Evaluation and Applying Feature Extraction Techniques for Face Detection and Recognition

Abstract

Detecting the image and identifying the face has become important in the field of computer vision in recognizing and analyzing, reconstructing into 3D, and labelling the image. Feature extraction is usually the first stage in detection and recognition of the image processing and computer vision. It supports the conversion of the image into a quantitative data. Later, this converted data can be used for labelling, classifying and recognizing a model. In this paper, performance of such feature extraction techniques viz. Local Binary Pattern (LBP), Histogram of Oriented Gradients (HOG) and Convolutional Neural Network (CNN) technique are applied to detect and recognize the face. The experiments conducted with a data set addressing the issues like pose variation, facial expression and intensity of light. The efficiency of the algorithms was evaluated based on the computational time and accuracy rate.

Full-convolution Siamese network algorithm under deep learning used in tracking of facial video image in newborns

“…At present, the technology of face detection and recognition include the knowledge-based, features-based, template matching-based, and statistics-based (Fig. 1) [24][25][26].…”