Iris recognition analysis using biorthogonal wavelets tranform for feature extraction

Human iris has a very unique pattern which is possible to be used as a biometric recognition. To identify texture in an image, texture analysis method can be used. One of method is wavelet that extract the image feature based on energy. Wavelet transforms used are biothogonal types, i.e. Haar and Daubechies. In this research, iris recognition based on Haar and Daubechies was done and then comparison analysis was conducted for which some conclusions taken. Some steps have to be done in the research. First, the iris image is segmented from eye image then enhanced with histogram equalization. The method used for extracting features are Haar and Daubechies (i.e. db5) wavelets transform. The features obtained is energy value. The next step is recognition using normalized Euclidean distance. Comparison analysis is done based on recognition rate percentage with two samples stored in database for reference images. As the result, the highest recognition rate is achieved using Haar with decomposition level 3 i.e. 84.375%, for which the highest recognition rate of db5 is 68.75% with decomposition level 2. The lowest recognition is achieved when db5 used with decomposition level 1, i.e. 38.231%, whereas the lowest recognition rate using Haar is 68.75% with decomposition level 1.