Open AccessApplying logarithmic density function to develop a segmentation model for intensity and noisy inhomogeneity images: towards informed decisions in computational and applied mathematics
Author(s) -
Shahed Ali Hamil,
Abdulrahman H. Majeed
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1362/1/012135
Subject(s) - logarithm , segmentation , artificial intelligence , intensity (physics) , noise (video) , function (biology) , computer science , image segmentation , homogeneity (statistics) , image (mathematics) , active contour model , noisy data , computer vision , mathematics , pattern recognition (psychology) , algorithm , machine learning , mathematical analysis , optics , physics , evolutionary biology , biology
This study examined the how a logarithmic intensity function could be used to develop a segmentation model for intensity and noisy homogeneity images. Whereas inhomogeneous images demand the use of local image data, it remains notable that it remains defective for noisy images. The eventuality is that active contour motions tend to be misguided by the local information. In the proposed model, the logarithmic function was able to capture minute details contained in selected images and also counter or ignore the perceived noise. As such, the model proved effective and robust, worth applying to such images. To verify the effectiveness of the model, its results were compared to the experimental outcomes previously reported after employing local Chan-Vese Model. Indeed, the proposed model exhibited superior performance in relation to the treatment of intensity and noisy inhomogeneity images.