Graph-based active learning for semi-supervised classification of SAR data

@inproceedings{Miller2022GraphbasedAL,
  title={Graph-based active learning for semi-supervised classification of SAR data},
  author={Kevin Miller and John Mauro and Jason Setiadi and Xoaquin Baca and Zhan Shi and Jeff Calder and A. Bertozzi},
  booktitle={Defense + Commercial Sensing},
  year={2022}
}
We present a novel method for classification of Synthetic Aperture Radar (SAR) data by combining ideas from graph-based learning and neural network methods within an active learning framework. Graph-based methods in machine learning are based on a similarity graph constructed from the data. When the data consists of raw images composed of scenes, extraneous information can make the classification task more difficult. In recent years, neural network methods have been shown to provide a promising… 
View on SPIE
[PDF] Semantic Reader
6 Citations
Background Citations
1
Methods Citations
2

6 Citations

Hamilton-Jacobi equations on graphs with applications to semi-supervised learning and data depth

It is shown that the continuum limit of the p-eikonal equation on a random geometric graph recovers a geodesic density weighted distance in the continuum, and the p→∞ limit recovers shortest path distances.

Joint reconstruction-segmentation on graphs

,

Research

Research

impact and the team lift; (5) reputation and aging effects on the success in creative markets; and (6) established patterns of socio-demographic homophily on online social media.

Research

Methods to identify substrates of ubiquitin ligases represent an important class of both oncogenes and tumor suppressors, and are applying these to both yeast and human cells.

Active Learning of Non-semantic Speech Tasks with Pretrained Models

ALOE is proposed, a novel system for improving the data- and label-efficiency of non-semantic speech tasks with active learning (AL).

References

SHOWING 1-10 OF 44 REFERENCES

Target Classification Using the Deep Convolutional Networks for SAR Images

A new all-convolutional networks (A-ConvNets), which only consists of sparsely connected layers, without fully connected layers being used, which can achieve an average accuracy of 99% on classification of ten-class targets and is significantly superior to the traditional ConvNets on the classification of target configuration and version variants.

SAR target recognition based on deep learning

  • Sizhe ChenHaipeng Wang
  • Computer Science
    2014 International Conference on Data Science and Advanced Analytics (DSAA)
  • 2014
This paper attempts to adapt the optical camera-oriented CNN to its microwave counterpart, i.e. synthetic aperture radar (SAR), as a preliminary study, a single layer of convolutional neural network is used to automatically learn features from SAR images.

SAR Automatic Target Recognition Based on Euclidean Distance Restricted Autoencoder

A deep learning method based on a multilayer autoencoder (AE) combined with a supervised constraint to use the limited training images well and to prevent overfitting caused by supervised learning is proposed.

Unsupervised Spatial–Spectral Feature Learning by 3D Convolutional Autoencoder for Hyperspectral Classification

Experimental results on several benchmark hyperspectral data sets have demonstrated that the proposed 3D-CAE is very effective in extracting spatial–spectral features and outperforms not only traditional unsupervised feature extraction algorithms but also many supervised feature extraction algorithm in classification application.

Model-Change Active Learning in Graph-Based Semi-Supervised Learning

This work considers a family of convex loss functions for which the acquisition function can be efficiently approximated using the Laplace approximation of the posterior distribution and shows a variety of multiclass examples that illustrate improved performance over prior state-of-art.

Application of deep-learning algorithms to MSTAR data

A new All-Convolutional Networks (A-convNets) is proposed and applied to Moving and Stationary Target Acquisition and Recognition (MSTAR) data and can significantly reduce the number of free parameters and the degree of overfitting.

A Deep Learning SAR Target Classification Experiment on MSTAR Dataset

This paper analyzes the potential of using additional radar information, such as phase information, in the deep learning process and proposes another deep learning method for automated target recognition in Synthetic Aperture Radar images.

Combining active learning and semi-supervised learning using Gaussian fields and harmonic functions

This work combines active and semi-supervised learning techniques under a Gaussian random field model, which requires a much smaller number of queries to achieve high accuracy compared with random query selection.

Target Recognition in SAR Images via Classification on Riemannian Manifolds

In this letter, synthetic aperture radar (SAR) target recognition via classification on Riemannian geometry is presented and two classification schemes are proposed, including a sparse representation-based classification.

Semi-Supervised Learning Using Gaussian Fields and Harmonic Functions

An approach to semi-supervised learning is proposed that is based on a Gaussian random field model, and methods to incorporate class priors and the predictions of classifiers obtained by supervised learning are discussed.