Image Compression Using PCA and Improved Technique with MLP Neural Network

Abstract

Computer images consist of large data and hence require more space to store in the memory. The compressed image requires less storing space of memory and less time to transmit. In this paper, feed forward back propagation neural network method with PCA technique is used for image compression. It is found that the results using PCA techniques are not satisfactory. To improve these results author suggest another technique which has better results. The Artificial Neural Network (ANN) used in these technique is trained by considering the different numbers of hidden neurons, epoch and reconstructed image compared with original image.