25-08-2017, 09:32 PM
Color Image Indexing Using BTC
Color Image Indexing Using BTC.pdf (Size: 1.83 MB / Downloads: 23)
INTRODUCTION
THE rapid expansion of the Internet and fast advancement
in color imaging technologies have made digital color images
more and more readily available to professional and amateur
users. The large amount of image collections available
from a variety of sources (digital camera, digital video, scanner,
the Internet, etc.) have posed increasing technical challenges to
computer systems to store/transmit and index/manage the image
data effectively and efficiently to make such collections easily
accessible.
The storage and transmission challenge is tackled by image
coding/compression, which has been studied for more than
30 years and significant advancements have been made. Many
successful, efficient and effective image-coding techniques
have been developed and the body of literature on image coding
is huge. Well-developed and popular international standards,
e.g., [6], on image coding have also long been available and
widely used in many applications.
BTC FOR COLOR IMAGE CODING
Block truncation coding (BTC) was first developed in 1979
for greyscale image coding [5]. This method first divides the
image to be coded into small nonoverlapping image blocks (typically
of size 4 4 pixels to achieve reasonable quality). The
small blocks are coded one at a time. For each block, the original
pixels within the block are coded using a binary bit-map
the same size as the original block and two mean pixel values.1
The method first computes the mean pixel value of the whole
block and then each pixel in that block is compared to the block
mean. If a pixel is greater than or equal to the block mean, the
corresponding pixel position of the bitmap will have a value of
1, otherwise it will have a value of 0. Two mean pixel values,
one for the pixels greater than or equal to the block mean and
the other for the pixels smaller than the block mean are also calculated.
At decoding stage, the small blocks are decoded one
at a time. For each block, the pixel positions where the corresponding
bitmap has a value of 1 is replaced by one mean pixel
value and those pixel positions where the corresponding bitmap
has a value of 0 is replaced by another mean pixel value.
CONCLUDING REMARKS
In this paper, we have presented a method for using a well
known image coding technique to achieve both image coding
and content based image retrieval in the compressed domain.
Two image content description features derived directly from
the compressed stream have been developed. To demonstrate
the effectiveness of using the BTC code for image retrieval, we
presented experimental results using a texture database and a
large photographic image database. Results showed that the new
method performed at least as well as, sometime even better than
a state of the art technique. One significant advantage of the
current method is that it achieves coding and retrieval simultaneously.
As with other techniques that use multimodality features,
an open question is how to set the relative weightings of
the BCCM and BPH descriptors. In this paper, we followed the
common practice and determined the weightings empirically.
As a final remark, image coding has been studied for a
long time and has also been very successful. All image coding
methods essentially try to extract the most important visual
information and represent them in a compact manner. Indexing
and retrieval for image database applications have becoming
increasingly important. We believe results and experiences of
more than 30 years image coding/compression research can
play a fruitful role in the development of effective and efficient
methods for image indexing and retrieval in large image
database. We hope the present paper has shown a glimpse of
such potential.