Enhanced genetic algorithm for indoor low-illumination stereo matching energy function optimization

This study presents an enhanced genetic algorithm to generate high-quality, dense disparity maps in low-illumination indoor stereo matching scenarios. The algorithm utilizes a novel biomimetic mutation strategy, dynamically adjusted through common allelic theory, to effectively reduce the search spa...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhang Hongjin, Wei Hui
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Alexandria Engineering Journal
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1110016825001668
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study presents an enhanced genetic algorithm to generate high-quality, dense disparity maps in low-illumination indoor stereo matching scenarios. The algorithm utilizes a novel biomimetic mutation strategy, dynamically adjusted through common allelic theory, to effectively reduce the search space and improve optimization efficiency. A multi-objective genetic approach is employed, optimizing both data fidelity and smoothness to achieve balanced and accurate results. The Integration of biological evolution principles, including a hierarchical performance index system, ensures the algorithm balances local and global features during optimization. Experimental results demonstrate that the proposed algorithm consistently converges within 10 iterations, reducing average disparity errors by 33.7% compared to state-of-the-art methods and by 20% compared to a control group. Additionally, the algorithm achieves a lower computational complexity of O(H), offering significant improvements in resource efficiency compared to standard algorithms. These findings validate the algorithm’s reliability, advancement, and effectiveness for stereo matching under challenging conditions.
ISSN:1110-0168