Apparatus for encoding/decoding an image signal having a still object
AssigneeDaewoo Electronics Co., Ltd.
InventorJong-Rak Kim
Life of this patent
Estimating the term…
Abstract
An encoding system effectively reduces the substantial amount of data for transmission by way of employing three encoding channels which perform a compression process for individually extracting and coding the contour of the approximated foreground region consisting of a still object, the boundary pixel values of the pixels located on the contour and the leveled-down/trimmed pixel data corresponding to the contour and the boundary pixel values. Further, a decoding system for use with the encoding system serves to effectively perform a decoding process for a coded image signal by using three decoding channels containing the decoding path for restoring pixel information including a set of reconstructed boundary pixel values and reconstructed contour information.
Claims9
An apparatus for encoding an image signal having foreground and background regions, wherein the foreground region consists of a still object and the image signal includes contour information relative to the positions of the pixels located on the contour of the foreground region and pixel information containing the pixel values of the foreground and the background regions in the image signal, which comprises:
- first encoding means, based on the contour information, for approximating the contour of the foreground region to provide contour data representing the contour of an approximated foreground region and for converting the contour data into a coded contour signal;
- second encoding means, based on the pixel information and the contour data, for masking the values of the pixels lying outside the approximated foreground region with a zero value to generate a zero-masked pixel signal and for extracting a set of boundary pixel values of the pixels located on the contour of the approximated foreground region from the zero-masked pixel signal to generate a coded boundary signal representing the set of boundary pixel values;
- third encoding means, based on the set of boundary pixel values, for leveling down the pixel values of the approximated foreground region and, based on the contour data, for trimming a portion of the background region to generate a coded pixel signal;
- and
- means for combining the coded contour signal, the coded boundary signal and the coded pixel signal to generate a coded image signal.
An apparatus for converting a coded image signal to a reconstructed image signal, wherein the coded image signal includes a coded contour signal, a coded boundary signal and a coded pixel signal, which comprises:
- means for demultiplexing the coded image signal into the coded contour signal, the coded boundary signal and the coded pixel signal;
- first decoding means, based on the coded contour signal, for generating reconstructed contour information representing the contour of the foreground region of the coded image signal;
- second decoding means, based on the reconstructed contour information, for converting the coded boundary signal to a set of reconstructed boundary pixel values of the pixels located on the contour of a foreground region;
- and
- third decoding means, in response to the set of reconstructed boundary pixel values and the reconstructed contour information, for converting the coded pixel signal to the reconstructed image signal.
Dependencies are inferred from the claim text.