Open AccessEdge Detection Algorithm Based on BEMD for Liver CT Images
Author(s) -
Gajendra Kumar Mourya,
Manashjit Gogoi,
Akash Handique
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.c8930.019320
Subject(s) - hilbert–huang transform , sobel operator , artificial intelligence , canny edge detector , pattern recognition (psychology) , edge detection , computer science , fourier transform , image (mathematics) , algorithm , computer vision , level set (data structures) , mathematics , image processing , mathematical analysis , filter (signal processing)
Liver edge identification requires for its volume estimation from CT image and this process is a prerequisite for liver diagnosis and treatment planning. In this article, an edge detection algorithm proposed based on Bi-dimensional Empirical Mode Decomposition (BEMD) and Fourier Transform. Intrinsic mode function (IMF) extracted from BEMD and mixed with the Fourier phase of the original image to get edge profile. The proposed method extensively evaluated on Berkeley Segmentation Data Set (BSDS-500) and compared with Sobel and Canny operators. Results achieved Mean Square error 0.04±0.01 and PSNR 62.27±1.1. In conclusion, The BEMD approach capable of identifying image edges with high accuracy compared with state of the art.