|
|
K- Means Clustering Based Image Compression in Wavelet Domain |
N. Venkateswaran
and Y.V. Ramana Rao
|
Abstract: This study suggests a novel image compression scheme, using the discrete wavelet transformation (DWT) and k-means clustering technique. The goal is to achieve higher compression rates based on clustering the wavelet coefficients of each DWT band. This methodology is different from other schemes wherein DWT is applied to the whole original image. Here the sub blocks of size 16x16 from an image are subjected to one level wavelet decomposition and the coefficients are clustered using the well-known K-means clustering algorithm. The centriods of each cluster is arranged in the form of a codebook and indexed. The index values are only transmitted over the line. By decreasing the number of cluster, high compression ratio can be achieved. Simulation results indicate that the compression ratio varies from 56.8 to 8.0 for varying the cluster size from 10 to 100 for an acceptable image quality. |
|
[Abstract]
[Fulltext
PDF]
[Fulltext
HTML]
[References]
[View
Citation]
[Report
Citation] |
|
COMMENTS |
22 March, 2017
shree:
hi sir/mam,
im studying mphil cs in my research i can select image compression . i read ur paper its very usefull for me really nice .its a new innovative for image compression side . can u plz send me the code for refer my research work its very useful for my research plz |
|
|
REPLY TO ABOVE COMMENTS |
|