Reconstruction-based change detection with image completion for a free-moving camera

研究成果: ジャーナルへの寄稿記事

1 引用 (Scopus)

抄録

Reconstruction-based change detection methods are robust for camera motion. The methods learn reconstruction of input images based on background images. Foreground regions are detected based on the magnitude of the difference between an input image and a reconstructed input image. For learning, only background images are used. Therefore, foreground regions have larger differences than background regions. Traditional reconstruction-based methods have two problems. One is over-reconstruction of foreground regions. The other is that decision of change detection depends on magnitudes of differences only. It is difficult to distinguish magnitudes of differences in foreground regions when the foreground regions are completely reconstructed in patch images. We propose the framework of a reconstruction-based change detection method for a free-moving camera using patch images. To avoid over-reconstruction of foreground regions, our method reconstructs a masked central region in a patch image from a region surrounding the central region. Differences in foreground regions are enhanced because foreground regions in patch images are removed by the masking procedure. Change detection is learned from a patch image and a reconstructed image automatically. The decision procedure directly uses patch images rather than the differences between patch images. Our method achieves better accuracy compared to traditional reconstruction-based methods without masking patch images.

元の言語英語
記事番号1232
ジャーナルSensors (Switzerland)
18
発行部数4
DOI
出版物ステータス出版済み - 4 17 2018

Fingerprint

change detection
Cameras
cameras
masking
Computer-Assisted Image Processing
Learning
learning

All Science Journal Classification (ASJC) codes

  • Analytical Chemistry
  • Atomic and Molecular Physics, and Optics
  • Biochemistry
  • Instrumentation
  • Electrical and Electronic Engineering

これを引用

@article{c12fec9ade1144eeaf57c811fd6009c7,
title = "Reconstruction-based change detection with image completion for a free-moving camera",
abstract = "Reconstruction-based change detection methods are robust for camera motion. The methods learn reconstruction of input images based on background images. Foreground regions are detected based on the magnitude of the difference between an input image and a reconstructed input image. For learning, only background images are used. Therefore, foreground regions have larger differences than background regions. Traditional reconstruction-based methods have two problems. One is over-reconstruction of foreground regions. The other is that decision of change detection depends on magnitudes of differences only. It is difficult to distinguish magnitudes of differences in foreground regions when the foreground regions are completely reconstructed in patch images. We propose the framework of a reconstruction-based change detection method for a free-moving camera using patch images. To avoid over-reconstruction of foreground regions, our method reconstructs a masked central region in a patch image from a region surrounding the central region. Differences in foreground regions are enhanced because foreground regions in patch images are removed by the masking procedure. Change detection is learned from a patch image and a reconstructed image automatically. The decision procedure directly uses patch images rather than the differences between patch images. Our method achieves better accuracy compared to traditional reconstruction-based methods without masking patch images.",
author = "Tsubasa Minematsu and Atsushi Shimada and Hideaki Uchiyama and Vincent Charvillat and Rin-Ichiro Taniguchi",
year = "2018",
month = "4",
day = "17",
doi = "10.3390/s18041232",
language = "English",
volume = "18",
journal = "Sensors",
issn = "1424-3210",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "4",

}

TY - JOUR

T1 - Reconstruction-based change detection with image completion for a free-moving camera

AU - Minematsu, Tsubasa

AU - Shimada, Atsushi

AU - Uchiyama, Hideaki

AU - Charvillat, Vincent

AU - Taniguchi, Rin-Ichiro

PY - 2018/4/17

Y1 - 2018/4/17

N2 - Reconstruction-based change detection methods are robust for camera motion. The methods learn reconstruction of input images based on background images. Foreground regions are detected based on the magnitude of the difference between an input image and a reconstructed input image. For learning, only background images are used. Therefore, foreground regions have larger differences than background regions. Traditional reconstruction-based methods have two problems. One is over-reconstruction of foreground regions. The other is that decision of change detection depends on magnitudes of differences only. It is difficult to distinguish magnitudes of differences in foreground regions when the foreground regions are completely reconstructed in patch images. We propose the framework of a reconstruction-based change detection method for a free-moving camera using patch images. To avoid over-reconstruction of foreground regions, our method reconstructs a masked central region in a patch image from a region surrounding the central region. Differences in foreground regions are enhanced because foreground regions in patch images are removed by the masking procedure. Change detection is learned from a patch image and a reconstructed image automatically. The decision procedure directly uses patch images rather than the differences between patch images. Our method achieves better accuracy compared to traditional reconstruction-based methods without masking patch images.

AB - Reconstruction-based change detection methods are robust for camera motion. The methods learn reconstruction of input images based on background images. Foreground regions are detected based on the magnitude of the difference between an input image and a reconstructed input image. For learning, only background images are used. Therefore, foreground regions have larger differences than background regions. Traditional reconstruction-based methods have two problems. One is over-reconstruction of foreground regions. The other is that decision of change detection depends on magnitudes of differences only. It is difficult to distinguish magnitudes of differences in foreground regions when the foreground regions are completely reconstructed in patch images. We propose the framework of a reconstruction-based change detection method for a free-moving camera using patch images. To avoid over-reconstruction of foreground regions, our method reconstructs a masked central region in a patch image from a region surrounding the central region. Differences in foreground regions are enhanced because foreground regions in patch images are removed by the masking procedure. Change detection is learned from a patch image and a reconstructed image automatically. The decision procedure directly uses patch images rather than the differences between patch images. Our method achieves better accuracy compared to traditional reconstruction-based methods without masking patch images.

UR - http://www.scopus.com/inward/record.url?scp=85045616582&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045616582&partnerID=8YFLogxK

U2 - 10.3390/s18041232

DO - 10.3390/s18041232

M3 - Article

VL - 18

JO - Sensors

JF - Sensors

SN - 1424-3210

IS - 4

M1 - 1232

ER -