Dot Text OCR via Dot Extraction and Distance Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optical character recognition (OCR) methods for dot text are limited by requiring extensive case-specific image pre-processing, being sensitive to dot pitch variations, and needing character or string segmentation, which makes them inefficient for recognizing dot text with non-uniform spacing, deformed, skewed, and rotated characters, as well as varying contrast and backgrounds.
Innovation Solution
A method that extracts sets of dots from an image based on selection criteria, transforms them into stroked characters by adjusting distances and orientations, and performs model-based character matching without prior segmentation, enabling robust recognition of dot text regardless of dot pitch variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If extensive case-specific image pre-processing is performed, then recognition accuracy may improve, but processing time and complexity increase significantly
Solution Approach 1:
The patent extracts only the essential features (dot positions and distances) needed for recognition, eliminating the need for extensive pre-processing. By focusing on the critical geometric relationships between dots rather than processing the entire image, the system achieves accurate recognition while significantly reducing processing time.
Solution Approach 2:
The patent segments the character recognition problem into two key components: detecting individual dot positions and measuring distances between adjacent dots. This segmentation allows the system to process only the relevant geometric information rather than performing comprehensive image pre-processing, thereby improving efficiency while maintaining accuracy.
2Ease of operation
If character or string segmentation is performed, then recognition can be simplified, but the process becomes more complex and time-consuming
Solution Approach 1:
The patent extracts the essential geometric relationships (dot positions and inter-dot distances) directly from the image without performing character or string segmentation. By focusing on the fundamental dot patterns that define characters, the system simplifies the recognition process while avoiding the complexity of segmentation algorithms.
Solution Approach 2:
The patent creates a universal approach that handles various character types, orientations, and spacings through a single dot-detection and distance-measurement process. This universal method eliminates the need for separate segmentation procedures for different character cases, thereby reducing process complexity while maintaining ease of operation.
3Adaptability or versatility
If traditional OCR methods are used, then standard text can be recognized, but dot text with non-uniform spacing and variations cannot be accurately recognized
Solution Approach 1:
The patent changes the approach from recognizing fixed-font characters to measuring variable distances between dots. By using distance measurements as the primary recognition parameter rather than relying on fixed character shapes, the system achieves high accuracy for dot text with non-uniform spacing, deformed, skewed, and rotated characters while expanding adaptability to various text types.
Data Source
AI summary
A method and apparatus for locating dot text in an image are described. A set of dots is extracted. A determination of whether a first region of interest (ROI) including the set of dots satisfies selection criteria is performed, where the first region of interest is oriented based on results from a principal component analysis of the set of dots. Responsive to determining that the first ROI does not satisfy the selection criteria, performing the following: removing an outlier dot from the first set of dots to obtain a second set of dots; when the second ROI satisfies the selection criteria, outputting the second ROI as a location of the dot text in the image, and when the second region of interest does not satisfy the selection criteria, repeating the operations until a resulting ROI is determined to satisfy the selection criteria.