Text Recognition Model Using Multi-Dimensional Feature Splicing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional text recognition methods face challenges with low accuracy due to limited annotation data, non-standard user expressions, and the diversity of text inputs, which complicates the classification process in human-computer interaction systems.
Innovation Solution
A text recognition method that employs primary and secondary classification stages, where primary classification extracts features from different dimensions and secondary classification uses a spliced feature to improve accuracy, utilizing a meta-classifier with local parameter spaces and a statistical machine learning model for enhanced classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional text classification methods are used, then the classification process is simple, but the text recognition accuracy is low
Solution Approach 1:
The classification process is divided into two distinct stages: primary classification that extracts features from different dimensions (semantic, syntactic, contextual), and secondary classification that performs final categorization. This segmentation allows each stage to focus on specific tasks, improving overall accuracy while maintaining manageable complexity through modular design
Solution Approach 2:
The patent introduces multi-dimensional feature extraction in the primary classification stage, analyzing text from semantic, syntactic, and contextual dimensions simultaneously. This dimensional expansion enriches the feature space without proportionally increasing system complexity, as the framework systematically processes each dimension
2Adaptability or versatility
If single-stage classification is used, then the system is simple to implement, but it cannot handle diverse text expressions effectively
Solution Approach 1:
The classification system is segmented into primary classification for multi-dimensional feature extraction and secondary classification for final decision-making. This segmentation enables the system to handle diverse text expressions by processing them through specialized stages, with each stage optimized for its specific function
Solution Approach 2:
The primary classification stage performs preliminary action by extracting and organizing features from different dimensions before the secondary classification stage makes the final categorization decision. This preliminary processing prepares the data in a structured manner that enhances the system's ability to handle diverse expressions
3Measurement precision
If limited annotation data is available, then data collection is efficient, but classification accuracy deteriorates
Solution Approach 1:
The patent compensates for limited annotation data by introducing multi-dimensional feature extraction, analyzing text from semantic, syntactic, and contextual dimensions. This dimensional enrichment creates more informative features from the available data, improving classification accuracy without requiring proportional increases in data volume
Solution Approach 2:
The system changes parameters by extracting multiple types of features (semantic features, syntactic features, contextual features) from the same text data. This parameter transformation allows the system to derive more information from limited annotation data, effectively increasing the informational content without increasing data quantity
Data Source
AI summary
Provided in the present disclosure are a text recognition method, and a model and an electronic device, which are applied to a mode in which primary classification is first performed from different dimensions, and secondary classification is then performed, such that the meaning of text is analyzed from different dimensions, thereby improving the accuracy of text recognition. The method includes: acquiring text to be recognized, and performing primary classification on the text to obtain a plurality of text features, wherein the primary classification is used for performing feature extraction on the text from different dimensions, and there are differences between features extracted from the different dimensions (100); splicing the plurality of text features, so as to obtain spliced features (101); and performing secondary classification on the spliced features to obtain a text category corresponding to the text, wherein the secondary classification is used for classifying the spliced features (102).