2018
DOI: 10.1109/tgrs.2018.2790262
|View full text |Cite
|
Sign up to set email alerts
|

Nonlocal Tensor Completion for Multitemporal Remotely Sensed Images’ Inpainting

Abstract: This is the pre-acceptance version, to read the final version please go to IEEE Transactions on Geoscience and Remote Sensing on IEEE Xplore. Remotely sensed images may contain some missing areas because of poor weather conditions and sensor failure. Information of those areas may play an important role in the interpretation of multitemporal remotely sensed data. The paper aims at reconstructing the missing information by a non-local low-rank tensor completion method (NL-LRTC). First, non-local correlations in… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
30
0
1

Year Published

2019
2019
2024
2024

Publication Types

Select...
9

Relationship

2
7

Authors

Journals

citations
Cited by 77 publications
(31 citation statements)
references
References 66 publications
0
30
0
1
Order By: Relevance
“…We present our experiments and comparisons of different cloud removal methods: Our proposed STGCN method, the non-local low-rank tensor completion method (NL-LRTC) [16], the recent spatial-temporal-spectral based cloud removal algorithm via CNN (STSCNN) [36], and the very recent temporal-based cloud removal network (CRN) [41]. A partial convolution-based in-painting technology for irregular holes (Pconv) [48] was also executed for quantitative and visual comparison.…”
Section: Cloud Removal Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…We present our experiments and comparisons of different cloud removal methods: Our proposed STGCN method, the non-local low-rank tensor completion method (NL-LRTC) [16], the recent spatial-temporal-spectral based cloud removal algorithm via CNN (STSCNN) [36], and the very recent temporal-based cloud removal network (CRN) [41]. A partial convolution-based in-painting technology for irregular holes (Pconv) [48] was also executed for quantitative and visual comparison.…”
Section: Cloud Removal Resultsmentioning
confidence: 99%
“…Cheng et al adopted an image in-painting technology based on non-local total variation, in which multi-band data were utilized to achieve spectral coherence. Non-local correlation in the spatial domain and low-rankness in the spatial-temporal-temporal domain were considered separately in reference [16] to reconstruct missing pixels. Recently, He et al [17] designed a new low-rank tensor decomposition method and a total variation model for missing information reconstruction.…”
Section: Of 19mentioning
confidence: 99%
“…Ng et al [34] and Cheng et al [35] designed adaptive weighted tensor completion methods to recover the cloud-contaminated pixels. Ji et al [36] suggested a nonlocal tensor completion method to restore the missing area. Chen et al [37] proposed a low-rank sparsity decomposition method regularized by total variation, called TVLRSDC, which could remove the cloud and cloud shadow simultaneously.…”
Section: Temporal Information-based Methodsmentioning
confidence: 99%
“…Following [49][50][51], we use lower-case letters for vectors, e.g., a; upper-case letters for matrices, e.g., A; and calligraphic letters for tensors, e.g., A. An N -mode tensor is defined as X ∈ R I1×I2×···×I N , and x i1,i2,··· ,i N denotes its (i 1 , i 2 , · · · , i N )-th component.…”
Section: Fmentioning
confidence: 99%