DocumentCode :
333408
Title :
Digital mammographic image analysis and data compression
Author :
Lin, Hong-Dun ; Lin, Kang-Ping ; Lou, Shyh-Liang
Author_Institution :
Dept. of Electr. Eng., Chung Yuan Christian Univ., Chung Li, Taiwan
Volume :
2
fYear :
1998
fDate :
29 Oct-1 Nov 1998
Firstpage :
1025
Abstract :
This paper proposes a digital mammogram segmentation method for mammographic images, which is based on the gradient detection between breast tissue and non-tissue area. Detection of tissue edge in medical images is closely related to the correctness of image segmentation. In this paper, we focus on digital mammograms and develop an edge detection method to extract breast tissue against background with noise and/or random artifacts, as a preprocess of image compression. The results show that the developed method performs a precise recovery of boundary between breast tissue and non-tissue area, which is then removed for increasing compression rate. Therefore, problems of image transmission efficiency and data storage amount can be greatly improved. The compression rate is promoted from 3:1 to 5-6:1 by a nonlossy compression method
Keywords :
data compression; edge detection; gradient methods; image coding; image segmentation; mammography; medical image processing; B-spline curve; background with noise; breast tissue; compression rate; data compression; data storage amount; digital mammographic image analysis; downhill reference points; gradient detection; image compression preprocess; image segmentation correctness; image transmission efficiency; mammogram segmentation method; nonlossy compression method; precise recovery of boundary; random artifacts; smoothing; tissue edge detection; Background noise; Biomedical imaging; Breast tissue; Data compression; Image analysis; Image coding; Image communication; Image edge detection; Image segmentation; Memory;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE
Conference_Location :
Hong Kong
ISSN :
1094-687X
Print_ISBN :
0-7803-5164-9
Type :
conf
DOI :
10.1109/IEMBS.1998.745624
Filename :
745624
Link To Document :
بازگشت