New local binary pattern approaches based on color channels in texture classification
- PDF / 2,814,763 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 29 Downloads / 293 Views
New local binary pattern approaches based on color channels in texture classification Ramazan Tekin 1
2
& Ömer Faruk Ertuğrul & Yılmaz Kaya
3
Received: 20 February 2020 / Revised: 12 August 2020 / Accepted: 21 August 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
In this paper, four novel, simple and robust approaches, which are left to right local binary patterns (LBPLL2R), top to down local binary patterns (LBPT2D), cube surface local binary pattern (LBPSurfaces), and cube diagonal local binary pattern (LBPDiagonal), were proposed in order to exact texture features in color images. These approaches were based on the local binary pattern (LBP), which is an effective statistical texture descriptor and can be employed in gray images. Proposed approaches were evaluated and validated in four datasets, which are Outex, KTH_TIPS, KTH_TIPS2, and USPtex datasets. The images in these datasets are in RGB, HSV, YIQ, and YCbCr color formats. Achieved results by these approaches were compared with the obtained results by the classical LBP and literature findings. As a result, the proposed approaches performed better than the traditional LBP method and they found effective in the classification of color texture images, especially in images, which are in RGB and HSV formats. Furthermore, noise robustness and time complexity of the proposed approaches were validated. Keywords Local binary patterns . Texture classification . Feature extraction . Image classification
1 Introduction Texture analysis methods have been employed to detect, exact or identify the characteristical properties of images [3]. Textures can be described as a summation of the brightness, color, slope, size, shape, and some other physical properties of an object or a group of objects in an image. Since textures give detailed information about an image, they can be employed to understand an
* Ramazan Tekin [email protected]
1
Department of Computer Engineering, Batman University, 72060 Batman, Turkey
2
Department of Electrical and Electronics Engineering, Batman University, 72060 Batman, Turkey
3
Department of Computer Engineering, Siirt University, 56100 Siirt, Turkey
Multimedia Tools and Applications
image, to determine the similarities/dissimilarities, and to classify images [12, 88]. Texture detection methods have been successfully employed in image processing issues such as remote sensing, medical processing, face, and fingerprint detection [41, 92]. Although many successful texture detection methods have been put forward, there are still some problems in assessing the real-world images such as having different luminance and scale of the images. Generally, texture detection methods are being processed according to the following stages: (1) preprocessing, (2) extracting features, and (3) machine learning stages [11, 53]. Texture extraction is generally done in images that are in gray formats. In order to detect the textures in color images, they have generally been converted to the gray images. Be
Data Loading...
