Image Processing Region Separation via Edge Strength Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing methods fail to accurately separate regions in images with diverse contents, often erroneously dividing a single region into multiple regions.

Innovation Solution

An image processing device that uses edge strength analysis to separate images into regions, employing a two-stage separation process where the first stage identifies uniform and non-uniform blocks, and the second stage refines the separation using threshold values based on surrounding region colors, followed by consolidation of sub-regions to improve precision.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a single threshold value is used to binarize the entire image, then the processing is simple and fast, but regions with different characteristics cannot be separated properly and a single region may be erroneously divided into multiple regions

Engineering Contradiction:
Improveprocessing speedVSAvoidregion separation precision
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The image processing is divided into multiple stages: first binarization using a first threshold value to obtain initial binary image data, then second binarization using a second threshold value to obtain final binary image data. This multi-stage segmentation approach allows different threshold values to be applied to different processing stages, improving region separation precision while maintaining processing efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The first binarization process is performed as a preliminary action to obtain initial binary image data that identifies candidate regions. This preliminary processing creates a foundation for the subsequent second binarization process, allowing the system to efficiently narrow down the processing scope before applying more precise thresholding.

Inventive Principle:
Principle #10Preliminary action

2Device complexity

If the entire image is binarized using a single threshold value determined from the whole image, then the process is simple, but it fails to account for local variations in different regions

Engineering Contradiction:
Improveprocess complexityVSAvoidlocal region separation precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies different threshold values (first threshold value and second threshold value) to different processing stages and regions. The first threshold value is used for initial binarization of the entire image, while the second threshold value is used for refined binarization of specific regions. This local quality approach ensures that each region is processed with appropriate thresholding parameters, improving local region separation precision without excessive complexity.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2645332B1Image processing device that separates image into plural regions
Publication Date: 2019.10.02 BROTHER KOGYO KK
  • EP2645332B1 patent drawingFigure 1
  • EP2645332B1 patent drawingFigure 2
  • EP2645332B1 patent drawingFigure 3

AI summary

An image processing device includes: a first separation unit; a second separation unit; and a generation unit. The first separation unit is configured to separate a target image represented by target image data into a plurality of regions that include a first region and a second region different from the first region. The second separation unit is configured to separate the first region into a plurality of sub-regions and to separate the second region into a plurality of sub-regions. The generation unit is configured to generate a consolidated region by consolidating at least two sub-regions among the plurality of sub-regions separated from the first and second regions.