Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection

Jiu Xu, Ning Jiang, Satoshi Goto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In this paper, a feature named Non-Redundant Gradient Semantic Local Binary Patterns (NRGSLBP) is proposed for pedestrian detection as a modified version of conventional Semantic Local Binary Patterns (SLBP). Calculations of this feature are carried out for both intensity and gradient magnitude image so that texture and gradient information are combined. Moreover, non-redundant patterns are adopted on SLBP for the first time, allowing better discrimination. Compared with SLBP, no additional cost of the feature dimensions NRGSLBP is necessary and the calculation complexity is considerably smaller than that of other features. Experimental results on several datasets show that the detection rate of our proposed feature outperforms those of other features such as Histogram of Orientated Gradient (HOG), Histogram of Templates (HOT), Bidirectional Local Template Patterns (BLTP), Gradient Local Binary Patterns (GLBP), SLBP and Covariance matrix (COV).

Original languageEnglish
Title of host publicationEuropean Signal Processing Conference
PublisherEuropean Signal Processing Conference, EUSIPCO
Pages1407-1411
Number of pages5
ISBN (Print)9780992862619
Publication statusPublished - 2014 Nov 10
Event22nd European Signal Processing Conference, EUSIPCO 2014 - Lisbon
Duration: 2014 Sep 12014 Sep 5

Other

Other22nd European Signal Processing Conference, EUSIPCO 2014
CityLisbon
Period14/9/114/9/5

Fingerprint

Semantics
Covariance matrix
Textures
Costs

Keywords

  • feature extraction
  • non-redundant gradient semantic local binary patterns
  • Pedestrian detection

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Xu, J., Jiang, N., & Goto, S. (2014). Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection. In European Signal Processing Conference (pp. 1407-1411). [6952501] European Signal Processing Conference, EUSIPCO.

Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection. / Xu, Jiu; Jiang, Ning; Goto, Satoshi.

European Signal Processing Conference. European Signal Processing Conference, EUSIPCO, 2014. p. 1407-1411 6952501.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xu, J, Jiang, N & Goto, S 2014, Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection. in European Signal Processing Conference., 6952501, European Signal Processing Conference, EUSIPCO, pp. 1407-1411, 22nd European Signal Processing Conference, EUSIPCO 2014, Lisbon, 14/9/1.
Xu J, Jiang N, Goto S. Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection. In European Signal Processing Conference. European Signal Processing Conference, EUSIPCO. 2014. p. 1407-1411. 6952501
Xu, Jiu ; Jiang, Ning ; Goto, Satoshi. / Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection. European Signal Processing Conference. European Signal Processing Conference, EUSIPCO, 2014. pp. 1407-1411
@inproceedings{72a11537edcc400aa3e1b14d81ce7095,
title = "Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection",
abstract = "In this paper, a feature named Non-Redundant Gradient Semantic Local Binary Patterns (NRGSLBP) is proposed for pedestrian detection as a modified version of conventional Semantic Local Binary Patterns (SLBP). Calculations of this feature are carried out for both intensity and gradient magnitude image so that texture and gradient information are combined. Moreover, non-redundant patterns are adopted on SLBP for the first time, allowing better discrimination. Compared with SLBP, no additional cost of the feature dimensions NRGSLBP is necessary and the calculation complexity is considerably smaller than that of other features. Experimental results on several datasets show that the detection rate of our proposed feature outperforms those of other features such as Histogram of Orientated Gradient (HOG), Histogram of Templates (HOT), Bidirectional Local Template Patterns (BLTP), Gradient Local Binary Patterns (GLBP), SLBP and Covariance matrix (COV).",
keywords = "feature extraction, non-redundant gradient semantic local binary patterns, Pedestrian detection",
author = "Jiu Xu and Ning Jiang and Satoshi Goto",
year = "2014",
month = "11",
day = "10",
language = "English",
isbn = "9780992862619",
pages = "1407--1411",
booktitle = "European Signal Processing Conference",
publisher = "European Signal Processing Conference, EUSIPCO",

}

TY - GEN

T1 - Non-Redundant Gradient Semantic Local Binary Patterns for pedestrian detection

AU - Xu, Jiu

AU - Jiang, Ning

AU - Goto, Satoshi

PY - 2014/11/10

Y1 - 2014/11/10

N2 - In this paper, a feature named Non-Redundant Gradient Semantic Local Binary Patterns (NRGSLBP) is proposed for pedestrian detection as a modified version of conventional Semantic Local Binary Patterns (SLBP). Calculations of this feature are carried out for both intensity and gradient magnitude image so that texture and gradient information are combined. Moreover, non-redundant patterns are adopted on SLBP for the first time, allowing better discrimination. Compared with SLBP, no additional cost of the feature dimensions NRGSLBP is necessary and the calculation complexity is considerably smaller than that of other features. Experimental results on several datasets show that the detection rate of our proposed feature outperforms those of other features such as Histogram of Orientated Gradient (HOG), Histogram of Templates (HOT), Bidirectional Local Template Patterns (BLTP), Gradient Local Binary Patterns (GLBP), SLBP and Covariance matrix (COV).

AB - In this paper, a feature named Non-Redundant Gradient Semantic Local Binary Patterns (NRGSLBP) is proposed for pedestrian detection as a modified version of conventional Semantic Local Binary Patterns (SLBP). Calculations of this feature are carried out for both intensity and gradient magnitude image so that texture and gradient information are combined. Moreover, non-redundant patterns are adopted on SLBP for the first time, allowing better discrimination. Compared with SLBP, no additional cost of the feature dimensions NRGSLBP is necessary and the calculation complexity is considerably smaller than that of other features. Experimental results on several datasets show that the detection rate of our proposed feature outperforms those of other features such as Histogram of Orientated Gradient (HOG), Histogram of Templates (HOT), Bidirectional Local Template Patterns (BLTP), Gradient Local Binary Patterns (GLBP), SLBP and Covariance matrix (COV).

KW - feature extraction

KW - non-redundant gradient semantic local binary patterns

KW - Pedestrian detection

UR - http://www.scopus.com/inward/record.url?scp=84911928230&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84911928230&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84911928230

SN - 9780992862619

SP - 1407

EP - 1411

BT - European Signal Processing Conference

PB - European Signal Processing Conference, EUSIPCO

ER -