Dot Text Recognition via Morphological Transformation
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.
Innovation Solution
A method that extracts sets of dots from an image based on selection criteria, transforms them into stroked characters using candidate distances along different 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 on dot text images, then recognition accuracy may improve, but processing time and computational complexity increase significantly
Solution Approach 1:
The patent applies morphological operations (dilation and erosion) as preliminary actions to close gaps between dots and connect broken character parts before recognition. This pre-processing step transforms dot-based characters into continuous stroke forms, making them recognizable without requiring extensive case-specific processing for each dot pitch variation.
Solution Approach 2:
The patent changes the structural parameter of the dot text by applying morphological transformations that convert discrete dots into connected regions. By adjusting the morphology parameters (structuring element size and shape), the system adapts to different dot pitches and character deformations, achieving accurate recognition across varying conditions without time-consuming manual preprocessing.
2Device complexity
If character or string segmentation is performed on dot text, then recognition may be simplified, but segmentation accuracy deteriorates due to non-uniform dot spacing, deformed, skewed and touching characters
Solution Approach 1:
Instead of segmenting dot text into individual characters or strings (the conventional approach), the patent inverts the strategy by applying morphological operations that connect dots within characters first. This transformation converts the segmented dot pattern into unified character forms, eliminating the segmentation problem entirely while handling non-uniform spacing, deformation, and touching characters effectively.
Solution Approach 2:
The patent extracts the essential structural information by applying morphological operations that preserve character identity while removing the problematic dot-based representation. By extracting and emphasizing the continuous stroke patterns through dilation and erosion, the system bypasses the need for accurate segmentation of deformed and touching characters.
3Measurement precision
If OCR methods are sensitive to dot pitch variations, then recognition accuracy improves for uniform dot text, but robustness deteriorates when dealing with non-uniform dot spacing and deformed characters
Solution Approach 1:
The patent introduces dynamic adaptability by applying morphological operations with adjustable structuring elements that can handle varying dot pitches. The dilation and erosion operations dynamically adapt to different spacing patterns, character deformations, and skew angles, making the recognition system robust to dot pitch variations while maintaining accuracy for uniform dot text.
Solution Approach 2:
The patent creates a universal pre-processing approach using morphological operations that serves multiple functions: closing gaps between dots, connecting broken character parts, removing noise, and handling touching characters. This multi-functional processing enhances robustness across diverse dot text conditions without sacrificing recognition accuracy for uniform cases.
Data Source
AI summary
A method and apparatus for extraction of dots in an image are described. An image is binarized according to an initial intensity threshold to obtain an initial binary image including foreground and background pixels. Each foreground pixel has a foreground intensity value and each background pixel has a background intensity value. A set of blobs including foreground pixels is selected from the initial binary image to be part of a selected set of dots, where each blob from the set of blobs has characteristics of a dot. Responsive to determining that a successive binarization is to be performed, the following operations are repeated: (1) binarization of the image according to a successive intensity threshold (2) selection of a successive set of blobs, where each blob has characteristics of a dot. Responsive to determining that a successive binarization is not to be performed, the selected set of dots is output.