Title: Local directional double ternary coding pattern for facial expression recognition

Authors: Chebah Ouafa; Laskri Mohamed Tayeb

Addresses: Laboratory of Research in Informatics (LRI), Department of Computer Science, Badji Mokhtar University, BP 12 Annaba 23000, Algeria ' Laboratory of Research in Informatics (LRI), Department of Computer Science, Badji Mokhtar University, BP 12 Annaba 23000, Algeria

Abstract: This paper presents a novel texture descriptor, the local directional double ternary coding pattern (LDDTCP) that combines the directional information from LDP and the ternary description from LTP for representing facial expression. The proposed LDDTCP operator encodes the image texture by computing the edge and line responses values using the 8-direction-based Frei-Chen masks. To achieve robustness, the obtained eight Frei-Chen masks are partitioned into two groups according to their directions. After calculating the average of each group, we assign three discrimination levels to each pixel based on the edge responses values in the first group and the line response values in the second group, obtaining LDDTCP-1 and LDDTCP-2 codes, respectively. The last feature descriptor vector LDDTCP is formed by concatenation of both LDDTCP-1 and LDDTCP-2 histograms. Experimental results using the CK and JAFFE database show that the LDDTCP descriptor achieves superior recognition performance compared to some existing local descriptor methods.

Keywords: facial expression recognition; human face; appearance descriptor; geometry descriptor; local binary pattern; LBP; local directional pattern; LDP; local ternary pattern; LTP; support vector machine; SVM.

DOI: 10.1504/IJCVR.2023.130646

International Journal of Computational Vision and Robotics, 2023 Vol.13 No.3, pp.259 - 284

Received: 03 Jan 2021
Accepted: 11 Feb 2022

Published online: 02 May 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article