Lossless Image Compression Using Causal Block Matching and 3D Collaborative FilteringTechnology #ua14-106
Questions about this technology? Ask a Technology Manager
Title: Lossless Image Compression Using Causal Block Matching and 3D Collaborative Filtering
Invention: This technology is a novel predictor that leverages non-local structure similarities to de-noise and de-blur images by extending block-matching and 3D collaborative filtering to image prediction and compression.
Background: Over the last few decades, substantial progress has been made in estimation methods that are adapted to the local structure in multidimensional image and video data. Block-matching and 3D collaborative filtering are algorithms developed in the last decade to aid in image de-noising, rendering older methods obsolete.
- Incorporation into a codec could provide excellent compression performance for image and video compression.
- Satellite imaging
- Medical imaging
- Time-dependent; addresses the issue of time-delay and increased resource use
- De-noises and de-blurs images
Assistant Director, Physical Sciences