Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/170934
Title: 3D image segmentation using multiresolution and wavelwts
Researcher: M NIVAS
Guide(s): M. RAMAKRISHNAN
Keywords: 
University: Bharath University
Completed Date: 2016
Abstract: quotMedical volume segmentation obtained the attraction of numerous researchers; as a result, several techniques have been implemented in terms of medical imaging including segmentations and other imaging processes. This research work focuses on an implementation of segmentation system which employs the Multiresolution analysis (MRA) techniques together or on their own to segment medical volumes, the system acquires a stack of 2D slices or a complete 3D volumes acquired from medical scanners as a data input. It aims at developing an automatic image segmentation scheme for categorizing region of interest (ROI) in medical images which are attained from diverse medical scanners such as PET, CT or MRI. Two major schemes have been implemented in this research for segmenting medical volume. The first approach employs 2D wavelet, Ridgelet and Curvelet transforms based (MRA) techniques for segmenting medical volume. It is mainly a tough task to classify cancers in the scanners output of the human organs by means of shape or gray-level information; organs shape changes throw different portions in medical heap and the gray-level intensity overlie in soft tissues. Multiresolution analysis based segmentation joined with thresholding as newlinepre- and post processing steps let precise detection of ROIs. Curvelet transform is a latest extension of wavelet and ridgelet transforms which focuses on dealing with exciting phenomena taking place along curves. Higher dimensions of discontinuity (line or curve singularity) have been taken out in medical images by means of a modified multi-resolution analysis transforms such as Ridgelet and Curvelet transforms. newline newline 3D volume segmentation aims at separating the voxels into 3D objects (sub-volumes) which signify significant physical entities. MRA permits for the conservation of an image according to definite levels of resolution or blurring. The excellence of this technique makes it valuable in image compression, de-noising, and classification or segmentation .The second implemented approach in
Pagination: 
URI: http://hdl.handle.net/10603/170934
Appears in Departments:Department of Computer Science and Engineering

Files in This Item:
File Description SizeFormat 
chapter 1.docAttached File623 kBMicrosoft WordView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.