Image compressing using k-means clustering:
K-means clustering is a popular method
of vector quantization used for image processing. If you consider a normal 512
X 512 color image, its size is 512X512X24 bits = 0.75MB (uncompressed). The
images can be compressed to have a lesser memory footprint on the hard-disk.
K-means clustering is used to compress the image by quantizing small blocks of
pixels (2 X2, 4 X 4 etc.) to a fixed code-book (blocks of pixels). Therefore,
the entire image can be represented using only a smaller no. of blocks
(code-book blocks) and hence requires lesser memory space. This project is
generalized for both gray-scale and color images of different dimensions.
Clustering technique is extensively used
in various applications in engineering, statistics and numerical
analysis. The k-means algorithm is by far the most popular
clustering tool used in scientific and industrial applications.
No comments:
Post a Comment