Character Recognition Using Symmetric Blank Areas

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current image processing techniques face challenges in accurately and efficiently identifying candidate regions for character recognition in image data, particularly due to poor image quality and unfavorable conditions such as dark areas proximate to graphical character representations, leading to misrecognition and resource wastage.

Innovation Solution

The method involves detecting symmetrically-located blank areas in the image data to identify candidate regions, where graphical character representations are likely to be sandwiched, and using these areas to isolate and recognize the characters, thereby improving detection accuracy and efficiency by avoiding pitfalls of existing techniques like MSER.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional OCR techniques search for patterns and features in image data, then character recognition can be performed, but detection accuracy deteriorates due to poor image quality and dark areas proximate to characters

Engineering Contradiction:

Improvecharacter detection accuracyVSAvoidimage quality degradation

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent introduces blank pixel detection as an intermediary approach to identify character regions. Instead of directly analyzing the character pixels which are affected by poor image quality and dark areas, the system detects the blank pixels surrounding the characters to infer character locations, thereby avoiding the harmful effects of image degradation on detection accuracy

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the traditional mechanical pattern-matching approach with a computational approach that analyzes pixel intensity distributions and blank areas. This substitution allows the system to overcome the limitations of conventional OCR methods in handling poor image quality by using a different detection mechanism altogether

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If the entire image data is processed for character recognition, then complete coverage is achieved, but computational overhead increases due to processing unnecessary regions

Engineering Contradiction:

Improverecognition efficiencyVSAvoidcomputational overhead

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent segments the image data into regions based on blank pixel detection. By identifying and isolating regions containing blank pixels that are characteristic of character areas, the system can process only these relevant segments rather than the entire image, significantly reducing computational overhead while maintaining complete coverage

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality analysis by detecting blank pixels and their spatial distributions to identify character regions. This localized approach allows the system to focus computational resources only on areas with specific blank pixel patterns, improving efficiency by avoiding processing of uniform or non-character regions

Inventive Principle:

Principle #3Local quality

3Ease of operation

If MSER (maximally stable extremal region) technique is used to identify character regions, then candidate region detection can be performed, but misrecognition occurs due to dark areas proximate to characters

Engineering Contradiction:

Improvecandidate region identificationVSAvoidrecognition accuracy

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent inverts the traditional approach by not detecting characters directly but rather detecting the blank spaces around them. This inversion allows the system to identify character regions through the presence of blank pixels, which are not affected by the same lighting conditions that cause misrecognition in direct character detection methods

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent converts the harmful effect of dark areas into a beneficial detection mechanism. By detecting blank pixels (which are not affected by darkness) rather than character pixels, the system uses the presence of bright blank areas to identify character regions, turning the lighting limitation into a strength for reliable detection

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentUS11074472B2Methods and apparatus for detecting and recognizing graphical character representations in image data using symmetrically-located blank areas

Publication Date: 2021.07.27 SYMBOL TECHNOLOGIES LLC

AI summary

Methods, systems and apparatus for detecting and recognizing graphical character representations in image data using symmetrically-located blank areas are disclosed herein. An example disclosed method includes detecting blank areas in image data; identifying, using the processor, a symmetrically-located pair of the blank areas; and designating an area of the image data between the symmetrically-located pair of the blank areas as a candidate region for an image processing function.