Texture descriptor based on local combination adaptive ternary pattern

Abstract

Material recognition has several applications, such as image retrieval, object recognition and robotic manipulation. To make the material classification more suitable for real-world applications, it is fundamental to satisfy two characteristics: robustness to scale and to pose variations. In this study, the authors propose a novel discriminant descriptor for texture classification based on a new operator called local combination adaptive ternary pattern (LCATP) descriptor used to encode both colour and local information. They start by building the LCATP descriptor using a combination of three different adaptive thresholding techniques. Moreover, they present a novel operator, mean histogram (MH), used jointly with the LCATP in order to incorporate colour information into the descriptor. This approach is then extended to four different colour spaces: LC 1 C 2 , I 1 I 2 I 3 , LSH uv and O 1 O 2 O 3 . The final descriptor, LCATP fusion (LCATP_F), is produced by fusing the basic histogram (H) and MH extracted from the different colour spaces. Finally, the LCATP_F descriptor properties, such as the robustness to scale and pose changes are evaluated using the challenging KTH-textures under varying illumination, pose and scale (TIPS2b) dataset along with the least squares support vector machines classifier. The obtained experimental results, using the LCATP_F descriptor, show a significant improvement with respect to the state-of-the-art results.

New content‐based image retrieval system based on optimised integration of DCD, wavelet and curvelet features

“…In DCD, the colour space is divided into partitions known as course partitions. Each partition has two main components, including the partition centre and percentage which are computed as [4]: [2] block colour histog co-occurrence matrix -Dubey et al [3] LCOD --Wang et al [4] DCD steerable filter decomposition Zernike moments Singha et al [6] colour histogram wavelet transform -Sandid et al [8] local ternary pattern -Farsi et al [10] wavelet transform -Gonde et al [12] curvelet transform -Talib et al [14] weighted DCD --Chun et al [20] -BDIP and BVLC of wavelet coefficients -Yildizer et al [21] wavelet transform -Wang et al [25] Zernike chromaticity distribution moments contourlet transform -ElAlami [26] CCV Gabor filters -Jhanwar et al [27] -Motif co-occurrence matrix -Lin et al [28] colour histogram for K-mean (CHKM) Motif co-occurrence matrix and Difference between pixels of scan pattern (DBPSP) -Dubey et al [22] local wavelet pattern -Murala et al [24] -LTrP -Dubey et al [13] RSHD as colour and texture features -In these equations, C i and P i are the partition centre and its…”

“…Texture is another important property of an image. Some basic primitive properties of image including the structural arrangement of its regions and the relationship between the surrounding regions can be obtained from texture features [8,9]. Texture features are conventionally divided into four different types including statistical, structural, model-based and signal processing-based features.…”

Texture Feature Extraction Methods: A Survey

“…The sensitivity to noise has also been the subject of several works such as [17] and [90] where, among others, the local ternary patterns (LTP) is proposed to overcome the high sensitivity to noise in near-uniform regions that is present in LBP, and to introduce a higher level of granularity. A local combination adaptive ternary pattern (LCATP) descriptor has also recently been proposed to encode both colour and local information [91]. A variant of LBP to obtain texture descriptors insensitive to blur is also available [92].…”

Dynamic texture classification using Gumbel mixtures in the complex wavelet domain

“…Texture is an important image property for image analysis and understanding [1, 2]. Dynamic texture (DT) is the texture with motion, which is the extension of image texture to the temporal domain.…”