Pattern-Based Optical Character Recognition for Bulk Text Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing Optical Character Recognition (OCR) technologies lack flexibility and ease of use, requiring users to manually select and copy multiple text items from images one at a time, which is laborious and prone to errors, especially when dealing with large datasets like email addresses.
Innovation Solution
Implementing an intelligent text recognition system with a pattern detection mode that allows users to select a pattern, train a model, and automatically identify and list occurrences of that pattern within an image, enabling bulk copying or selective pasting of relevant text.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual selection and copying of text items is used, then text extraction is possible, but user efficiency and productivity deteriorate when dealing with multiple text items
Solution Approach 1:
The system performs automatic text extraction by detecting patterns themselves without requiring manual user intervention for each text item. The OCR engine automatically identifies and extracts text matching the detected pattern, making the system serve itself rather than requiring continuous user guidance for each extraction task.
Solution Approach 2:
The patent replaces the mechanical manual process of selecting and copying text with an automated computational system. The pattern detection engine and OCR engine work together to automatically identify, select, and extract relevant text items, substituting the manual mechanical interaction with an intelligent automated process.
2Adaptability or versatility
If traditional OCR is used to recognize all text, then text identification is achieved, but user flexibility and ease of use worsen when needing to select specific patterns
Solution Approach 1:
The system performs preliminary pattern detection and identification before the actual text extraction process. By first detecting the pattern type (email, phone number, URL, etc.) and then using that information to guide the OCR extraction, the system prepares in advance what needs to be extracted, making the subsequent process more efficient and user-friendly.
Solution Approach 2:
The patent segments the text extraction process into distinct phases: pattern detection, pattern classification, and targeted text extraction. This segmentation allows the system to handle different text patterns (emails, phone numbers, URLs) separately and efficiently, providing adaptability while maintaining ease of operation through automated workflow management.
3Productivity
If manual iteration for selecting multiple text portions is performed, then text copying is achieved, but time consumption and productivity worsen
Solution Approach 1:
The system maintains continuous automated operation throughout the text extraction process. Once the pattern is detected and the OCR engine is activated, the extraction continues automatically through all matching text items without interruption or manual intervention, eliminating the need for repeated start-stop manual iterations and significantly reducing time loss.
Solution Approach 2:
The patent replaces the iterative manual process of selecting, copying, and pasting text multiple times with a single automated computational process. The system detects the pattern once, then automatically extracts all matching text items in one continuous operation, substituting multiple manual iterations with a single automated workflow.
Data Source
AI summary
Disclosed are various embodiments for intelligent text recognition based upon a selected pattern detection mode. First, text can be identified in an image. A pattern detection mode can be selected by a user or autonomously. In some instances, the pattern detection mode can be selected based at least in part on a user account. Next, the text can be parsed for occurrences of a pattern associated with the selected pattern detection mode. A list of occurrences of the pattern can be generated from the text and presented to a user. In some instances, a user can train a model to learn a new pattern.