Author:
Minh Ly Duc ,Dong Phan Tan
Abstract
This paper studies the application of the Binary particle swarm optimization (BPSO) algorithm to the optimal search for facial features and gender classification by K-Nearest Neighbors (K-NN) model. The results show that the accuracy and processing time of the model is much better than that of VGG16, VGG19, Resnet50, Senet50, Face Net, Open Face and FbDeep Face models. a large-scale GenderFace80K dataset with 80,000 facial images with gender annotation used in the research model.
Reference15 articles.
1. Mai T. V. Le T. D. Vinh T. H. and Tram A. N. T. (2021), Unsupervised gender prediction based on deep facial features, Zooming Innovation in Consumer Technologies Conference (ZINC), pp.1-4, DOI: 10.1109/ZINC52049.2021.9499250.
2. Muhammad. N. F. and Adhi. P. (2019), Gender classification using fisher face and support vector machine on face image, Signal and Image Processing Letters,Vol. 1, No. 1, pp. 32-40, ISSN: 2714-6677, doi. Prefix 10.31763.
3. Hui. C. L. and Bao. L. L. (2006), Multi-view Gender Classification Using Local Binary Patterns and Support Vector Machines, Lecture Notes in Computer Science (LNCS), Vol. 3972, pp. 202–209, ISBN: 3-540-34439-X.
4. Seda. Ç. and Gökhan. Ş. (2016), Gender Prediction by Using Local Binary Pattern and K Nearest Neighbor and Discriminant Analysis Classifications, Signal Processing and Communication Application Conference (SIU), Vol. 24, pp.1-4, ISBN: 978-1-5090-1679-2, DOI: 10.1109/SIU.2016.7496201.
5. Temilola. M. A. Matthias. O. O. Adebanjo. A. and Oluwatobi. A. A. (2018), Gender Classification based on Local Binary Pattern and K-Nearest Neighbour, Proceedings of the International Conference on Global & Emerging Trends (ICGET) 2018, pp. 74-78 ,ISSN:2645-2405.