Shading Dot Removal in Document Image Processing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing techniques for removing shading dots from document images, such as those discussed in Japanese Patent Application Laid-Open No. 2000-293630, are inadequate as they require predefined frequency components and may fail to remove shading dots in contact with characters, especially when the contact is variable or the shape of projections is complex.

Innovation Solution

An image processing apparatus and method that includes a processor to perform binarization, shading dot removal by isolating and removing dots not in contact with characters based on their size, and dynamically determining the size of shading dots in contact with characters for accurate removal without pre-defined parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If Fourier transform is used to remove shading dots by deleting frequency components, then shading dots with predefined frequency patterns can be removed, but shading dots with variable frequency components or complex contact patterns cannot be removed

Engineering Contradiction:

Improveshading dot removal effectivenessVSAvoidadaptability to different shading dot patterns

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent applies dynamics by making the frequency component selection adaptive rather than fixed. The system dynamically determines which frequency components to delete based on the actual shading dot patterns detected in the scanned image, allowing the removal process to adapt to various shading dot patterns including those with variable frequency components and complex contact patterns that cannot be removed by predefined frequency deletion

Inventive Principle:

Principle #15Dynamics

2Reliability

If frequency components are deleted to remove shading dots, then isolated shading dots can be removed, but shading dots in contact with characters cannot be removed without affecting character recognition

Engineering Contradiction:

Improveshading dot removal effectivenessVSAvoidcharacter recognition accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by performing frequency component deletion selectively in different regions of the image. The system identifies regions containing shading dots and applies frequency component deletion only to those specific regions, while preserving frequency components in regions containing characters. This localized approach allows effective removal of shading dots in contact with characters without affecting character recognition accuracy

Inventive Principle:

Principle #3Local quality

3Reliability

If smoothing processing is applied to remove projections, then isolated shading dots can be removed, but complex projection shapes may cause character distortion

Engineering Contradiction:

Improveshading dot removal effectivenessVSAvoidcharacter shape accuracy

Core Design Contradiction:

ReliabilityVSManufacturing precision

Solution Approach 1:

The patent applies dynamics by making the smoothing processing adaptive and controlled. The system dynamically adjusts the smoothing parameters and applies smoothing selectively based on the detected projection shapes and their relationship with characters. This dynamic control allows effective removal of shading dots with complex projection shapes while preventing character distortion by adjusting the smoothing intensity and scope

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS11778122B2Apparatus, method, and storage medium for removing shading dots

Publication Date: 2023.10.03 CANON KK

AI summary

According to an exemplary embodiment of the present disclosure, an isolated dot is removed from an image including a shading dot, and a shading dot in contact with a character is removed from the isolated dot-removed image based on a size of the removed isolated dot.