Multiclass Non-Randomized Spectral–Spatial Active Learning for Hyperspectral Image Classification

Active Learning (AL) for Hyperspectral Image Classification (HSIC) has been extensively studied. However, the traditional AL methods do not consider randomness among the existing and new samples. Secondly, very limited AL research has been carried out on joint spectral–spatial information. Thirdly,...

Full description

Saved in:
Bibliographic Details
Published inApplied sciences Vol. 10; no. 14; p. 4739
Main Authors Ahmad, Muhammad, Mazzara, Manuel, Raza, Rana Aamir, Distefano, Salvatore, Asif, Muhammad, Sarfraz, Muhammad Shahzad, Khan, Adil Mehmood, Sohaib, Ahmed
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.07.2020
Subjects
Active learning
Active Learning (AL)
Algorithms
Classification
Datasets
ELM
Fuzzy sets
Hyperspectral Image Classification (HSIC)
Hyperspectral imaging
Image classification
KNN
Learning algorithms
Methods
Pipelines
Probabilistic methods
Query Function
Spatial data
Spatial discrimination learning
Statistical analysis
Statistical methods
Statistical tests
Support vector machines
SVM
Online AccessGet full text

Cover

More Information
Summary:Active Learning (AL) for Hyperspectral Image Classification (HSIC) has been extensively studied. However, the traditional AL methods do not consider randomness among the existing and new samples. Secondly, very limited AL research has been carried out on joint spectral–spatial information. Thirdly, a minor but still worth mentioning factor is the stopping criteria. Therefore, this study caters to all these issues using a spatial prior Fuzziness concept coupled with Multinomial Logistic Regression via a Splitting and Augmented Lagrangian (MLR-LORSAL) classifier with dual stopping criteria. This work further compares several sample selection methods with the diverse nature of classifiers i.e., probabilistic and non-probabilistic. The sample selection methods include Breaking Ties (BT), Mutual Information (MI) and Modified Breaking Ties (MBT). The comparative classifiers include Support Vector Machine (SVM), Extreme Learning Machine (ELM), K-Nearest Neighbour (KNN) and Ensemble Learning (EL). The experimental results on three benchmark hyperspectral datasets reveal that the proposed pipeline significantly increases the classification accuracy and generalization performance. To further validate the performance, several statistical tests are also considered such as Precision, Recall and F1-Score.
ISSN:2076-3417
2076-3417
DOI:10.3390/app10144739