Vectorized Drawing Instruction Symbol Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing OCR technologies face challenges in recognizing texts from electronic documents that use vectorized drawing instructions, especially when custom character codes or missing codes are present, leading to inefficiencies and inaccuracies.
Innovation Solution
The implementation of methods and systems that utilize vectorized drawing instructions to identify symbols directly, employing neural network models to process these instructions and generate probabilities for symbol identification, thereby bypassing the reliance on character codes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If OCR technology uses traditional character code-based recognition, then the process is simple and fast, but it fails when character codes are absent or incorrect
Solution Approach 1:
The patent introduces vectorized drawing instructions as an intermediary between the symbol and its character code. Instead of directly relying on character codes, the system uses VDI data (mathematical descriptions of symbol geometry) as a mediator to identify symbols when character codes are missing or incorrect. This intermediary layer enables reliable recognition without being blocked by code failures.
Solution Approach 2:
The patent replaces the mechanical character code-based recognition system with a neural network-based pattern recognition system that processes vectorized drawing instructions. This substitution allows the system to recognize symbols based on their geometric characteristics rather than relying on predefined character codes, significantly improving reliability for difficult symbol sets.
2Reliability
If the system processes vectorized drawing instructions using neural networks, then recognition accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The patent applies partial action by using neural networks only when necessary - specifically when character code-based recognition fails or when processing difficult symbol sets. For standard symbols with valid character codes, the system uses the simpler and faster traditional method. This selective application of complex processing minimizes overall processing time while maintaining high accuracy for problematic cases.
3Productivity
If the system relies on character codes for symbol identification, then the process is efficient, but it cannot handle custom or missing character codes
Solution Approach 1:
The patent creates a universal symbol recognition system that can handle multiple types of input scenarios through a single integrated approach. The system can process both standard symbols with character codes and custom symbols without proper codes by using vectorized drawing instructions as a universal descriptor. This multi-functionality allows the same system to efficiently handle diverse symbol types and encoding situations.
Data Source
AI summary
Aspects and implementations provide for techniques of fast and efficient recognition of texts in electronic documents. The disclosed techniques include, for example, accessing a description of a symbol in a page description file for a document and identifying, responsive to a character code failure, the symbol using a vectorized drawing instruction for the symbol. The character code failure includes an absence of a character code in the description of the symbol or a bad character code in the symbol description of the symbol. The techniques further include identifying a text of the document using the identified symbol.