Image Processing Apparatus Geometric Distortion Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Image processing technologies face challenges in effectively monochromatizing foreground portions with similar colors in captured images, particularly in portable terminals, where environmental conditions and lighting variations lead to geometric distortion and inclusion of unwanted objects, making it difficult to maintain the originality and visibility of documents.

Innovation Solution

An image processing apparatus and method that identifies and corrects geometric distortion in captured images, and then performs shade removal and foreground extraction to unify similar colors into a single color, allowing for optional monochromatization of text portions, enhancing visibility and maintaining document originality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If shade component removing processing is performed on captured images, then visibility of foreground portions is improved, but minute color differences in foreground portions are lost

Engineering Contradiction:
ImprovevisibilityVSAvoidcolor differences
Core Design Contradiction:
Illumination intensityVSLoss of information

Solution Approach 1:

The image processing is divided into distinct segments: first shade component removal is performed, then foreground portion extraction is applied selectively. This segmentation allows the system to remove shades from the background while preserving and enhancing the color information of foreground portions through subsequent extraction processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different processing qualities are applied to different regions of the image. The shade component removing processing is applied to the background area to improve visibility, while the foreground portions undergo extraction processing to preserve their color characteristics. This local quality approach ensures that each region receives the appropriate treatment to maintain both visibility and color accuracy.

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If geometric distortion correction is applied to captured images, then document shape accuracy is improved, but processing time increases

Engineering Contradiction:
Improveshape accuracyVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

Geometric distortion correction is performed as a preliminary step before other image processing operations. By correcting the geometric distortion early in the processing pipeline, subsequent operations such as shade removal and foreground extraction can be performed more efficiently on the already-corrected image, reducing overall processing time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system performs geometric distortion correction quickly by using efficient algorithms that prioritize speed while maintaining adequate accuracy for the application. The correction is rushed through as a necessary preliminary step to enable faster subsequent processing of the document image.

Inventive Principle:
Principle #21Skipping (Rushing through)

Data Source

PatentEP3407589B1Image processing apparatus, image processing method, and storage medium
Publication Date: 2023.02.15 CANON KK
  • EP3407589B1 patent drawingFigure 1A~1B
  • EP3407589B1 patent drawingFigure 2
  • EP3407589B1 patent drawingFigure 3

AI summary

An image processing apparatus includes an acquisition unit configured to acquire an image, an image processing unit configured to perform image processing on the acquired image, and an output unit configured to output the image having undergone the image processing to an output destination. The image processing unit is able to perform replacement processing for replacing colors of predetermined image portions in the acquired image with a color selected from a plurality of predetermined different colors based on the colors of the predetermined image portions. The output unit is able to output the acquired image having undergone the replacement processing to the output destination and to output the image not having undergone the replacement processing to the output destination.