DocumentCode
316660
Title
Texture analysis of mammograms using the two-dimensional Hurst operator
Author
Chandrasekhar, Ramachandran ; Attikiouzel, Yianni ; Desilva, Christopher J S
Author_Institution
Dept. of Biomed. Eng., Singapore Gen. Hosp., Singapore
Volume
1
fYear
1997
fDate
2-4 Jul 1997
Firstpage
97
Abstract
A new texture operator; the two-dimensional Hurst operator, H, is introduced in this paper. It is the two-dimensional extrapolation of an operator originally devised by H.E. Hurst (1951) to analyze time-series. If an image is visualized as a three-dimensional relief map, and the intensity taken to be the height above an x-y plane, this texture operator is useful in determining the relative jaggedness of different image regions. Initial results with mammograms reveal that this operator can discriminate between the predominantly smooth “fatty” and relatively rougher “fibroglandular” regions on mammograms, and has the potential for lesion detection
Keywords
X-ray applications; diagnostic radiography; extrapolation; image segmentation; image texture; mathematical operators; medical image processing; time series; X-ray images; fatty regions; fibroglandular regions; image intensity; image regions; lesion detection; mammograms; relative jaggedness; texture analysis; texture operator; three-dimensional relief map; time-series analysis; two-dimensional Hurst operator; two-dimensional extrapolation; Extrapolation; Fractals; Hospitals; Image processing; Image segmentation; Intelligent systems; Lesions; Tiles; Visualization; X-ray imaging;
fLanguage
English
Publisher
ieee
Conference_Titel
Digital Signal Processing Proceedings, 1997. DSP 97., 1997 13th International Conference on
Conference_Location
Santorini
Print_ISBN
0-7803-4137-6
Type
conf
DOI
10.1109/ICDSP.1997.627977
Filename
627977
Link To Document