Authors: T.M.P. Rajkumar; Mrityunjaya V. Latte
Addresses: JNTU Hyderabad, Faculty Department of Electronics & Communication Engineering, Anjuman Engineering College, Bhatkal, India ' JNTU Hyderabad, Faculty Department of Electronics & Communication Engineering, Anjuman Engineering College, Bhatkal, India
Abstract: A novel search algorithm called superimposed diamond search algorithm (SDSA) based on lifting wavelet transform (LWT) for medical image compression is proposed in this paper. Fuzzy C means clustering (FCM) is applied to extract the region of interest (ROI) from the medical image. MAXSHIFT method is used to scale the coefficients so that the bits associated with the ROI are placed in higher bit planes than the bits associated with the background without the requirement of the shape information and without the need for calculating the ROI mask. SDSA keeps track of significant pixels of wavelet sub-band in hexagonal search in the scan order of left to right and top to bottom. The experimental results show the good compression ratio over the other existing methods such as set partitioning in hierarchical trees (SPIHT) and embedded zerotree wavelet (EZW).
Keywords: image compression; LWT; lifting wavelet transform; FCM; fuzzy C-means clustering; region of interest; RoI; superimposed diamond search algorithm; medical images.
International Journal of Computer Applications in Technology, 2015 Vol.51 No.3, pp.185 - 192
Received: 08 May 2021
Accepted: 12 May 2021
Published online: 11 May 2015 *