Face Recognition And Human Image Processing Ability

Introduction

Human being relies heavily on vision or eyesight for most of their daily activities. The eye is the most frequently used sensor in our bodies. It is said that four-fifths of all information our brain receives comes from the eyes. It enables visual perception, allowing individuals to assimilate and interpret information about their surrounding environments. The human visual system is one of the most complex systems in our anatomy, it’s been studied for centuries, neuroscientists are still deciphering how image is generated and processed. Psychological studies show that the human image processing ability is exceedingly dynamic. Computer vision tries to emulate or copy the human visual system. One of the most studied system is the people/facial recognition. A seemingly trivial function of the Human brain, proved challenging to recreate artificially. Humans can successfully recognize and identify a variety of faces regardless of time span (years of separation), ageing factor (bearded, greying hair, gain or loss of weight…), artificial changes (wearing glasses, piercings, tattoos, hair styles, plastic surgeries…), facial expressions (anger, happiness, sadness…), multi-dimensional view of the face, and various other changing conditions. This technology can be widely used in Biometrics, crowd surveillance, bank security, airport security, cell phone security, inline production, search and rescue missions, access control systems, human computer interface, and others.

Background

The widespread of facial Biometric systems in identification, authentication, safety, and security systems triggered major research in this field. Biometric identification is the technique of automatically identifying or verifying an individual by a physical characteristic or personal trait. The term “automatically” means the biometric identification system must identify or verify a human characteristic or trait quickly with little or no intervention from the user. Biometric technology was developed for use in high-level security systems and law enforcement markets. The key element of Biometric technology is its ability to identify a human being and enforce security. Face biometric system is non-invasive, it doesn’t require subjects physical contact with the recognition equipment. Thus, making it preferable over other Biometric Technologies.Research on fully automated face recognition started in 1960s, featuring works Woody Bledsoe, Helen Chan Wolf, and Charles Bisson especially in face feature (eyes, ears, nose, and mouth) localization. In 1964 & 1965, Bledsoe, along with Helen Chan and Charles Bisson used computer to recognize human faces. Since then various techniques have been developed such as principal component analysis (PCA), Independent component analysis (ICA), Linear Discriminant Analysis (LDA),Neural networks, machine learning, information theory, geometrical modelling, skin color template matching, Hough transform, extraction of geometrical facial features, motion extraction, colour analysis, feature descriptors and extraction.

The Study Concept

Face processing is based on the principle that information about a user’s identity can be extracted from an images and the computers can process this information accordingly. The face is detected once a person’s face comes into view. Once it is detected, the face region is cropped from the image to be used as a “Probe(input)” into the knowledge to check for possible matches. In the Face recognition phase, the input image is compared to every image in database. The input image is also called a probe and the database is called gallery or knowledge. The face image is pre- processed for factors such as image size and illumination and to detect particular features. The data from the image is then matched against the knowledge (database of stored face images). The matching algorithm will produce a similarity measure for the match of the probe face into the knowledge. To build fully automated systems, robust and efficient face detection algorithms are required.

Face recognition algorithms consists of three fundamental steps:

1. Detection
2. Feature extraction
3. Recognition.

The first step in face recognition system is to detect the face in an image. The main objective of face detection is to find whether there are any faces in the image or not. If the face is present, it returns the location of the image and extent of the each face. Pre-processing is done to remove noise and reliance on the precise registration. Several factors makes this process challenging; varying illumination, facial expression, face and image orientation, occlusion, pose. Facial feature detection detects the presence and location of features (nose, eyebrow, eyes, lips, nostrils, mouth, ears…) on the face.