Image Compression using Multilayer Feed Forward Artificial Neural Network and DCT

The Neural Networks are good alternative for solving many complex problems. In this paper multi-layer neural network has been employed to achieve image compression. The proposed technique breaks down large images into smaller windows and applies Discrete Cosine Transform (DCT) to these windows as a pre-process technique. The input pixels will be used as target values so that assigned mean square error can be obtained, and then the hidden layer output will be the compressed image. A compression among different back propagation training algorithms were introduced with different compression ratio ,and different block sizes were expressed. Block sizes play role in image compression even with the same compression ratio. The proposed technique has been implemented using Matlab®.