Useful Terminology Related to Image Processing and Face Recognition

0
328

Digital Image Processing and Face Recognition are very wide field and hence there are thousands of terms. But, here we thought to share brief definition of those useful terms that we encountered during research work.

Discontinuity measure

Discontinuity measure determines the level of discontinuity at each pixel.

Distance matrix

In mathematics, computer science and graph theory, a distance matrix is a matrix (two-dimensional array) containing the distances, taken pair-wise, of a set of points. This matrix will have a size of N xN , where N is the number of points, nodes or vertices (often in a graph).

Enrolment phase

Enrolment phase in bio-metric recognition application refers to the phase of recognition process during which either system is trained using gallery set (or feature vectors that are derived from gallery set) for future use (testing and deployment).

Face recognition

Face recognition is a process of automatically identifying or verifying a person from a digital image or a video frame from a video source by using facial features.

Feature vector

A feature vector is an n-dimensional vector of numerical features that represent some object. Many pattern recognition algorithms require a numerical representation of objects, since such representations facilitate processing and statistical analysis.

Filtering

In image processing, a filtering is a process that changes the appearance of an image or part of an image by altering the shades and colours of the pixels in some manner. Filters are used to increase brightness and contrast, to eliminate noise, to sharpen edges, as well as to add a wide variety of textures, tones and special effects to a picture.

Fuzzy partition matrix

In clustering, fuzzy partition matrix is a matrix that defines fuzzy partitions and stores a membership value for each data element with respect to each of the clusters.

Fuzzy set

A Fuzzy Set is any set that allows its members to have different grades of membership (membership function) in the interval [0,1].

Gallery set

It is a set of images that are already known to a bio-metric recognition system.

Grayness ambiguity

Grayness ambiguity can be quantified by considering the fuzzy boundaries of regions based on global gray value distribution and the rough resemblance between nearby gray levels.

Homogeneity

Homogeneity is defined as the quality or state of being homogeneous. Homogeneous means- of the same or similar nature.

Identification system

An identification system recognises an individual by searching the entire database for a match. It conducts one-to-many comparisons to establish the identity of the individual. In an identification system, the system establishes a subject’s identity (or fails if the subject is not enrolled in the system database).

Illumination

Illumination is the state of being illuminated. Also, it is referred as the effect of light on different parts of the image.

Illumination normalisation

Illumination normalisation is a process of reducing or overcoming the effect of illumination from a given subject or image.

Image segmentation

Segmentation refers to the process of partitioning a digital image into multiple segments (sets of pixels). The goal of segmentation is to simplify and/or change the representation of an image into something that is more meaningful and easier to analyse. Image segmentation is typically used to locate objects and boundaries (lines, curves, etc.) in images. More precisely, image segmentation is the process of assigning a label to every pixel in an image such that pixels with the same label share certain visual characteristics.

Imposter image

An image which is not present in gallery set. It is used for finding false acceptance rate.

In-homogeneity measure

It is any quantitative metric of measuring in-homogeneity. This term is highly used in Digital Image Processing for measuring homogeneity within images.

Linear discriminant analysis

Linear Discriminant Analysis (LDA) and the related Fisher’s linear discriminant are methods used in statistics, pattern recognition and machine learning to find a linear combination of features which characterise or separate two or more classes of objects or events. The resulting combination may be used as a linear classifier. Many times, it can also be used for feature vector dimension reduction before the actual classification.

Membership function

Membership function is the characteristic function of a fuzzy set, which assigns to each element in a universal set a membership value between 0 and 1.

Neighbourhood size

In a given image, neighbourhood size is the number of pixels near to a given pixel within a radius. Radius is a spatial distance between a given pixel and neighbouring pixels under consideration.

Retinex

The word “retinex” is formed from retina and cortex, suggesting that both the eye and the brain are involved in the image perception/processing by human beings. Retinex theory is related to image formation process.

ROC

A Receiver Operating Characteristic (ROC), or simply ROC curve, is a graphical plot of the recognition rate vs. false acceptance rate, for a classifier system as its discrimination threshold is varied. The ROC can also be represented equivalently by plotting the FAR, FRR, and RR with different thresholds. ROC analysis provides tools to select possibly optimal models and to discard sub-optimal ones.

Similarity matrix

A similarity matrix is a matrix of scores which express the similarity between two data points. Similarity matrices are strongly related to their counterparts, distance matrices and substitution matrices.

Universe of discourse

The universe of discourse is the range of all possible values for an input to a fuzzy system.

Verification system

A verification system authenticates a person’s identity by comparing the captured bio-metric characteristic with his/her own bio-metric template(s) pre-stored in the system. It conducts one-to-one comparison to determine whether the identity claimed by the individual is true. A verification system either rejects or accepts the submitted claim of identity.

SHARE
Previous articleWhy Fuzzy Image Processing?
Next articleVarying Illumination – Challenging Issue in Face Recognition
Dr R M Makwana
Dr. Makwana is Ph.D. in Computer Engineering, specialized in Artificial Intelligence from Sardar Patel University, Anand, Gujarat, India. Accelerated career growth from lecturer to professor in short span, having teaching experience of more than 13 years. He is TechSavvy with Research interest in Artificial Intelligence, Image Processing, Computer Vision, and Internet of Things. Actively supporting research community by providing service as a member of technical program committees of national and international conferences and workshops, as well as by reviewing journal and conference papers.

LEAVE A REPLY