Abstract
In this paper, a novel scheme for multilevel detection and segmentation of suspicious lesions from mammograms is proposed. First, non-decimated wavelet transform (NDWT) of selective scale is applied to input mammograms. In coarse segmentation, NDWT sub-images go through decimated wavelet transform, and thereafter multiscale analysis of image probability distribution function is used. In order to improve the morphological feature of coarse segmented image, a morphological filter is applied to NDWT transformed sub-image and convolved with it. Fractional Fourier transform (FRFT) domain homomorphic filter is applied on convoluted image for improved fine segmentation. FRFT domains are used to enhance the intensity gradient by controlling the reflectance slope with the help of fractional order \(\upomega\). Second level local adaptive thresholding is used to obtain final segmented image. Efficiency of the proposed scheme is verified by simulating mammograms in the mammographic image analysis society (MIAS) database. The proposed algorithm obtains improved detection with 100% sensitivity and 0.667 false positive per image for CIRC and MISC lesions and comparable results for other masses in comparison to the nearest schemes.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
