Texture Representation via Joint Statistics of Local Quantized Patterns

Tiecheng SONG; Linfeng XU; Chao HUANG; Bing LUO

doi:10.1587/transinf.E97.D.155

Texture Representation via Joint Statistics of Local Quantized Patterns

Tiecheng SONG, Linfeng XU, Chao HUANG, Bing LUO

Full Text Views

0

Cite this

Summary :

In this paper, a simple yet efficient texture representation is proposed for texture classification by exploring the joint statistics of local quantized patterns (jsLQP). In order to combine information of different domains, the Gaussian derivative filters are first employed to obtain the multi-scale gradient responses. Then, three feature maps are generated by encoding the local quantized binary and ternary patterns in the image space and the gradient space. Finally, these feature maps are hybridly encoded, and their joint histogram is used as the final texture representation. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art LBP based and even learning based methods for texture classification.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.1 pp.155-159

Publication Date: 2014/01/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E97.D.155

Type of Manuscript: LETTER

Category: Image Recognition, Computer Vision

Authors

Tiecheng SONG
  University of Electronic Science and Technology of China
Linfeng XU
  University of Electronic Science and Technology of China
Chao HUANG
  University of Electronic Science and Technology of China
Bing LUO
  University of Electronic Science and Technology of China

Keyword

texture classification, Local Binary Patterns (LBP), Gaussian derivative filter, ternary coding

Cite this

Copy

Tiecheng SONG, Linfeng XU, Chao HUANG, Bing LUO, "Texture Representation via Joint Statistics of Local Quantized Patterns" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 1, pp. 155-159, January 2014, doi: 10.1587/transinf.E97.D.155.
Abstract: In this paper, a simple yet efficient texture representation is proposed for texture classification by exploring the joint statistics of local quantized patterns (jsLQP). In order to combine information of different domains, the Gaussian derivative filters are first employed to obtain the multi-scale gradient responses. Then, three feature maps are generated by encoding the local quantized binary and ternary patterns in the image space and the gradient space. Finally, these feature maps are hybridly encoded, and their joint histogram is used as the final texture representation. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art LBP based and even learning based methods for texture classification.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E97.D.155/_p

Copy

@ARTICLE{e97-d_1_155,
author={Tiecheng SONG, Linfeng XU, Chao HUANG, Bing LUO, },
journal={IEICE TRANSACTIONS on Information},
title={Texture Representation via Joint Statistics of Local Quantized Patterns},
year={2014},
volume={E97-D},
number={1},
pages={155-159},
abstract={In this paper, a simple yet efficient texture representation is proposed for texture classification by exploring the joint statistics of local quantized patterns (jsLQP). In order to combine information of different domains, the Gaussian derivative filters are first employed to obtain the multi-scale gradient responses. Then, three feature maps are generated by encoding the local quantized binary and ternary patterns in the image space and the gradient space. Finally, these feature maps are hybridly encoded, and their joint histogram is used as the final texture representation. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art LBP based and even learning based methods for texture classification.},
keywords={},
doi={10.1587/transinf.E97.D.155},
ISSN={1745-1361},
month={January},}

Copy

TY - JOUR
TI - Texture Representation via Joint Statistics of Local Quantized Patterns
T2 - IEICE TRANSACTIONS on Information
SP - 155
EP - 159
AU - Tiecheng SONG
AU - Linfeng XU
AU - Chao HUANG
AU - Bing LUO
PY - 2014
DO - 10.1587/transinf.E97.D.155
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2014
AB - In this paper, a simple yet efficient texture representation is proposed for texture classification by exploring the joint statistics of local quantized patterns (jsLQP). In order to combine information of different domains, the Gaussian derivative filters are first employed to obtain the multi-scale gradient responses. Then, three feature maps are generated by encoding the local quantized binary and ternary patterns in the image space and the gradient space. Finally, these feature maps are hybridly encoded, and their joint histogram is used as the final texture representation. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art LBP based and even learning based methods for texture classification.
ER -