Abstract

The main challenge in a rain removal algorithm is to differentiate rain streak from moving objects. This paper addresses this problem using the spa-tiotemporal appearance technique (STA). Although the STA-based technique can significantly remove rain from video, in some cases it cannot properly retain all the moving object regions. The photometric feature of rain streak was used to solve this issue. In this paper, a new algorithm combining STA and the photo-metric correlation between rain streak and background is proposed. Rain streak and moving objects were successfully detected and separated by combining both techniques, then fused to obtain well-recovered moving objects with rain-free video. The experimental results reveal that the proposed algorithm significantly outperforms the state-of-the-art methods for both real and synthetic rain streak.
Original languageEnglish
Title of host publicationImage and Video Technology
Subtitle of host publicationPSIVT 2019 International Workshops Sydney, NSW, Australia, November 18-22, 2019 Revised Selected Papers
EditorsJoel Janek Dabrowski, Ashfaqur Rahman, Manoranjan Paul
Place of PublicationCham, Switzerland
PublisherSpringer Nature Switzerland AG
Pages3-13
Number of pages11
Edition1
ISBN (Electronic)9783030397708
ISBN (Print)9783030397692
DOIs
Publication statusPublished - 2020
Event9th Pacific-Rim Symposium on Image and Video Technology: PSIVT 2019 - Charles Sturt University Study Centre, Sydney, Australia
Duration: 18 Nov 201922 Nov 2019
http://www.psivt.org/psivt2019/index.html
http://www.psivt.org/psivt2019/program.html (program)
https://link.springer.com/book/10.1007/978-3-030-34879-3 (proceedings)

Publication series

NameLecture Notes in Computer Science
PublisherSpringer, Cham
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference9th Pacific-Rim Symposium on Image and Video Technology
Country/TerritoryAustralia
CitySydney
Period18/11/1922/11/19
Internet address

Fingerprint

Dive into the research topics of 'Rain streak removal with well-recovered moving objects from video sequences using photometric correlation'. Together they form a unique fingerprint.

Cite this