Halim, Marina (2025) Improvement of the stereo vision algorithm using an adaptive guided filter for depth map measurement. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)
![]() |
Text (Full Text)
Improvement of the stereo vision algorithm using an adaptive guided filter for depth map measurement.pdf - Submitted Version Download (3MB) |
Abstract
Stereo vision techniques play a crucial role in computer vision and photogrammetry by estimating depth information through matching corresponding points in stereo image pairs. Despite its significance in applications such as 3D reconstruction and autonomous vehicles, traditional stereo matching algorithms face challenges in textureless regions, occlusions, and varying illumination conditions, leading to inaccuracies in depth estimation. This study aims to address these challenges by developing a stereo vision system to enhance the accuracy and robustness of stereo matching, particularly in low-texture regions. The project objective includes evaluating the proposed method using benchmark datasets like the Middlebury datasets to demonstrate its effectiveness and superiority in achieving reliable stereo matching results. The methodology involves utilizing MATLAB for depth estimation of 3D images, implementing sustainable stereo vision algorithms prioritizing energy efficiency and resource optimization, and enhancing stereo vision algorithms with adaptive guided filtering to address occlusions, noise, and efficiency for robust depth estimation. The results include the development of a comprehensive stereo matching method that improves accuracy and robustness in challenging imaging conditions, validated through experimental results and discussions showcasing the effectiveness and efficiency of the proposed algorithm.
Item Type: | Final Year Project (Project Report) |
---|---|
Uncontrolled Keywords: | Stereo vision, Depth estimation, Stereo matching, 3D reconstruction, Autonomous vehicles, Textureless regions, Occlusions, Illumination variations, MATLAB, Adaptive guided filtering, Resource optimization, Middlebury datasets, Robust depth estimation |
Subjects: | T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General) |
Divisions: | Library > Final Year Project > FTKEK |
Depositing User: | Norfaradilla Idayu Ab. Ghafar |
Date Deposited: | 08 Oct 2025 02:52 |
Last Modified: | 08 Oct 2025 02:52 |
URI: | http://digitalcollection.utem.edu.my/id/eprint/36579 |
Actions (login required)
![]() |
View Item |