Face Recognition

INTRODUCTION

• In recent years face recognition has received substantial attention from researchers in biometrics, pattern recognition.

• The strong need for user friendly system that can secure our assets and protect our privacy can be easily catered to, using systems utilizing the concept of face recognition.

What is Face Recognition?

• Given still or video images of a scene, identify or verify one or more persons in the scene using a stored database of faces.

• Fields of requirement or the aspects of face recognition are

• Face Identification – Here the face to be recognized is to be identified among various pre-captured images stored in the database. Ex: Surveillance systems

• Face Verification – Here the system has to verify the claimed identity of the person whose image the camera captured. The comparison has to occur with single existence image of the person in the database. Security systems apply this principle

The Human Approach

FACE  IDENTIFICATION

FACE VERIFICATION

• BASIC STEPS

• DETECTION

• ALIGNMENT

• MEASUREMENT

• REPRESENTATION

• MATCHING

• VERIFICATION

• ACTUAL STEPS EXPLAINATION

• INPUT IMAGE

• The input image can in the form of 2D image such as a photograph or a 3D image taken from a frame of a live video capture.

• IMAGE PROCESSING

• Image processing is done based on the input image, and there are various algorithms for processing an image. As the input image can be either a 2D image or from a video, algorithms and implementation vary accordingly.

• For image based processing(2D), algorithms are

• PCA – Principle Component Analysis,

• IDA - Independent Component Analysis,

• LDA - Linear Discriminant Analysis,

• EP – Evolutionary Pursuit,

• Trace Transform, etc.

• For video based processing(3D), there is no algorithm that can be used to implement this technology directly, but the modification are present.

• CODE GENERATION

• Based on the algorithm chosen, suitable form of code is generated that is used for further processing.

• DATABASE

• It is the collection of the images that are stored for the references and comparison. Implementation of the database can vary.

• CODE MATCHING

• Based on the output produced from the code generation, the codes are matched with those present in the database.

• SECONDARY MODULES

•  These modules are the usage blocks, i.e., the result generated from the code match, such as, storing the new picture in database, etc.

An illustrative example of FACE RECOGNITION in surveillance systems:

• Detection: 

• Accomplished by digitally scanning an existing photograph (2D) or by using a video image to acquire a live picture of a subject (3D).

An illustrative example of FACE RECOGNITION in surveillance systems:

• Alignment:

• The system determines the head's position, size and pose

• In case of 3D, the subject has the potential to be recognized up to 90 deg,

• While with 2D, the head must be turned at least 35 degrees toward the camera.

An illustrative example of FACE RECOGNITION in surveillance systems:

• Measurement:

• The system then measures the curves of the face on a sub-millimeter (or microwave) scale and creates a template.

An illustrative example of FACE RECOGNITION in surveillance systems:

• Representation:

• The system translates the template into a unique code. This coding gives each template a set of numbers to represent the features on a subject's face.

An illustrative example of FACE RECOGNITION in surveillance systems:

• Matching:

• If image is 3D and the database contains 3D images, then matching will be easy.

• When a 3D image is taken, different points are identified and measured. Once those measurements are in place, an algorithm will be applied to the image to convert it to a 2D image, so that the software can  compare the image with the 2D images in the database to find a potential match .

An illustrative example of FACE RECOGNITION in surveillance systems:

• Verification:

• The final stage- an image is matched to only one image in the database.

REFERENCES