Shadow detection for mobile robots:: Features, evaluation, and datasets
Shadow detection for mobile robots:: Features, evaluation, and datasets
Shadows have long been a challenging topic for computer vision. This challenge is made even harder when we assume that the camera is moving, as many existing shadow detection techniques require the creation and maintenance of a background model. This article explores the problem of shadow modelling from a moving viewpoint (assumed to be a robotic platform) through comparing shadow-variant and shadow-invariant image features — primarily color, texture and edge-based features. These features are then embedded in a segmentation pipeline that provides predictions on shadow status, using minimal temporal context. We also release a public dataset of shadow-related image sequences, to help other researchers further develop shadow detection methods and to enable benchmarking of techniques.
1-23
Newey, Charles C.
f92bf8ff-1cfd-4db1-b58d-6ec36d76f7fa
Jones, Owain D.
19371ba6-7d52-4eae-8c01-6a17606e3cef
Dee, Hannah M.
3fddc966-dd20-4a68-9dfb-76992d0cb934
Newey, Charles C.
f92bf8ff-1cfd-4db1-b58d-6ec36d76f7fa
Jones, Owain D.
19371ba6-7d52-4eae-8c01-6a17606e3cef
Dee, Hannah M.
3fddc966-dd20-4a68-9dfb-76992d0cb934
Newey, Charles C., Jones, Owain D. and Dee, Hannah M.
(2017)
Shadow detection for mobile robots:: Features, evaluation, and datasets.
Spatial Cognition & Computation, .
(doi:10.1080/13875868.2017.1322088).
Abstract
Shadows have long been a challenging topic for computer vision. This challenge is made even harder when we assume that the camera is moving, as many existing shadow detection techniques require the creation and maintenance of a background model. This article explores the problem of shadow modelling from a moving viewpoint (assumed to be a robotic platform) through comparing shadow-variant and shadow-invariant image features — primarily color, texture and edge-based features. These features are then embedded in a segmentation pipeline that provides predictions on shadow status, using minimal temporal context. We also release a public dataset of shadow-related image sequences, to help other researchers further develop shadow detection methods and to enable benchmarking of techniques.
Text
13875868.2017.1322088
- Accepted Manuscript
More information
Accepted/In Press date: 28 April 2017
e-pub ahead of print date: 28 April 2017
Identifiers
Local EPrints ID: 413327
URI: http://eprints.soton.ac.uk/id/eprint/413327
ISSN: 1387-5868
PURE UUID: 34b348b0-03c7-4574-afd8-29520e829382
Catalogue record
Date deposited: 22 Aug 2017 16:31
Last modified: 16 Mar 2024 05:40
Export record
Altmetrics
Contributors
Author:
Charles C. Newey
Author:
Owain D. Jones
Author:
Hannah M. Dee
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics