Tumor detection in digitized mammograms by image texture analysis

Richard Pfisterer; Farzin Aghdasi

doi:10.1117/1.1338998

1 February 2001 Tumor detection in digitized mammograms by image texture analysis

Richard Pfisterer, Farzin Aghdasi

Optical Engineering, Vol. 40, Issue 2, (February 2001). https://doi.org/10.1117/1.1338998

Four schemes to detect masses in digitized mammograms based on textural information are demonstrated and compared. Enhancement and segmentation of the image is performed using hexagonal wavelets, steerable filters, and Laws texture energy maps. Hexagonal wavelets are used because spectral orientation is partitioned into three bands covering the frequency domain equally. The lifting scheme is used to construct these wavelets, because it can be extended to uneven sampling, is not susceptible to boundary conditions, and is computationally efficient. It is found that with Laws, the specific convolutional mask is not as important as the pre-enhancement used. Steered filter enhancement shows the best results, closely followed by adaptive histogram equalization, with wavelet enhancement performing the poorest. Oriented filters are used for enhancement and orientation detection. A set of filters are designed that can be efficiently steered along the dominant orientation of the image.

©(2001) Society of Photo-Optical Instrumentation Engineers (SPIE)

Citation Download Citation

Richard Pfisterer and Farzin Aghdasi "Tumor detection in digitized mammograms by image texture analysis," Optical Engineering 40(2), (1 February 2001). https://doi.org/10.1117/1.1338998

Published: 1 February 2001

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available