34N-LBPH Algorithm Incorporated with CNN For Face Recognition Under Constrained Environment

Solomon Anab, Godbless Mensah, Ebenezer Komla Gavua, Mustapha Adamu Mohammed

Abstract


Along the security systems and interfaces up to human computer interaction, face recognition is implemented in various applications and usages. In this domain, the Local Binary Pattern Histogram (LBPH) algorithm has come up as probable due to the versatility in the texture based features and also due to the inherent resistance. In this work, the 34N-LBPH is integrated with CNN, further optimization through the machine learning approach to enhance face recognition in a restricted illumination condition. A comprehensive experimentation was undertaken to evaluate the performance of the hybridised algorithms using constrained face datasets: ORL Dataset. The hybridised technique was then compared with the traditional LBPH technique under constrained environment. The performance analysis revealed that the proposed hybridised face recognition technique achieved a higher Accuracy of 95.25% than the original LBPH of 94.23%, which constitutes a significant improvement and indicates that the suggested technique accurately classified 1.02% more samples than LBPH normal. The specific conclusion drawn therefore posits that the hybridised technique is more capable of handling the facial recognition tasks as compared to the more conventional LBPH algorithm in cases of varying lighting conditions and even variations in pose to improve the face recognition system

Keywords


Accuracy, Algorithm, Face Recognition, LBPH.

References


Y. Liu, S. Li, X. Wu, and Y. Jiang, "Hybrid LBPH- CNN based approach for face recognition in complex environments," Pattern Recognition Letters, vol. 131, pp. 123-130, 2020.

J. Wang, H. Chen, and P. Hu, "Improved face recognition using enhanced LBPH and CNN fusion model," IEEE Access, 2021.

T. Zhang, Z. Luo, and X. Li, "Robust face recognition under occlusion and varying lighting using LBPH-CNN hybrid models," Computer Vision and Image Understanding, vol. 204, p. 103189, 2021.

X. Chen, M. Wang, and W. Zhang, "Real-time face recognition in dynamic environments: Enhancing LBPH with deep learning," Neural Computing and Applications, vol. 34, no. 12, pp. 9371-9380, 2022.

T. Ojala, M. Pietikäinen, and D. Harwood, "A comparative study of texture measures with classification based on featured distributions," Pattern Recognition, vol. 29, no. 1, pp. 51-59, 1996..

A. Singh, M. Sharma, and S. Pandey, "A hybrid approach combining LBPH and HOG for face recognition with expression variations," Journal of Image Processing and Computer Vision, vol. 23, no. 2, pp. 88-95, 2022.

T. Ahonen, A. Hadid, and M. Pietikäinen, "Face recognition with local binary patterns," in Computer Vision – ECCV 2004: 8th European Conference on Computer Vision, Prague, Czech Republic, May 11- 14, 2004. Proceedings, Part I, vol. 3021, pp. 469- 481. Springer, Berlin, Heidelberg, 2004.

X. Sun, J. Wang, and R. Li, "Enhancing LBPH with preprocessing techniques for improved face recognition," Journal of Computer Vision and Pattern Recognition, vol. 45, no. 3, pp. 234-245, 2021.

V. M. Tiryaki, U. Adia-Nimuwa, V. M. Ayres, I. Ahmed, and D. I. Shreiber, "Texture-based segmentation and a new cell shape index for quantitative analysis of cell spreading in AFM images," Cytometry Part A, vol. 87, no. 12, pp. 1090-1100, 2015

J. Urdal, "Image processing and classification of urothelial carcinoma using tissue sample images," M.S. thesis, University of Stavanger, Stavanger, Norway, 2016.

S. Anab, G. Mensah, E. Y. Baagyere, and M. M. Adamu, "An enhanced LBPH algorithm for robust face feature extraction," Int. J. Multidisciplinary Res., vol. 6, no. 5, pp. 27853, Sep.-Oct. 2024, Doi: 10.36948/ijfmr.2024.v06i05.27853.

C. Shi, S. Luo, M. Xu, and J. Tang, "Learning gradient fields for molecular conformation generation," in Proc. Int. Conf. Mach. Learn., PMLR, Jul. 2021, pp. 9558-9568.

Y. Li, Y. Zhang, R. Timofte, L. Van Gool, L. Yu, Y. Li, and X. Wang, "NTIRE 2023 challenge on efficient super-resolution: Methods and results," in Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition., 2023, pp. 1922-1960.

J. Zhang, W. Li, and G. Sun, "LBPH-CNN hybrid for robust face recognition under pose variations," IEEE Trans. Biometrics, Behaviour., Identity Sci., vol. 4, no. 1, pp. 53-63, 2022.

X. Wang, H. Liu, and Y. Feng, "Multi-scale optimization of radius and neighbouring points for LBPH," Int. J. Pattern Recognition. Artificial. Intelligence., vol. 34, no. 10, p. 2058003, 2020.


Full Text: PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

IT in Innovation IT in Business IT in Engineering IT in Health IT in Science IT in Design IT in Fashion

IT in Industry @ http://www.it-in-industry.com . ISSN (Online): 2203-1731; ISSN (Print): 2204-0595