Image Reconstruction Using Local and Global Character Thickness Thresholds

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing techniques for improving image quality prior to optical character recognition (OCR) are limited in their ability to handle low-quality images, particularly those with distortion, and fail to differentiate between bold and normal characters, requiring high-quality input and frequent updates to character dictionaries.

Innovation Solution

A system and method that preprocesses images to generate character images, determines local and global character thickness threshold values, and reconstructs characters based on these values to enhance image quality, allowing for improved OCR accuracy by mitigating distortions and differentiating between character types.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If existing image processing techniques are used to improve image quality, then some correction capabilities are provided, but they cannot differentiate between bold and normal characters and require frequent dictionary updates

Engineering Contradiction:

Improveimage qualityVSAvoidcharacter dictionary maintenance

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies parameter changes by computing character thickness values for each character in the image and using these thickness parameters to differentiate between bold and normal characters. Instead of relying on dictionary updates, the system dynamically adjusts reconstruction thresholds based on measured character thickness, allowing automatic adaptation to different fonts and styles without maintaining extensive character dictionaries.

Inventive Principle:

Principle #35Parameter changes

2Manufacturing precision

If existing image processing techniques are used, then some reconstruction is performed, but they do not differentiate between bold and normal characters

Engineering Contradiction:

Improvecharacter reconstruction accuracyVSAvoidcharacter style information

Core Design Contradiction:

Manufacturing precisionVSLoss of information

Solution Approach 1:

The patent applies local quality by computing individual thickness values for each character and applying reconstruction thresholds specific to each character's measured thickness. This allows bold characters to be reconstructed with different parameters than normal characters, preserving character style information through localized adaptation rather than uniform processing.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes the reconstruction parameter (threshold value) based on the measured thickness of each character. By dynamically adjusting the reconstruction threshold according to character thickness, the system preserves distinctions between bold and normal characters while improving overall reconstruction accuracy.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If high quality input image is provided to OCR, then accurate text recognition is achieved, but low quality distorted images cannot be processed effectively

Engineering Contradiction:

Improvetext recognition accuracyVSAvoidimage quality tolerance

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent applies preliminary action by performing image reconstruction and quality improvement before the OCR process. The system reconstructs distorted characters, corrects skew, and improves image quality in advance, enabling subsequent OCR to achieve high accuracy even from low-quality input images.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes physical parameters of the image during reconstruction, including character thickness, skew angle, and pixel density. By adjusting these parameters to optimize character appearance before OCR, the system enables accurate text recognition from distorted images that would otherwise be unsuitable for OCR.

Inventive Principle:

Principle #35Parameter changes

4Ease of operation

If manual data extraction is performed, then flexibility is maintained, but time consumption and human error increase

Engineering Contradiction:

Improveextraction flexibilityVSAvoiddata extraction time

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical data extraction with an automated computer-based image reconstruction and recognition system. The system automatically performs image preprocessing, character reconstruction, and OCR without human intervention, eliminating time consumption and human error while maintaining flexibility through configurable processing parameters.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11386687B2System and method for reconstructing an image

Publication Date: 2022.07.12 WIPRO LTD

AI summary

This disclosure relates generally to image processing, and more particularly to method and system for reconstructing an image. In one embodiment, the method includes pre-processing an input image to generate character images corresponding to characters in the input image, determining a local character thickness threshold value for each character image, determining a global character thickness threshold value for the input image based on the local character thickness threshold values for the character images, and reconstructing each character image based on the local character thickness threshold value for each character image and the global character thickness threshold value to generate reconstructed character images. The local character thickness threshold value in a character image may be based on a set of character pixel values in a pre-determined number of segments in the character image. The method further includes re-constructing the input image based on the reconstructed character images.