2021
DOI: 10.3390/rs13112234
|View full text |Cite
|
Sign up to set email alerts
|

A Novel Query Strategy-Based Rank Batch-Mode Active Learning Method for High-Resolution Remote Sensing Image Classification

Abstract: An informative training set is necessary for ensuring the robust performance of the classification of very-high-resolution remote sensing (VHRRS) images, but labeling work is often difficult, expensive, and time-consuming. This makes active learning (AL) an important part of an image analysis framework. AL aims to efficiently build a representative and efficient library of training samples that are most informative for the underlying classification task, thereby minimizing the cost of obtaining labeled data. B… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
4
3

Relationship

0
7

Authors

Journals

citations
Cited by 7 publications
(3 citation statements)
references
References 73 publications
0
3
0
Order By: Relevance
“…Moreover, the spatial resolution of remote sensing images is low when sea fluctuations are significant [5]. Under this condition, this paper determines whether the pixels in the image are edge pixels by extracting the coastline differential edge features of the remote sensing image [6]. The closer the detected pixel gray value is to the current pixel, the more likely that point will be identified as an edge point [7].…”
Section: Extracting the Difference Edge Features Of The Coastline Rem...mentioning
confidence: 99%
See 1 more Smart Citation
“…Moreover, the spatial resolution of remote sensing images is low when sea fluctuations are significant [5]. Under this condition, this paper determines whether the pixels in the image are edge pixels by extracting the coastline differential edge features of the remote sensing image [6]. The closer the detected pixel gray value is to the current pixel, the more likely that point will be identified as an edge point [7].…”
Section: Extracting the Difference Edge Features Of The Coastline Rem...mentioning
confidence: 99%
“…In formula (6), Eki  is the shoreline edge preservation index of remote sensing image; ^( , ) ( , ) i i P x y is the coastline target pixel of the remote sensing image after noise reduction;…”
Section: Building Coastline Detection Model Of Remote Sensing Imagementioning
confidence: 99%
“…In the process of classifying multi-source remote sensing data, Pastorino et al ( 2021 ) designed a hierarchical probabilistic graphical model, which combines Markov framework and decision tree method, which has certain effectiveness and feasibility (Pastorino et al, 2021 ). In order to improve the classification effect of remote sensing images, Luo et al ( 2021 ) designed a combination strategy based on sorting batch mode, combined with spectral information divergence, and good classification effect can be obtained (Luo et al, 2021 ). Dong R. et al ( 2020 ) proposed a fast depth-aware network that combines multiple advantages to achieve simultaneous extraction of deep and shallow features (Dong R. et al, 2020 ).…”
Section: Related Workmentioning
confidence: 99%