| Summary: | 碩士 === 逢甲大學 === 資訊工程研究所 === 85 === This thesis extends and modifies Classified Vector
Quantization (CVQ) to solve the problem of inverse halftoning
from error-diffused images. The process consists of two phases:
the encoding phase and decoding phase. The encoding process
needs a codebook for the encoder which transforms a halftoned
image to a set of codewords. The decoding process also requires
a different codebook for the decoder which reconstructs a gray-
scale image from a set of codewords. According the relationships
between these two codebooks, we modified the traditional
generalized Lloyd Algorithm to fit our purpose. On the other
hand, we also developed a process for inverse halftoning of
error-diffused color image. Compared with other available
techniques, our approach has the better quality for
reconstructed image. Using CVQ, the reconstructed gray-scale
image and color image can be stored in compressed form and no
further compression maybe required. This is contrast to the
existing algorithms which reconstruct a halftoned image in an
uncompressed form. The bit-rate is about 0.52 bits per pixel for
encoding a reconstructed gray-scale image, and about 0.96 bits
per pixel for encoding a reconstructed color image. The main
contribution of this thesis is it opens an area of application
on signal reconstruction for VQ.
|
|---|
