Facial Verification via Multi-Color Channel Shading Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial verification technologies face challenges in accurately verifying users in varying illumination environments, leading to decreased accuracy due to shading and illumination variations.
Innovation Solution
The method involves separating a query face image into color channel images, generating a multi-color channel target face image with reduced shading by combining smoothed and gradient images, and using a weight parameter to fuse these images, which is determined based on image quality assessment or learned through neural networks, to enhance edge detection and minimize illumination effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional facial verification is used in varying illumination environments, then the verification process is simple, but the accuracy decreases due to shading and illumination variations
Solution Approach 1:
The query face image is separated into multiple color channel images (e.g., red, green, blue channels). This segmentation allows independent processing of each color channel to reduce the impact of illumination variations, as different color channels respond differently to varying light conditions, thereby improving verification accuracy
Solution Approach 2:
The patent transforms the original color channel images into target face images by applying image processing operations (smoothing and gradient calculations) and combining them with weight parameters. This parameter transformation reduces shading effects and illumination variations while preserving essential facial features for accurate verification
2Adaptability or versatility
If color channel separation and image fusion are applied to reduce illumination effects, then the robustness to illumination variations improves, but the processing complexity increases
Solution Approach 1:
By separating the image into color channels and processing each channel independently with standardized operations (smoothing, gradient calculation), the patent achieves adaptability to illumination variations while maintaining manageable processing complexity through modular, repeatable steps
Solution Approach 2:
The use of weight parameters in fusing color channel images provides a flexible mechanism to adjust the contribution of each channel based on illumination conditions. This parameter-based approach enables adaptability without requiring complex decision logic, balancing robustness and processing complexity
Data Source
AI summary
A facial verification method includes separating a query face image into color channel images of different color channels, obtaining a multi-color channel target face image with a reduced shading of the query face image based on a smoothed image and a gradient image of each of the color channel images, extracting a face feature from the multi-color channel target face image, and determining whether face verification is successful based on the extracted face feature.