UDC 004.932
MODIFICATION OF SGM STEREO IMAGE MATCHING ALGORITHM BASED ON ADAPTIVE PENALTY FUNCTION
I. A. Zinin, post-graduate student, RSREU, Ryazan, Russia;
orcid.org/0009-0007-6166-677X, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The article presents the implementation of semi-global matching (SGM) algorithm, improved with adap tive regularization strategies to enhance disparity estimation in stereo vision. Traditional SGM relies on fixed penalty parameters for smoothness constraints, which can lead to inaccuracies in complex scenes with different textures, occlusions, and depth gaps. The proposed approach introduces adaptive regularization functions that dynamically adjust penalty coefficients based on local image characteristics such as intensity gradients and disparity confidence measures.
Key words: : disparity, dynamic penalties, semi-global matching, gradient, connectivity, smoothness, depth map, regularization.