CoChat

Local feature extraction based facial emotion recognition: a survey

Open access
Slimani, Khadija, Kas, Mohamed, El Merabet, Youssef, Ruichek, Yassine, Messoussi, Rochdi · 2020 · International Journal of Electrical and Computer Engineering (IJECE)

The DOI matches its Crossref record; 64 of the 66 references it cites were found in a registry or an index; none of the 13 sources its claims cite is checked for support yet.

Abstract

Notwithstanding the recent technological advancement, the identification of facial and emotional expressions is still one of the greatest challenges scientists have ever faced. Generally, the human face is identified as a composition made up of textures arranged in micro-patterns. Currently, there has been a tremendous increase in the use of local binary pattern based texture algorithms which have invariably been identified to being essential in the completion of a variety of tasks and in the extraction of essential attributes from an image. Over the years, lots of LBP variants have been literally reviewed. However, what is left is a thorough and comprehensive analysis of their independent performance. This research work aims at filling this gap by performing a large-scale performance evaluation of 46 recent state-of-the-art LBP variants for facial expression recognition. Extensive experimental results on the well-known challenging and benchmark KDEF, JAFFE, CK and MUG databases taken under different facial expression conditions, indicate that a number of evaluated state-of-the-art LBP-like methods achieve promising results, which are better or competitive than several recent state-of-the-art facial recognition systems. Recognition rates of 100%, 98.57%, 95.92% and 100% have been reached for CK, JAFFE, KDEF and MUG databases, respectively.

Overview

CoChat Summary

## TL;DR

This is a survey and experimental comparison of 46 recent local binary pattern (LBP)-like texture descriptors for facial emotion recognition. Across four benchmark datasets, several graph-based LBP variants performed very strongly, with reported recognition rates of 100% on CK and MUG, 98.57% on JAFFE, and 95.92% on KDEF for the best methods.

## Key findings

  • The paper evaluates 46 state-of-the-art LBP variants on four facial expression datasets: CK, JAFFE, KDEF, and MUG.
  • On CK, 27 LBP-methods (including RALBGC, BGC1, BGC2, BGC3, dLBPα, ELGS) achieved 100% average accuracy; all evaluated descriptors reached 100% for Happy and Surprise.
  • On JAFFE, the best results were 98.57% from DSLGS, ELGS, and SLGS; on the class sadness, accuracies dropped to 50%–90% for the tested descriptors.
  • On KDEF, LGS was best at 95.92%; on MUG, 8 descriptors (ELGS, LDTP, LDENP, LGCP, LNDP, LTP, LQPAT, SMEPOP) reached 100%, while LBP reached 98.73%, DBC 98.05%, XCS LBP 97.40%, and GTUC 97.08%.

## Methods

  • The study uses a supervised image classification pipeline: face preprocessing/resizing, feature extraction with LBP variants, region-wise histogram formation over non-overlapping sub-regions, and SVM classification.
  • Datasets and splits stated in the paper: CK subset of 10 sequences, with the four latest frames used and 2 training / 2 test images per emotion; JAFFE with 213 images; KDEF with 70 subjects, 980 images at 256×256; and MUG with 924 images at 256×256.
  • The paper reports average per-class accuracies for the 7-expression problem and compares the best-performing descriptor on each dataset against recent state-of-the-art methods.

## Limitations

  • The authors state that comparisons with other state-of-the-art methods are not directly comparable because of differences in dataset partitioning, number of classes, number of subjects, and features used.
  • The paper notes that it used the half-half configuration on its own evaluation, while many existing systems use settings with more training than test images; this makes the comparison more difficult (stated in the paper).
  • Evident from the text: the study is a survey/comparative experiment on handcrafted LBP-like features only, so it does not evaluate deep features or deep classifiers; the authors explicitly list these as future work.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.11591/ijece.v10i4.pp4080-4092

Cites

64 of 64 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Outline of the adopted facial emotion recognition system.
Figure 1 Outline of the adopted facial emotion recognition system.CC-BY-SA · From the PDF

Verification

Claims the paper makes

0 Valid · 0 Not valid · 7 Not checked · 39 Own results and interpretations

24 more claims

Create a free account to see all 46 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 15
  • The quote only partly states this claim. 1
  • Probably not a substantial claim. 3

Create a free account to see these claims.

Citations

Cites

Show all 64 references

Research works printed but not linked: 6

  1. [3]3. S. Bourekkadi, S. Khoulji et al., "The design of a psychotherapy remote intelligent system," Journal of Theoretical and Applied Information Technology, vol. 93, no. 1, p. 116, 2016.
  2. [22]22. M. Islam, "Local gray code pattern (lgcp): A robust feature descriptor for facial expression recognition," International Journal of Science and Research (IJSR), India Online ISSN, pp. 2319-7064, 2013.
  3. [30]30. S. Rajput and J. Bharti, "A face recognition using linear-diagonal binary graph pattern feature extraction Int J Elec & Comp Eng, Vol. 10, No. 4, August 2020 : 4080 -4092 Int J Elec & Comp Eng ISSN: 2088-8708 Ì 4091
  4. [31]31. method," International Journal in Foundations of Computer Science & Technology, vol. 6, no. 2, 2016.
  5. [56]56. pressions," in Proceedings of the Tenth Indian Conference on Computer Vision, Graphics and Image Processing. ACM, 2016, p. 25.
  6. [58]58. S. A. Shah, G. Ali et al., "Cross-cultural emotion analysis a clustering approach," International Journal of Computer Science and Network Security, vol. 17, no. 8, pp. 35-41, 2017.

Cited by

See all 15

Metadata

Authors
Slimani, K., Kas, M., El Merabet, Y., Ruichek, Y., & Messoussi, R.
Affiliations
  • Université Ibn-Tofail (MA)
  • Université Bourgogne Franche-Comté (FR)
Venue
International Journal of Electrical and Computer Engineering (IJECE)
Published
2020-08-01
Volume
10
Issue
4
Pages
4080
ISSN
  • 2722-256X (ISSN-L)
  • 2722-2578
  • 2088-8708
OpenAlex
W3016519690
Cited by
15
References
66
Citations by year
2026: 1 · 2025: 1 · 2024: 3 · 2022: 7 · 2021: 2 · 2020: 1
FWCI
0.97
Type
journal article
Language
en
Open access
Locations
  • International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering (publishedVersion, cc-by-sa)
  • Zenodo (CERN European Organization for Nuclear Research) (submittedVersion, cc-by)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face and Expression Recognition
  • Physical Sciences › Computer Science › Artificial Intelligence › Emotion and Mood Recognition
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
Keywords
facial emotion recognition, local binary pattern, texture features, local feature extraction, facial expression recognition, JAFFE database, CK+ database

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-06 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Local feature extraction based facial emotion recognition: a survey. https://app.cochat.ai/papers/10.11591/ijece.v10i4.pp4080-4092
The paper
Slimani, K., Kas, M., El Merabet, Y., Ruichek, Y., & Messoussi, R. (2020). Local feature extraction based facial emotion recognition: a survey. International Journal of Electrical and Computer Engineering (IJECE). https://doi.org/10.11591/ijece.v10i4.pp4080-4092

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.