Recognition of Retain Images Based on Association Rules

Abstract

The objective of feature extraction is the transformation of input data into a set of features. Features are distinctive properties of input patterns that help in differentiating between the categories of input patterns. The main idea of the proposed system depends on the feature extraction where the system uses wavelet transforms. In this paper apply colour information (RGB colour space) in a specific form of two-level discrete wavelet transforms (DWT) of Retain image; obtain seven bands of texture features are extracted from wavelet coefficients. In the paper, the algorithm depends on the idea of extracting features from retaining images person to extract association rules between these features retina images of person recognition. These rules obtain from using Frequent Pattern (FP-growth) Association Rules (AR) where give for each features code. These rules are id for retina image for the person. The system was tested over a database collected from 30 volunteers, where 15 images for each person were available on the Internet. The used algorithm is Association Rules has been used to extract association rules between features to recognize retain images of person and The achieved training rate was 100% for feature extraction, while the achieved recognition rate for wavelet transform was 82%; an excellent recognition rate and the proposed system implemented in programing language Visual Basic 6.0.