Dynamic Threshold Character Segmentation for OCR Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Poor word segmentation in optical character recognition (OCR) processes can significantly lower recognition accuracy, as incorrect segmentation cannot be overcome by excellent character recognition, especially when different font families with varying character spacing are involved.
Innovation Solution
A character segmentation system that analyzes character distances and generates dynamic thresholds to determine whether to merge or split characters into words, using computer vision techniques to tailor the recognition process to the specific font family, thereby improving word segmentation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fixed threshold methods are used for character segmentation, then the processing speed is fast, but the segmentation accuracy deteriorates when different font families with varying character spacing are involved
Solution Approach 1:
The patent implements dynamic threshold adjustment by analyzing the actual spacing between consecutive characters in the document. Instead of using a fixed threshold, the system calculates spacing metrics from the document image and adapts the segmentation threshold accordingly. This allows the segmentation process to automatically adjust to different font families and spacing styles, resolving the contradiction between maintaining simple processing and achieving high accuracy across varied document formats.
Solution Approach 2:
The system changes the segmentation threshold parameter dynamically based on document-specific characteristics. By computing actual character spacing from the image and using this information to adjust the threshold, the system transforms a static parameter into a dynamic one that adapts to different font families and spacing conventions, thereby improving segmentation accuracy without requiring manual intervention.
2Measurement precision
If dynamic threshold calculation is performed to improve segmentation accuracy, then the recognition accuracy improves, but the processing time increases
Solution Approach 1:
The system performs preliminary analysis of character spacing and font characteristics before the main segmentation process. By pre-calculating spacing metrics and determining appropriate thresholds in advance, the system avoids repeated calculations during segmentation, thereby reducing overall processing time while maintaining high accuracy.
Solution Approach 2:
The system applies dynamic threshold adjustment selectively based on document characteristics. Rather than performing exhaustive analysis on all documents, the system identifies when dynamic adjustment is necessary (e.g., when multiple font families are detected) and applies the computationally intensive process only in those cases, balancing accuracy improvement with processing efficiency.
Data Source
AI summary
A method clusters each character on a document into one of a plurality of clusters based on widths of at least a portion of the characters on the document and measures distances between characters on the document. A threshold for each of the plurality of clusters is calculated based on at least a portion of the distances between characters in each cluster. The method then segments characters into units using the thresholds for the plurality of clusters. A distance between two characters in the document is compared to a threshold for a cluster to classify the two characters as being part of a unit when the distance is less than the threshold and not being part of the unit when the distance is greater than the threshold. Then, the method performs a recognition process on the document using the units.