Hexagonal wavelets for the detection of masses in digitized mammograms
Richard Pfisterer, Farzin Aghdasi · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1999
Wavelets for the detection of masses in digitized mammograms are proposed in this study. Enhancement and segmentation of the image, based on texture, is performed using wavelets and Laws texture maps. Hexagonal wavelets are proposed in this study for improved detection and speed. In hexagonal sampling, orientations are partitioned into three bands of 60 degrees, equally covering the frequency domain such that no particular orientation is favored. The lifting scheme is used to construct the wavelets. This scheme has several advantages over the conventional Fourier approach. Most importantly the lifting scheme can easily be extended to uneven sampling, is not susceptible to boundary conditions, and is also computationally more efficient by a factor of two. One method uses wavelets to enhance the image before the Laws texture energy map is found. The second segments the image based on the wavelet coefficients for low texture. Both of the methods show promising results.