Lossless image compression using neural network

• Main Author: Chin, Heng Ee.
• Other Authors: Soon, Ing Yann
• Format: Theses and Dissertations
• Language: English
• Published: 2009
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/19781
• Institution: Nanyang Technological University

The purpose of the project is to assess the feasibility of using Neural Network as the prediction mechanism for lossless compression of images. The Neural Network model used is the Multilayer Perceptron or the Backpropagation model with one input layer, one hidden layer and one output layer. The image used for compression is a SPOT Multispectral Satellite image with three image bands: Green, Red and Infra-Red. In this project, both intra band and inter band predictive coding were implemented. In intra-band prediction, samples from the same single band are used to predict the sample to be coded. However, for inter band prediction, samples from more than one band are used.