Biometric Skin Color Threshold Adaptation for Lighting Noise

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Biometric determination techniques face accuracy drops due to changes in lighting, illuminance, and background color, especially when camera automatic correction is activated, leading to noise interference beyond skin color changes due to blood flow.

Innovation Solution

An image processing apparatus that determines whether a person in an image is a living body by analyzing skin color thresholds over a time-series image, updating these thresholds based on changes in skin color pixels and average hue, and adjusting biometric determination processing accordingly to maintain accuracy despite changes in imaging conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If camera automatic correction function is activated to adapt to lighting changes, then adaptability to imaging conditions is improved, but measurement precision of skin color changes due to blood flow deteriorates

Engineering Contradiction:
Improveadaptability to lighting changesVSAvoidprecision of skin color change detection
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent extracts and removes the automatic correction processing from the image processing pipeline during biometric determination. By stopping the automatic correction function, the system eliminates the source of noise that interferes with skin color analysis, allowing pure blood flow-induced color changes to be detected without camera processing interference.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent dynamically adjusts the state of the automatic correction function based on the operational phase. During biometric determination, the correction function is stopped to ensure measurement accuracy, while it can be active during other phases. This dynamic control allows the system to optimize for precision when needed while maintaining adaptability when not performing biometric analysis.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If automatic correction is stopped during biometric determination, then measurement precision is improved, but adaptability to lighting changes deteriorates

Engineering Contradiction:
Improveprecision of biometric determinationVSAvoidresponse to lighting condition changes
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent performs preliminary actions to compensate for lighting changes before biometric determination begins. By detecting skin color thresholds in advance and establishing baseline values under current lighting conditions, the system prepares reference data that allows accurate blood flow detection even without continuous automatic correction during the measurement process.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If skin color threshold is determined based on initial image, then processing speed is improved, but reliability under changing conditions deteriorates

Engineering Contradiction:
Improveprocessing speedVSAvoidreliability of biometric determination
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements periodic updates of skin color thresholds during the biometric determination process. Rather than using a single initial threshold, the system periodically recalibrates threshold values based on ongoing analysis, ensuring that the determination remains reliable even as lighting or imaging conditions change throughout the measurement period.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentUS20230290186A1Image processing apparatus, image processing method, and storage medium
Publication Date: 2023.09.14 CANON KK
  • US20230290186A1 patent drawing
  • US20230290186A1 patent drawing
  • US20230290186A1 patent drawing

AI summary

An image processing apparatus is configured to determine whether a person in an image is a living body. The image processing apparatus includes a processor, and a memory storing executable instructions which, when executed by the processor, cause the image processing apparatus to perform operations including determining, based on a time-series image acquired during a predetermined time period, a threshold indicating a skin color range, obtaining color information describing a skin color area in the time-series image based on the threshold, and controlling, based on the color information, whether to continue biometric determination processing for determining whether a person in the time-series image is a living body.