Face Recognition Using Various Kinds of Analysis

The proposed method is based on eigenspace [14], and WF, FFP and RFF are included in vectorized set of three features obtained by principal component analysis (PCA) [15]. Positioning of facial feature points is obtained by a new algorithm. For this purpose we will review briefly some of the previous methods in this section. 1 Face Recognition Principal Component Analysis (PCA) using Principal Component Analysis is a technique to reduce dimensions based on the number of principal components needed to extract multidimensional data.

Face recognition algorithms can be categorized into image based face recognition algorithms and video based face recognition algorithms. Both types can use some common algorithms to perform face recognition processes. This algorithm is used for image-based face recognition principal component analysis (PCA), independent component analysis (ICA), Bayesian framework, and so on. On the other hand, video-based face recognition algorithms include principal component analysis (PCA), linear discriminant analysis (LDA), 2D Fisher face algorithm, and the like.

Face emotion recognition is a computer vision application that can be used for security, entertainment, work, education, and human interface. Using a genetic algorithm and a neural network, we show a simple automatic emotion recognition and classification method that proves that any computer scientist can easily implement it. The demo system (including the code and checking the github link) contains three steps. In the first step preprocessing such as contrast adjustment, color segmentation, filtering, edge detection etc. is applied to the input image. In the second stage, features are extracted using the projected contour method. Finally, in the third stage, we calculate optimal parameters of the eyes and lips with a genetic algorithm, and classify the emotions (neutral, happy, sad, tired, angry, surprised, fearful) using artificial neural networks . To illustrate these points, the system was tested with the image of the first author Harman's face.

Due to the dynamic nature of facial images, face recognition systems encounter various problems during the recognition process. Face recognition systems can be classified as "robust" or "weak" based on recognition performance under these circumstances. The purpose of the robust face recognition system is given below: Noise invariance. A powerful face recognition system should be insensitive to noise generated by a frame grabber or camera. In addition, it should run under partially hidden images. If the face image has already been stored in the face database, the robust face recognition system should be able to classify the face image as "known" even under the above conditions.