A Hybrid Approach for Facial Expression Detection using Principal Component Analysis and Feature Extraction
Shefali Goyal, Nongmeikapam Thoiba Singh, Tanu Dhiman · 2023
A number of systems have been proposed to deal with the issue of automatic emotion detection from facial images. The challenges in this field of research are enormous. The face complexion of different persons belongs to different regions of the globe, different facial accessories, and different poses are further made the detection challenging. A facial expression detection system has been proposed based on a principal component analysis (PCA) approach and feature-based technique (FBT). The system considered four categories of emotions: happy, normal, surprising, and sad. Overall recognition accuracy achieved is 98.09%. The proposed approach can be further enhanced for use in the detection of other emotion categories as well.