DocumentCode :
3001373
Title :
Study of medical image segmentation algorithm based multi-wavelet analysis
Author :
Wencang, Zhao ; Hongli, Wei
Author_Institution :
Autom. & Electron. Eng. Coll., Qingdao Univ. of Sci. & Technol., Qingdao
fYear :
2008
fDate :
1-3 Sept. 2008
Firstpage :
2325
Lastpage :
2328
Abstract :
Recently, with the rapid development of biomedical imaging technology, image segmentation technology has become the key technology of the medical image processing and analysis in order to extract the information from the medical images and to do the various following processing accurately. In this paper, considering the margin and the region of the object synthetically, we use the multi-resolution analysis theory and the method of minimizing regional uniformity to get the optimal threshold value to do the medical image segmentation. Through image segmentation processing and comparative experiments of the computer tomography medical images, we prove that this method can get comparatively clearer interested areas, and relatively legibled edge features, and the definition of the image also has been improved obviously.
Keywords :
edge detection; feature extraction; image resolution; image segmentation; medical image processing; wavelet transforms; biomedical imaging processing; edge feature extraction; medical image segmentation algorithm; multi resolution analysis theory; multi wavelet analysis; optimal image threshold value; regional uniformity minimization; Algorithm design and analysis; Automation; Biomedical engineering; Biomedical imaging; Data mining; Educational institutions; Image analysis; Image segmentation; Industrial electronics; Information analysis; Image segmentation; Medical image; Multi-resolution analysis; Multiwavelet transform;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Automation and Logistics, 2008. ICAL 2008. IEEE International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-1-4244-2502-0
Electronic_ISBN :
978-1-4244-2503-7
Type :
conf
DOI :
10.1109/ICAL.2008.4636555
Filename :
4636555
Link To Document :
بازگشت