Defective Pixel Detection Using Interquartile Range Filtering

Solution Overview

Problem

Existing image processing methods struggle to accurately detect defective pixels, particularly in cases where noise and edge or texture presence lead to incorrect detection, and they require significant computational resources.

Innovation Solution

An image processing device and method that performs order-statistic filter processing on pixel values, calculates a difference value and statistical variation, and determines defective pixels by evaluating the difference value against a threshold relative to the statistical variation, using interquartile range as a statistical measure to reduce computational load and suppress incorrect detections.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If standard deviation of pixel values is used to calculate degree of prominence for defective pixel detection, then defective pixels can be identified, but computational complexity increases and incorrect detections occur in noisy or textured regions

Engineering Contradiction:

Improvedefective pixel detection accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the statistical parameter from standard deviation to interquartile range (IQR). The IQR is calculated as the difference between the 75th percentile and 25th percentile values of pixel intensities in a local region. This parameter change provides robustness against outliers (defective pixels) and reduces sensitivity to noise and texture, while maintaining computational efficiency through sorted pixel value analysis.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent employs a simplified detection metric based on IQR that requires less computational resources compared to standard deviation methods. By using percentile-based statistics rather than squared differences, the computation becomes more efficient and suitable for real-time processing, effectively using a 'simpler' statistical approach that achieves the goal with reduced complexity.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

2Measurement precision

If standard deviation-based prominence calculation is used, then defective pixels can be detected, but noise and edge/texture presence cause incorrect detections

Engineering Contradiction:

Improvedefective pixel detection accuracyVSAvoidnoise and texture interference

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent changes the statistical parameter from standard deviation to interquartile range (IQR). The IQR is calculated as the difference between the 75th percentile and 25th percentile values of pixel intensities in a local region. This parameter change provides robustness against outliers (defective pixels) and reduces sensitivity to noise and texture, while maintaining computational efficiency through sorted pixel value analysis.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If standard deviation calculation is performed on pixel values, then defective pixels can be identified, but computational power consumption increases

Engineering Contradiction:

Improvedefective pixel detection accuracyVSAvoidcomputational power consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent employs a simplified detection metric based on IQR that requires less computational resources compared to standard deviation methods. By using percentile-based statistics rather than squared differences, the computation becomes more efficient and suitable for real-time processing, effectively using a 'simpler' statistical approach that achieves the goal with reduced complexity.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Data Source

PatentUS10728476B2Image processing device, image processing method, and image processing program for determining a defective pixel

Publication Date: 2020.07.28 OLYMPUS CORPORATION(JP)

AI summary

An image processing device includes a computer that is configured to: perform order-statistic filter processing on pixel values of individual pixels in a block including a pixel of interest and peripheral pixels in an acquired image; calculate a difference value between the pixel values of the pixel of interest before and after the order-statistic filter processing is performed; calculate a statistical variation of the pixel values of the peripheral pixels subjected to the order-statistic filter processing; and determine the pixel of interest as being a defective pixel in the case where a value obtained by dividing the difference value by the statistical variation is greater than or equal to a predetermined threshold.