|
|||||||
|
Wavelet for Medical Image Enhancement to Assist Resizing Varsha H. Patil1 and Dattatraya S. Bormane2 1Assistant Professor, Computer Engineering Department, K. K. Wagh Institute of Engineering Education and Research, Nashik (MS) India 2Principal, Bharati Vidyapeeth’s College of Engineering for Women, Pune (MS), India. ABSTRACT The Medical Image Processing, Visualization and Analysis involve quantitative analysis and visualization of medical images of numerous modalities such as PET, MRI, CT, or microscopy. Often a medical image analysis involves resizing of the image. Synthetic zooming of region of interest is important application in medical imaging. If the specified size does not produce the same aspect ratio as the input image, the output image will be distorted. Even with low pass filtering, resizing an image can introduce artifacts, because information is always lost when you reduce the size of an image. Image interpolation is widely used in resizing the image. But it cannot recover high frequency components lost. Extrapolating image from a blurred image increases sharpness. We present a wavelet based super resolution algorithm to suit the need of image resizing. In wavelet technique, the image is decomposed into approximation, vertical details, horizontal details and diagonal details. In multilevel wavelet the image is decomposed into multiple levels and the high frequency components are separated out. High frequency component is used in extrapolation using which high-resolution image is constructed. The objective is the humanizing the physician's ability to analyze image. Keywords: Enhancement, Interpolation, Objective evaluation and Subjective evaluation, Super resolution, Wavelet.
Biography:
BibTex: @ARTICLE{P1150608001,
AUTHOR = {Varsha
H. Patil and Dattatraya S.
Bormane}, TITLE = {Wavelet for Medical Image Enhancement to Assist Resizing }, JOURNAL = {ICGST International Journal on Graphics, Vision and Image Processing}, YEAR = {2005}, MONTH={DECEMBER}, VOLUME={05}, ISSUE = {9}, PAGES={63--68} } |
|||||||
|
|