Image feature fusion for human detection with multi-sensor based on FHOG

Meng Wang; Ya Ping Dai

doi:10.15918/j.tbit1001-0645.2015.02.016

Image feature fusion for human detection with multi-sensor based on FHOG

Meng Wang, Ya Ping Dai

School of Automation

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

Proposed a novel image feature fusion approach based on histogram of oriented gradient (HOG). The visual activation measure (VAM) was used to select the statistics of local gradients with significant direction, and forms fused histogram of gradient (FHOG), which effectively solved the existing deficiencies of multi-resolution (MR) image fusion. These fused features were plugged into support vector machine (SVM), and train human/background binary classifier for human detection. Experiments show that compared with the traditional MR image fusion approaches, in reference points the missing rate of the proposed approach decreases by 3%~10%, and the false alarm rate drops by an average of more than 20%.

Original language	English
Pages (from-to)	192-196 and 202
Journal	Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology
Volume	35
Issue number	2
DOIs	https://doi.org/10.15918/j.tbit1001-0645.2015.02.016
Publication status	Published - 1 Feb 2015

Keywords

Fused histogram of oriented gradient (FHOG)
Human detection
Image feature fusion
Visual active measurement (VAM)

Access to Document

10.15918/j.tbit1001-0645.2015.02.016

Cite this

Wang, M., & Dai, Y. P. (2015). Image feature fusion for human detection with multi-sensor based on FHOG. Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology, 35(2), 192-196 and 202. https://doi.org/10.15918/j.tbit1001-0645.2015.02.016

@article{7984e1766b7e45878aa2374808b1675f,

title = "Image feature fusion for human detection with multi-sensor based on FHOG",

abstract = "Proposed a novel image feature fusion approach based on histogram of oriented gradient (HOG). The visual activation measure (VAM) was used to select the statistics of local gradients with significant direction, and forms fused histogram of gradient (FHOG), which effectively solved the existing deficiencies of multi-resolution (MR) image fusion. These fused features were plugged into support vector machine (SVM), and train human/background binary classifier for human detection. Experiments show that compared with the traditional MR image fusion approaches, in reference points the missing rate of the proposed approach decreases by 3%~10%, and the false alarm rate drops by an average of more than 20%.",

keywords = "Fused histogram of oriented gradient (FHOG), Human detection, Image feature fusion, Visual active measurement (VAM)",

author = "Meng Wang and Dai, {Ya Ping}",

year = "2015",

month = feb,

day = "1",

doi = "10.15918/j.tbit1001-0645.2015.02.016",

language = "English",

volume = "35",

pages = "192--196 and 202",

journal = "Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology",

issn = "1001-0645",

publisher = "Beijing Institute of Technology",

number = "2",

}

TY - JOUR

T1 - Image feature fusion for human detection with multi-sensor based on FHOG

AU - Wang, Meng

AU - Dai, Ya Ping

PY - 2015/2/1

Y1 - 2015/2/1

N2 - Proposed a novel image feature fusion approach based on histogram of oriented gradient (HOG). The visual activation measure (VAM) was used to select the statistics of local gradients with significant direction, and forms fused histogram of gradient (FHOG), which effectively solved the existing deficiencies of multi-resolution (MR) image fusion. These fused features were plugged into support vector machine (SVM), and train human/background binary classifier for human detection. Experiments show that compared with the traditional MR image fusion approaches, in reference points the missing rate of the proposed approach decreases by 3%~10%, and the false alarm rate drops by an average of more than 20%.

AB - Proposed a novel image feature fusion approach based on histogram of oriented gradient (HOG). The visual activation measure (VAM) was used to select the statistics of local gradients with significant direction, and forms fused histogram of gradient (FHOG), which effectively solved the existing deficiencies of multi-resolution (MR) image fusion. These fused features were plugged into support vector machine (SVM), and train human/background binary classifier for human detection. Experiments show that compared with the traditional MR image fusion approaches, in reference points the missing rate of the proposed approach decreases by 3%~10%, and the false alarm rate drops by an average of more than 20%.

KW - Fused histogram of oriented gradient (FHOG)

KW - Human detection

KW - Image feature fusion

KW - Visual active measurement (VAM)

UR - http://www.scopus.com/inward/record.url?scp=84927138708&partnerID=8YFLogxK

U2 - 10.15918/j.tbit1001-0645.2015.02.016

DO - 10.15918/j.tbit1001-0645.2015.02.016

M3 - Article

AN - SCOPUS:84927138708

SN - 1001-0645

VL - 35

SP - 192-196 and 202

JO - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

JF - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

IS - 2

ER -

Image feature fusion for human detection with multi-sensor based on FHOG

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this