Image Processing Apparatus Quantization Threshold Adjustment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing methods using peripheral vision and involuntary eye movement during fixation are prone to excessive sensitivity, leading to the unintended detection of minute noise, which decreases inspection efficiency by requiring unnecessary reconfirmation of noise pop-ups.

Innovation Solution

An image processing apparatus that adjusts the quantization threshold value based on the division size to stabilize singular portion extraction, preventing the detection of smaller-than-necessary defects or noise, while maintaining sensitivity by using a specific relationship between division size and quantization threshold value.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the division region is enlarged or the quantization threshold value is reduced to increase sensitivity, then the singular portion becomes more conspicuous in the image, but minute noise is also popped up unintendedly

Engineering Contradiction:
Improvesingular portion detection sensitivityVSAvoidnoise detection
Core Design Contradiction:
Measurement precisionVSObject-generated harmful factors

Solution Approach 1:

The patent applies parameter changes by dynamically adjusting the quantization threshold value based on the division region size. Specifically, when the division region is enlarged, the quantization threshold value is increased proportionally, maintaining an appropriate sensitivity level that prevents noise detection while still making singular portions conspicuous. This coordinated parameter adjustment resolves the contradiction between detection sensitivity and noise suppression.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If the quantization threshold value is reduced to detect smaller defects, then detection sensitivity increases, but unnecessary reconfirmation of noise is required

Engineering Contradiction:
Improvedefect detection sensitivityVSAvoidinspection efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent implements parameter changes by establishing a specific relationship between quantization threshold value and division region size. The quantization threshold value is set to increase in proportion to the division region size, which maintains optimal detection sensitivity while preventing the pop-up of minute noise. This eliminates unnecessary reconfirmations and preserves inspection efficiency.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If the division region is enlarged to make singular portions more conspicuous, then the singular portion extraction sensitivity increases, but the risk of detecting minute noise increases

Engineering Contradiction:
Improvesingular portion extraction sensitivityVSAvoidnoise detection accuracy
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies parameter changes by coordinating the adjustment of two parameters: division region size and quantization threshold value. When the division region is enlarged to improve singular portion conspicuousness, the quantization threshold value is simultaneously increased to maintain reliable noise filtering. This coordinated parameter adjustment preserves both extraction sensitivity and detection reliability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3239928B1Image processing apparatus, image processing method, and program
Publication Date: 2019.06.12 CANON KK
  • EP3239928B1 patent drawingFigure 1A~1D
  • EP3239928B1 patent drawingFigure 2
  • EP3239928B1 patent drawingFigure 3

AI summary

An image processing apparatus includes a setting unit, an averaging unit, a determination unit, a quantization unit, an addition unit and a detection unit. The averaging unit divides image data based on a division size and phase set by the setting unit to subject pixels included in the resultant respective division regions to an averaging processing to calculate an average value. The quantization unit configured to obtain a quantization value for each of the plurality of pixels. The addition unit adds the quantization values, that are obtained so that at least one of the division size and the phase is different from the other, to generate addition image data. The determination unit determines the quantization threshold value so that the quantization threshold value in a case where the division size is a first size is higher than the quantization threshold value in a case where the division size is a second size larger than the first size.