Person Identification Using Oodb - A Fuzzy Logic Approach

Vol.3 No.4

Year: 2009

Issue: April-June

Title: Person Identification Using Oodb - A Fuzzy Logic Approach

Author Name: Thanga Ramya S 

Synopsis: 

As image databases become more and more pervasive, finding the image in large databases becomes a major problem. This thesis intends to give a solution to this problem by proposing a novel neural-fuzzy based approach for identifying the personality by comparing his/her eye image with the eye image extracted from an image database. Image database is in the form of object oriented model. In the object oriented model, operator/operand model is replaced by the object message model. In this model all the information is represented in the form of objects. An object is a self-contained entity consisting of its own private memory, a set of operations which constitute the external interface of the object with the rest of the system. Fuzzy logic provides an effective method for handling the problems with uncertain information or for dealing with the problem of knowledge representation in an uncertain and imprecise environment. Fuzzy logic is used for expressing queries in terms of the natural language. The queries designed are based on the feature “Color”. Since a gray scale image possesses some ambiguity within pixels due to the possible multi-valued levels of brightness and noise, it is justified to apply the concept of fuzzy logic to person identification. Neural network is designed to learn the meaning of the queries raised by fuzzy logic. Neural Network is a learned function that maps a list of real valued inputs to one or more Boolean or real-valued outputs. This thesis uses an Error back propagation algorithm which is used to learn the meaning of queries in fuzzy terms such as “very similar”, “similar” and “some what similar”.