Design and Implementation of Facial Expressions Recognition based on LBP & DRLBP with CNN

Abstract

Facial Expression Recognition (FER), which is the main way to figure out what someone means without them saying it, is an important and promising area of research in computer vision and artificial intelligence. FER is also the main way that nonverbal intentions are figured out. This report gives a clear and concise summary of what's been going on in FER recently. To start, we will divide the current methods for FER into two main groups: those that are considered "traditional" and those that use "deep learning." We suggest a general structure for a classic FER method and look at the technologies that could be used for its different parts. Our goal is to show how they are similar and how they are different. Four different state-of-the-art neural network-based FER approaches are given as examples of deep learning-based techniques, and each one is then looked at in detail. We also give a description of seventeen FER datasets that are often used, as well as a summary of four characteristics of datasets that are linked to FER and could affect the choice of method and the way it is processed. We talk about evaluation methods and metrics, compare the results of different FER approaches using benchmark datasets, Design and implementation of Facial Expressions Recognition Based on Local Binary Pattern (LBP)&Dominant Rotated Local Binary Patterns (DRLBP).