Iterative Image Character Recognition Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional OCR systems face challenges in accurately retrieving data from images, particularly when image quality is poor, leading to misinterpretation of characters and inefficiencies in manual correction processes.
Innovation Solution
A system that segments images into sections, performs OCR, queries a linguistic database to identify errors, calculates a statistical score, and iteratively modifies image characteristics until the score meets a threshold, selecting optimal sections with minimal errors to generate a processed image with reduced errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional OCR systems are used to retrieve data from images, then text recognition can be performed, but accuracy deteriorates when image quality is poor leading to character misinterpretation
Solution Approach 1:
The patent segments the image into multiple candidate regions (first candidate image, second candidate image, etc.) that all correspond to the same character position. Each candidate is processed independently through OCR, and results are compared to identify the most accurate character recognition, thereby improving reliability under poor image quality conditions.
Solution Approach 2:
The system implements a feedback mechanism where OCR results from multiple candidates are evaluated, compared, and used to select the optimal character recognition. The process includes calculating confidence scores, comparing character patterns, and iteratively refining the selection until the most accurate result is identified, thus improving measurement precision.
2Measurement precision
If manual correction processes are used to fix OCR errors, then accuracy can be improved, but productivity deteriorates due to time-consuming manual intervention
Solution Approach 1:
The system performs self-correction by automatically comparing OCR results from multiple candidate images, evaluating confidence scores, and selecting the most accurate character recognition without requiring manual intervention. This self-service approach maintains high accuracy while preserving processing efficiency.
Solution Approach 2:
The patent performs preliminary OCR processing on multiple candidate images simultaneously before final selection. By pre-processing all candidates and preparing their recognition results in advance, the system enables rapid automatic selection and correction, eliminating the need for time-consuming manual correction while maintaining high accuracy.
3Measurement precision
If multiple candidate images are processed through iterative OCR and comparison, then data retrieval accuracy is improved, but device complexity increases
Solution Approach 1:
The system uses a universal comparison mechanism that can handle multiple candidate images through the same OCR and evaluation process. The same pattern matching algorithm and confidence score calculation are applied universally to all candidates, simplifying the overall system architecture while maintaining the ability to process multiple images and improve accuracy.
4Reliability
If iterative image processing with linguistic database queries is performed, then error correction capability is improved, but use of energy increases
Solution Approach 1:
The system performs linguistic database queries and iterative processing only when necessary - specifically when confidence scores indicate potential errors or when comparing multiple candidates. By applying these energy-intensive operations selectively rather than universally, the system improves error correction capability while controlling energy consumption through partial action.
Data Source
AI summary
Various embodiments illustrated herein disclose a method that includes receiving a plurality of images from an image capturing unit. Thereafter, an image evaluation process is executed on each of plurality of sections in each of the plurality of images. The image evaluation process includes performing optical character recognition (OCR) on each of the plurality of sections in each of the plurality of images to generate text corresponding to the plurality of respective sections. Further, the image evaluation process includes querying a linguistic database to identify one or more errors in the generated text. Further, the method includes modifying one or more image characteristics of each of the plurality of images and repeating the execution of the image evaluation process on the modified plurality of images until at least the calculated statistical score is less than a pre-defined statistical score threshold.


