Abstract
A new method for image coding and reconstruction is described. The gray level of each pixel of an image is estimated from a weighted sum of gray levels of its neighbors on every side. The weight coefficients are determined so that the mean square estimation error is minimized. An optimal size of the neighborhood is also determined from the statistical properties of the image. Estimation errors are transmitted together with the weight coefficients and boundary values. on the receiver side, an iterative algorithm based on the conjugate gradient method is used to reconstruct the original image.
| Original language | English |
|---|---|
| Pages | 530-532 |
| Number of pages | 3 |
| Publication status | Published - 1979 Jan 1 |
| Event | Proc of the Int Jt Conf on Pattern Recognition, 4th - Kyoto, Jpn Duration: 1978 Nov 7 → 1978 Nov 10 |
Other
| Other | Proc of the Int Jt Conf on Pattern Recognition, 4th |
|---|---|
| City | Kyoto, Jpn |
| Period | 78/11/7 → 78/11/10 |
ASJC Scopus subject areas
- Engineering(all)