Fuzzy Kernel Extreme Learning Machine for Face Recognition
Bhawna Ahuja, Virendra Prasad Vishwakarma · Proceedings of the 2022 Fourteenth International Conference on Contemporary Computing · 2022
Kernel Extreme Learning Machine (KELM) provides deterministic solution to pattern recognition problems in non-iterative manner. On contrary to ELM, the feature mapping is implicitly accomplished in KELM by practicing kernel matrix function. The performance of a learning technique relies on the capability of feature vectors. Keeping this perspective, we have applied fuzzy concept for feature extraction to develop an enhancement of KELM named as Fuzzy KELM (FKELM). In the introduced technique, the capability of fuzzy set is utilized to obtain the association of pixel elements of facial images with distinct classes. The validity of presented technique is tested on face databases. The empirical results illustrates that the presented technique outperform ELM variants for multi-class classification problem.