A new function of stereo matching algorithm based on hybrid convolutional neural network

Hamid, Mohd Saad and Kadmin, Ahmad Fauzan and Abd Manap, Nurulfajar and Hamzah, Rostam Affendi and Abd Gani, Shamsul Fakhar and Herman, Adi Irwan (2022) A new function of stereo matching algorithm based on hybrid convolutional neural network. Indonesian Journal of Electrical Engineering and Computer Science, 25 (1). pp. 223-231. ISSN 2502-4752

Text 25074-51436-1-PB.PDF Download (1MB)

Abstract

This paper proposes a new hybrid method between the learning-based and handcrafted methods for a stereo matching algorithm. The main purpose of the stereo matching algorithm is to produce a disparity map. This map is essential for many applications, including three-dimensional (3D) reconstruction. The raw disparity map computed by a convolutional neural network (CNN) is still prone to errors in the low texture region. The algorithm is set to improve the matching cost computation stage with hybrid CNN-based combined with truncated directional intensity computation. The difference in truncated directional intensity value is employed to decrease radiometric errors. The proposed method’s raw matching cost went through the cost aggregation step using the bilateral filter (BF) to improve accuracy. The winner-take-all (WTA) optimization uses the aggregated cost volume to produce an initial disparity map. Finally, a series of refinement processes enhance the initial disparity map for a more accurate final disparity map. This paper verified the performance of the algorithm using the Middlebury online stereo benchmarking system. The proposed algorithm achieves the objective of generating a more accurate and smooth disparity map with different depths at low texture regions through better matching cost quality.

Item Type:	Article
Uncontrolled Keywords:	Convolutional neural network, Directional intensity, Stereo matching algorithm, Stereo vision
Divisions:	Faculty of Electronics and Computer Engineering
Depositing User:	mr eiisaa ahyead
Date Deposited:	28 Mar 2023 10:41
Last Modified:	28 Mar 2023 10:41
URI:	http://eprints.utem.edu.my/id/eprint/26743
Statistic Details:	View Download Statistic

Actions (login required)

View Item