Speech and Audio Coding for Wireless and Network Applications
Speech and Audio Coding for Wireless and Network Applications
Digital Image Restoration
| Hi-index | 0.04 |
In this paper, we present an image decompression schemethat emphasizes visual perception rather than objective signalintegrity between the orginal and the reconstructed images. Thedecoding scheme is developed in the context of vector quantization(VQ). The technique includes a projection based image recoveryalgorithm as part of the decoding process. The quantized blocksare classified into edge and shade blocks. The edge blocks arefiltered using optimized templates while the shade blocks areleft unaltered. The method can be considered as a constrainedsmoothing of edges, which does not cause any degradation to theshade (background) regions in the decoded image. The result isa reconstruction that is subjectively better than a simple look-uptable decoding.