Building recognition on subregion’s multi-scale gist feature extraction and corresponding columns information based dimensionality reduction

Bin Li, Wei Pang, Yuhao Liu, Xiangchun Yu, Zhezhou Yu (Corresponding Author)

Research output: Contribution to journalArticle

1 Citation (Scopus)
3 Downloads (Pure)

Abstract

In this paper, we proposed a new building recognition method named subregion’s multiscale gist feature (SM-gist) extraction and corresponding columns information based dimensionality reduction (CCI-DR). Our proposed building recognition method is presented as a two-stage model: in the first stage, a building image is divided into 4 × 5 subregions, and gist vectors are extracted from these regions individually. Then, we combine these gist vectors into a matrix with relatively high dimensions. In the second stage, we proposed CCI-DR to project the high dimensional manifold matrix to low dimensional subspace. Compared with the previous building recognition method the advantages of our proposed method are that (1) gist features extracted by SM-gist have the ability to adapt to nonuniform illumination and that (2) CCI-DR can address the limitation of traditional dimensionality reduction methods, which convert gist matrices into vectors and thus mix the corresponding gist vectors from different feature maps. Our building recognition method is evaluated on the Sheffield buildings database, and experiments show that our method can achieve satisfactory performance.
Original languageEnglish
Article number898705
Number of pages10
JournalJournal of Applied Mathematics
Volume2014
DOIs
Publication statusPublished - 27 Apr 2014

Fingerprint

Dimensionality Reduction
Feature Extraction
Feature extraction
Two-stage Model
Lighting
Reduction Method
Higher Dimensions
Convert
Illumination
High-dimensional
Subspace
Experiments
Experiment

Cite this

Building recognition on subregion’s multi-scale gist feature extraction and corresponding columns information based dimensionality reduction. / Li, Bin; Pang, Wei; Liu, Yuhao; Yu, Xiangchun; Yu, Zhezhou (Corresponding Author).

In: Journal of Applied Mathematics, Vol. 2014, 898705, 27.04.2014.

Research output: Contribution to journalArticle

@article{61e2281d305345e0ba033093640e38af,
title = "Building recognition on subregion’s multi-scale gist feature extraction and corresponding columns information based dimensionality reduction",
abstract = "In this paper, we proposed a new building recognition method named subregion’s multiscale gist feature (SM-gist) extraction and corresponding columns information based dimensionality reduction (CCI-DR). Our proposed building recognition method is presented as a two-stage model: in the first stage, a building image is divided into 4 × 5 subregions, and gist vectors are extracted from these regions individually. Then, we combine these gist vectors into a matrix with relatively high dimensions. In the second stage, we proposed CCI-DR to project the high dimensional manifold matrix to low dimensional subspace. Compared with the previous building recognition method the advantages of our proposed method are that (1) gist features extracted by SM-gist have the ability to adapt to nonuniform illumination and that (2) CCI-DR can address the limitation of traditional dimensionality reduction methods, which convert gist matrices into vectors and thus mix the corresponding gist vectors from different feature maps. Our building recognition method is evaluated on the Sheffield buildings database, and experiments show that our method can achieve satisfactory performance.",
author = "Bin Li and Wei Pang and Yuhao Liu and Xiangchun Yu and Zhezhou Yu",
year = "2014",
month = "4",
day = "27",
doi = "10.1155/2014/898705",
language = "English",
volume = "2014",
journal = "Journal of Applied Mathematics",
issn = "1110-757X",
publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Building recognition on subregion’s multi-scale gist feature extraction and corresponding columns information based dimensionality reduction

AU - Li, Bin

AU - Pang, Wei

AU - Liu, Yuhao

AU - Yu, Xiangchun

AU - Yu, Zhezhou

PY - 2014/4/27

Y1 - 2014/4/27

N2 - In this paper, we proposed a new building recognition method named subregion’s multiscale gist feature (SM-gist) extraction and corresponding columns information based dimensionality reduction (CCI-DR). Our proposed building recognition method is presented as a two-stage model: in the first stage, a building image is divided into 4 × 5 subregions, and gist vectors are extracted from these regions individually. Then, we combine these gist vectors into a matrix with relatively high dimensions. In the second stage, we proposed CCI-DR to project the high dimensional manifold matrix to low dimensional subspace. Compared with the previous building recognition method the advantages of our proposed method are that (1) gist features extracted by SM-gist have the ability to adapt to nonuniform illumination and that (2) CCI-DR can address the limitation of traditional dimensionality reduction methods, which convert gist matrices into vectors and thus mix the corresponding gist vectors from different feature maps. Our building recognition method is evaluated on the Sheffield buildings database, and experiments show that our method can achieve satisfactory performance.

AB - In this paper, we proposed a new building recognition method named subregion’s multiscale gist feature (SM-gist) extraction and corresponding columns information based dimensionality reduction (CCI-DR). Our proposed building recognition method is presented as a two-stage model: in the first stage, a building image is divided into 4 × 5 subregions, and gist vectors are extracted from these regions individually. Then, we combine these gist vectors into a matrix with relatively high dimensions. In the second stage, we proposed CCI-DR to project the high dimensional manifold matrix to low dimensional subspace. Compared with the previous building recognition method the advantages of our proposed method are that (1) gist features extracted by SM-gist have the ability to adapt to nonuniform illumination and that (2) CCI-DR can address the limitation of traditional dimensionality reduction methods, which convert gist matrices into vectors and thus mix the corresponding gist vectors from different feature maps. Our building recognition method is evaluated on the Sheffield buildings database, and experiments show that our method can achieve satisfactory performance.

U2 - 10.1155/2014/898705

DO - 10.1155/2014/898705

M3 - Article

VL - 2014

JO - Journal of Applied Mathematics

JF - Journal of Applied Mathematics

SN - 1110-757X

M1 - 898705

ER -