An Effective Method for Recognition of Facial Expressions from Occluded Images
Shilpa Shiva · Journal of Emerging Technologies and Innovative Research · 2021
The occluded facial expression recognition (FER) technology is challenge in PC vision space. Numerous strategies have been applied to acquire exact and proficient outcomes in facial expressions. In this paper we proposed Local Binary pattern (LBP) method to recognize the occluded facial expressions from the images or faces. And also doing accuracy based comparison with the Wasserstein Generative Adversarial Network (WGAN) method and LBP method. The LBP which is a surface depiction strategy, that portrays the surface component of an image. The WGAN network which is used to measure the real/fake of the generated images. Finally, after the comparison of LBP and WGAN find which one is better method based on the accuracy performance. That is, the proposed method or already existing method. The exploratory outcome shows that the proposed method is superior to the existing method. Therefore, this method proposed in the paper realizes face recognition with occlusion in complex environment and meets the needs of practical applications. The proposed methods are tested using FER dataset.