Character Recognition Device Using Segmented Finite Automata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current character recognition systems face challenges in accurately identifying character strings from input images while efficiently excluding prohibited strings, as they often rely on finite automata models that are not optimized for filtering out unwanted characters.
Innovation Solution
A recognition device is designed with a candidate detection unit, recognition unit, matching unit, and prohibition processing unit, utilizing deterministic finite automata to model recognition and prohibition target character strings, which allows for efficient matching and filtering of character candidates based on knowledge and prohibition dictionaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If finite automata models are used for character recognition, then matching capability is improved, but filtering efficiency of prohibited strings deteriorates
Solution Approach 1:
The patent segments the character recognition system into two distinct finite automata: one dedicated to recognition (matching valid characters) and another dedicated to prohibition (filtering invalid characters). This segmentation allows each automaton to be optimized for its specific function, resolving the contradiction between recognition accuracy and filtering efficiency.
Solution Approach 2:
The patent introduces a prohibition dictionary as an intermediary component that contains prohibited character strings. This intermediary enables the system to efficiently filter unwanted characters by comparing recognition results against the prohibition dictionary using a dedicated finite automaton, without compromising the recognition capability.
2Measurement precision
If knowledge dictionary matching is performed, then recognition accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary action by building and storing finite automata models for both recognition and prohibition dictionaries in advance. During actual character recognition, the system simply executes the pre-built automata rather than performing complex dictionary matching, significantly reducing processing time while maintaining high accuracy.
Solution Approach 2:
The patent replaces the traditional mechanical dictionary matching process with finite automata-based pattern recognition. This substitution transforms the time-consuming string matching operation into efficient state transitions within the automata, dramatically improving processing speed while preserving recognition accuracy.
Data Source
AI summary
According to an embodiment, a recognition device includes a candidate detection unit, a recognition unit, a matching unit, and a prohibition processing unit. The candidate detection unit detects, from an input image, character candidates each being a set of pixels estimated to include a character. The recognition unit recognizes each of the character candidates and generates one or more recognition candidates each being a character of a candidate as a recognition result. The matching unit matches each of the one or more recognition candidates with a knowledge dictionary in which a recognition target character string is modeled, and generates matching results obtained by matching a character string estimated to be included in the input image with the knowledge dictionary. The prohibition processing unit deletes, from the matching results, a matching result obtained by matching a character string including a prohibition target character string with the knowledge dictionary.