CoChat

Hybrid LBPH-CNN for Interactive Web Interface-Based Multi-Face Emotion Recognition

Open access
Katti, Arpana G., M V, Chidananda Murthy · 2026 · International Journal of Reconfigurable and Embedded Systems (IJRES)

The DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index.

Abstract

The foundation of contemporary human-computer interaction is facial expression recognition; yet, it is still difficult to develop algorithms that are accurate and useful for multi-face, real-world situations. In order to achieve robust, multi-face emotion classification, this research introduces a hybrid architecture that combines a proprietary Convolutional Neural Network (CNN) with Local Binary Pattern Histogram (LBPH) texture data. Streamlit is used to deploy the system as an interactive online application that allows users to observe real-time detections, upload photographs, and examine face histograms. The hybrid model outperforms pure LBPH (58.4%) on the FER2013 benchmark with 68.5% test accuracy, and it can compete with deeper CNNs that need a lot more processing power. More significantly, the system achieves nearly flawless predictions on a real-world group shot with four faces: 99.8% for Happy and 86.7% for Neutral. 48x48 greyscale inputs reduce accuracy by 10.2%, according to an ablation study on input resolution, whereas values larger than 96x96 give marginal improvements at high computational expense. The article offers a thorough analysis of LBPH-CNN fusion, contrasts it with cutting-edge techniques, and addresses the trade-offs between learned and hand-crafted features. The source code is completely open-source and can be used as a model for easily accessible and efficient emotion-aware applications.

Overview

Integrity signals

Registry

Crossref · 10.2139/ssrn.6831425

Cites

16 of 16 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Citations

Cites

Show all 16 references

Research works printed but not linked: 25

  1. K Priya, 2025, Multimodal approach to decoding facial expressions using LBPH and deep learning techniques, 2025 International Conference on Intelligent Systems, Security, and Communication (ICISSC), 1
  2. A G Katti, 2025, A fusion convolution neural network-local binary pattern histogram algorithm for emotion recognition in human, IAES International Journal of Artificial Intelligence (IJ-AI), 14, 2734, 10.11591/ijai.v14.i4.pp2734-2740 · DOI
  3. K Priya, 2025, Multimodal Approach to Decoding Facial Expressions Using LBPH and Deep Learning Techniques, 2025 International Conference on Intelligent Systems, Security, and Communication (ICISSC), 1
  4. H I Dino, Facial expression classification based on K-Nearest Neighbors, Journal of Physics: Conference Series, 1530
  5. P Kumar, 2023, Multimodal emotion recognition using facial expressions and speech: A deep learning approach, Journal of Ambient Intelligence and Humanized Computing, 14, 2345
  6. H M S Hussain, 2025, The memory efficiency in a receptionist robot's face recognition system using LBPH algorithm, Core.ac.uk
  7. 2025, Real-time emotion recognition from facial expressions using convolutional neural network with Fer2013 dataset, Amrita Vishwa Vidyapeetham
  8. F Xue, 2023, Transferable facial expression recognition with vision transformer, IEEE Transactions on Affective Computing, 14, 1156
  9. Y Zhang, 2023, Contrastive learning for facial expression recognition in the wild, IEEE Transactions on Image Processing, 32, 1234
  10. P A Baffour, 2024, Feature Extraction by Traditional Machine Learning and Deep Learning for Facial Expression Classification, IJFMR
  11. M Abdeldayem, Improving Facial Expression Recognition in real-world Environments, IAES International Journal of Artificial Intelligence
  12. A Deshmukh, 2023, Lightweight convolutional neural networks for facial expression recognition on edge devices, Proc. IEEE/CVF Winter Conf. Applications of Computer Vision (WACV), 2980
  13. Y Fan, 2023, Efficient emotion recognition from facial expressions with attention-guided knowledge distillation, IEEE Transactions on Multimedia, 25, 4321
  14. K Wang, 2022, A dual-stream network with LBP and depthwise separable convolutions for real-time facial expression recognition, IEEE Signal Processing Letters, 29, 1565
  15. R Jain, 2023, An ensemble of lightweight CNNs for facial expression recognition in resource-constrained environments, Journal of Real-Time Image Processing, 20
  16. M Z Uddin, 2023, Facial expression recognition using hybrid CNN and LBP features with attention mechanism for human-robot interaction, IEEE Transactions on Industrial Informatics, 19, 2980
  17. T Zhang, 2023, Cross-dataset facial expression recognition via unsupervised domain adaptation with style transfer, IEEE Transactions on Biometrics, Behavior, and Identity Science, 5, 537
  18. H S Kang, 2024, Real-time facial expression recognition on mobile devices using quantized CNNs and temporal smoothing, IEEE Access, 12, 23456
  19. P Gupta, 2024, Federated learning for privacy-preserving facial expression recognition in edge networks, Proc. IEEE Int. Conf. Communications (ICC), 1
  20. L Chen, 2024, Grad-CAM guided explainable facial expression recognition for clinical depression screening, IEEE Journal of Biomedical and Health Informatics, 28, 2890
  21. S R Dubey, 2023, Multi-scale LBP and lightweight CNN for facial expression recognition in the wild, Pattern Recognition Letters, 168, 89
  22. A K Tyagi, 2023, Streamlit-based deployment of deep learning models for affective computing: A case study on FER2013, Software Impacts, 18
  23. M A H Akhand, 2022, Hybrid feature extraction from LBP and CNN for robust facial expression recognition under variable illumination, Journal of King Saud University -Computer and Information Sciences, 34, 5678
  24. N Otberdout, 2022, Dynamic facial expression recognition using spatiotemporal LBP and residual networks, IEEE Transactions on Image Processing, 31, 4502
  25. T O Ajayi, 2023, Comparative analysis of Haar cascade, MTCNN, and RetinaFace for real-time face detection in emotion recognition systems, Proc. IEEE AFRICON, 1

Metadata

Authors
Katti, A. G., & M V, C. M.
Affiliations
Sri Siddhartha Academy of Higher Education (IN)
Venue
International Journal of Reconfigurable and Embedded Systems (IJRES)
Publisher
Elsevier BV
Published
2026
Volume
15
Issue
3
Pages
859-859
ISSN
2089-4864 (ISSN-L)
OpenAlex
W7162434333
Cited by
0
References
10
FWCI
0.00
Type
journal article
Language
en
Open access
Licence
cc-by-sa
Locations
  • International Journal of Reconfigurable and Embedded Systems (IJRES) (publishedVersion, cc-by-sa)
  • SSRN Electronic Journal (acceptedVersion)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Emotion and Mood Recognition
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face and Expression Recognition
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
Keywords
facial expression recognition, local binary pattern histogram, convolutional neural networks, hybrid feature fusion, Streamlit web interface, input resolution, real-time emotion recognition

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.

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

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.