2021
DOI: 10.1007/978-3-030-86517-7_28
|View full text |Cite|
|
Sign up to set email alerts
|

Automated Machine Learning for Satellite Data: Integrating Remote Sensing Pre-trained Models into AutoML Systems

Abstract: Current AutoML systems have been benchmarked with traditional natural image datasets. Differences between satellite images and natural images (e.g., bit-wise resolution, the number, and type of spectral bands) and lack of labeled satellite images for training models, pose open questions about the applicability of current AutoML systems on satellite data. In this paper, we demonstrate how AutoML can be leveraged for classification tasks on satellite data. Specifically, we deploy the Auto-Keras system for image … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
4
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6

Relationship

1
5

Authors

Journals

citations
Cited by 6 publications
(4 citation statements)
references
References 29 publications
0
4
0
Order By: Relevance
“…We implemented our methods in AutoKeras, like the work on image classification by Palacios Salinas et al [64]. The AutoKeras library is a natural choice for SR because it already contains functionality for image tasks.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…We implemented our methods in AutoKeras, like the work on image classification by Palacios Salinas et al [64]. The AutoKeras library is a natural choice for SR because it already contains functionality for image tasks.…”
Section: Methodsmentioning
confidence: 99%
“…In atmospheric science, Zheng et al, employed the FLAML framework to estimate particulate matter concentrations in satellite measurements [60]. In image classification, Palacios Salinas et al, proposed a network architecture search (NAS) system optimised for classifying EO images with blocks that were pre-trained on four EO datasets (e.g., [16,[61][62][63]) by customising the search space of AutoKeras [64]. Another approach for object recognition in EO images was presented by Polonskaia et al, who proposed an automated evolutionary NAS approach for designing CNNs implemented in Auto-Pytorch [65].…”
Section: Automl For Eo Tasksmentioning
confidence: 99%
“…Another challenge in DL is hyperparameter tuning. Due to the complexity and configuration variability of large DL models, we must use the best-performing configuration based on hyperparameter tuning to obtain the optimal inference results from the model [6], [18]. In some cases, the hyperparameter space can be prohibitively large, and the requirement to iterate over possible training configurations to arrive at the optimum using large RS images can be a serious computational challenge.…”
Section: Challenges In Deep Learningmentioning
confidence: 99%
“…Two main obstacles in using DL methods in RS applications are; 1) the limited availability of training data sets and 2) the prohibitive nature of determining the configuration parameters for the DL network. In literature, the search for optimal hyperparameters is referred to as hyperparameter tuning and is imperative when designing DL models that can capture high-level semantics [6].…”
mentioning
confidence: 99%