Implementation of Image Classification Using Wavelet Based Method for Eye Detection and Neural Networks
Nikita Bhardwaj, Er. Mithlesh Kumar Singh
ANN, Image, Eye Detection, Wavelet
The aim of this paper eye detection with software simulation packages such as wavelet and NN ', such as the Matlab 7.0 toolbox to verify the specificity of the human eye and its performance as a biometric. A sequential feature selection algorithm is then used to select the most appropriate features for classification. The classification is performed using neural networks that provide high accuracy. The eye detection system includes an automated segmentation system based on wavelet transform, and then wavelet analysis is used as a pre-processor for a posterior diffusion neural network with conjugate gradient learning. Inputs of neural networks are the waveform maxima neighborhood coefficients of facial images on a particular scale. The output of a neural network is the classification of inputs into an eye or non-eye region. The program code is generated using Matlab and the results are analyzed. The output is such that it classifies the dataset into normal and image classifications, using wavelet based methods for eye detection. Datasets are categorized using neural networks. The main objective is to examine the image compression of a gray scale image using wave theory. It is implemented in software using MATLAB wavelet toolbox and 2D-DWT technology. Experiments and results are performed on .jpg format images. These results provide a good reference for application developers to choose a good waveform compression system for their application.