GENERALIZING GABOR FILTERS BASED ON ATEB-FUNCTIONS
Open Access
- 1 January 2019
- journal article
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- Vol. 4 (4), 72-84
- https://doi.org/10.28925/2663-4023.2019.4.7284
Abstract
Image filtering attempts to achieve greater resolution. There is a large number of filters that allows you to bring images with clear borders. In addition, noise is present when digitizing images. One of the most common types of filtering is the Gabor filter. It allows you to restore the image with the contour allocation at a certain frequency. Its core looks like elements of the Fourier basis, which is multiplied by Gaussian. The widespread use of Gabor filters for filtration is due to the fact that it gives a strong response at those points of the image where there is a component with local features of frequency in space and orientation. It is proposed to use the Ateb-Gabor filter, which greatly expands the well-known Gabor filter. The Ateb-Gabor filter combines all the properties of a harmonic function, which is multiplied by Gaussian. As a harmonic function, it is proposed to use the Ateb-functions that greatly extend the trigonometric effect. The developed filter is applied to the images. The Ateb-Gabor filter depends on the frequency and directions of the quasiperiodic structure of the image. Usually, to simplify the task, the average image frequency is calculated. It is unchanged at every point. Filtration of images is based on the generalized Ateb-Gabor filter. Influence of filtering parameters on images is investigated. The properties of periodic Ateb-functions are investigated. The value of the period from which the filtering results depend on is calculated. Ateb-Gabor filtering allowed for wider results than the classic Gabor filter. The one-dimensional Gabor filter based on the Ateb-functions gives the possibility to obtain more lenient or more convex forms of function at the maximum described in this study. In this way, filtration with a large spectrum of curves can be realized. This provides quick identification, since a more versatile kind of filtering has been developed.Keywords
This publication has 14 references indexed in Scilit:
- Threshold changeable secret image sharing scheme based on interpolation polynomialMultimedia Tools and Applications, 2019
- SimBaACM Journal of Experimental Algorithmics, 2019
- Ensemble Kalman filter-based data assimilation for three-dimensional multi-phase-field model: Estimation of anisotropic grain boundary propertiesMaterials & Design, 2019
- Application perfected wave tracing algorithmPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- Editing raster images and digital rating with softwarePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Functional luminal imaging probe topography: an improved method for characterizing esophageal distensibility in eosinophilic esophagitisTherapeutic Advances in Gastroenterology, 2012
- Geodesic active contoursPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Texram: a smart memory for texturingIEEE Computer Graphics and Applications, 1996
- Fingerprint enhancement by directional Fourier filteringIEE Proceedings - Vision, Image, and Signal Processing, 1994
- Wavelet transform domain filters: a spatially selective noise filtration techniqueIEEE Transactions on Image Processing, 1994