Edge-adaptive structure tensor nonlocal kernel regression for removing cloud

Guohong Liang, Ying Li, Junqing Feng

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper the authors applies structure tensor matrix as a tool in order to survey the anisotropic structure of remote sensing image and combines nonlocal kernel regression methods for removing cloud tasks. The method utilizes both local structural regularity and the nonlocal self-similarity properties in remote sensing images. The nonlocal self-similarity takes advantages of observation that image patches incline to repeat themselves in remote sensing images. The non-local prior avails of the redundancy of similar patches remote sensing images, while the local prior consider that a target pixel can be computed by a weighted average of its neighbors. Experimental results indicate that our new algorithm better than both the steering kernel regression in persevering edge and improve the visual quality.

Original languageEnglish
Title of host publicationRecent Developments in Intelligent Systems and Interactive Applications - Proceedings of the International Conference on Intelligent and Interactive Systems and Applications, IISA 2016
EditorsFatos Xhafa, Srikanta Patnaik, Zhengtao Yu
PublisherSpringer Verlag
Pages383-389
Number of pages7
ISBN (Print)9783319495675
DOIs
StatePublished - 2017
EventInternational Conference on Intelligent and Interactive Systems and Applications, IISA2016 - Shanghai, China
Duration: 25 Jun 201626 Jun 2016

Publication series

NameAdvances in Intelligent Systems and Computing
Volume541
ISSN (Print)2194-5357

Conference

ConferenceInternational Conference on Intelligent and Interactive Systems and Applications, IISA2016
Country/TerritoryChina
CityShanghai
Period25/06/1626/06/16

Keywords

  • Cloud
  • Kernel regression
  • Nonlocal Self-similarity
  • Structure tensor

Fingerprint

Dive into the research topics of 'Edge-adaptive structure tensor nonlocal kernel regression for removing cloud'. Together they form a unique fingerprint.

Cite this