correct potential reading errors before they propagate through the processing system. By proactively addressing accuracy issues, the system ensures high data accuracy without requiring subsequent manual verification that would reduce efficiency.
3Reliability
If inference processing is applied to all check images, then data accuracy is improved, but processing time increases
Solution Approach 1:
The system applies inference processing selectively only to check images where OCR confidence is below the threshold or errors are detected, rather than uniformly to all checks. This localized application of inference processing maintains high data accuracy for problematic cases while minimizing additional processing time for already-clear images.
Solution Approach 2:
The system performs inference processing partially, applying it only when necessary based on confidence thresholds and error detection, rather than excessively to all images. This partial application strategy achieves sufficient data accuracy improvement without the full time cost of universal inference processing.
Data Source
AI summary
Various embodiments herein each include at least one of systems, methods, and software for check image data inference processing. Another example method embodiment includes inferring a check amount of a check image included in an account group of check images stored in a memory device. Where the check amount is missing in check data associated with the check image or was poorly read by an optical character recognition process, the method includes inferring of the check amount based at least in part on one or more check amounts of check data associated with other check images of the account group. Once inferred, the method includes updating the check amount of the check data associated with the respective check image with the inferred check amount of the check image. Some embodiments also or alternatively include inferring a check date.