Image compression using Multilayer Feed Forward Artificial Neural Network with Conjugate Gradient

The performance of Multilayer Feed Forward Artificial Neural Network in image compression using Conjugate Gradient algorithm is examined in this paper. One of the essential factors that affect the performance of Artificial Neural Networks is the learning algorithm. In this paper we presented the algorithm for implementation of digital image compression using MFFANN with 64 input neurons, 13 or 64 neurons in hidden layer that are determining compression rate and 64 output neurons. Based on Conjugate Gradient algorithm compressed for TIF, JPEG, PNG and BMP images. Compression of image in any form is an active field and big business. Image compression is a subset of this huge field of data compression, where we undertake the compression of image data specifically. The performance of compression is evaluated using some standard images. It is shown that the development of architecture and training algorithm provide high compression ratio. The results of simulation are shown and compared different quality parameter of it´s by applying on various images.