Video Compression Using Neural Networks

The U.S. Navy has identified a need for development of an innovative data compression technique to facilitate distribution of video and animated environmental products over low bandwidth communication channels. To address this need, American Research Corporation of Virginia (ARCOVA) proposes the development of an intelligent data compression interface for transmission of video imagery, graphics and multi-dimensional array fields to environmental computer systems at shore sites and ships. The program technical objectives include identification of environmental database specifications, selection of an optimal neural network paradigm, training and testing of the neural network paradigm selected, and design of a user interface. Successful completion of these objectives will result in an automated data reduction system capable of achieving maximal compression with minimal loss. The significance of this effort is the innovative use of neural networks for compression and distribution of video and animated data for use on newly developed environmental computer equipment and systems. Since neural networks have been shown to be faster and more accurate than traditional data compression techniques in similar applications, ARCOVA will demonstrate a proof-of-concept system for video image compression that is superior to current methods.