Selective OCR Text and Image Display via Garbage Score Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Optical Character Recognition (OCR) technology often introduces errors in text extraction from scanned documents, affecting the usability and readability of the converted text, especially when displayed on computer screens or mobile devices.
Innovation Solution
A method and system that allows users to toggle between OCR'ed text and the original image segments, with each text segment tagged with a 'garbage score' that determines whether to display the text or the corresponding image based on a threshold value, enabling users to choose between editable text and the original image for improved readability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If OCR technology is used to convert scanned documents to editable text, then text size is reduced and transmission efficiency is improved, but errors are introduced that interfere with use and readability
Solution Approach 1:
The document is divided into multiple text segments, each with its own image counterpart. The system segments the OCR'd document into discrete units that can be independently evaluated and displayed, allowing selective presentation of text or image based on quality metrics for each segment.
Solution Approach 2:
An intermediary system is introduced that sits between the OCR output and the final display. This intermediary evaluates quality metrics for each text segment and automatically selects whether to display the text or the original image, acting as a mediator that resolves the conflict between text efficiency and accuracy.
2Ease of operation
If OCR'ed text is displayed to improve readability and editability, then text can be copied and searched, but errors from the OCR process detract from the reading experience
Solution Approach 1:
The display mode becomes dynamic rather than static. The system automatically adjusts between text and image display based on quality metrics, and allows user interaction to toggle between modes. This dynamic adaptation ensures the best possible reading experience while maintaining editability when text quality is sufficient.
Solution Approach 2:
The system changes the display parameter (text vs. image) based on quality metric thresholds. When quality metrics indicate acceptable accuracy, text is displayed for ease of operation; when quality degrades below thresholds, the system switches to image display to preserve reading experience integrity.
3Quantity of substance
If all text segments are displayed to maximize transmission efficiency, then file size remains small, but error-prone segments reduce overall document quality
Solution Approach 1:
Different quality standards are applied to different segments of the document. High-quality text segments are displayed as text to maintain small file size, while low-quality segments are displayed as images to ensure document quality. This local quality approach optimizes the overall document by treating each segment according to its specific quality characteristics.
Solution Approach 2:
Instead of applying a uniform threshold to all segments, the system uses quality metrics to selectively apply image display only where necessary. This partial action approach minimizes the impact on file size while ensuring that only the segments that truly need it are displayed as images, maintaining overall document quality without excessive resource consumption.
Data Source
AI summary
Text is extracted from a source image of a publication using an Optical Character Recognition (OCR) process. A document is generated containing text segments of the extracted text. The document includes a control module that responds to user interactions with the displayed document. Responsive to a user selection of a displayed text segment, a corresponding image segment from the source image containing the text is retrieved and rendered in place of the selected text segment. The user can select again to toggle the display back to the text segment. Each text segment can be tagged with a garbage score indicating its quality. If the garbage score of a text segment exceeds a threshold value, the corresponding image segment can be automatically displayed instead.