Abstract:Recent years have witnessed great progress in synthetic aperture radar (SAR) target detection methods based on deep learning. However, these methods generally assume the training data and test data obey the same distribution, which does not always hold when the radar parameters, imaging algorithm, viewpoints, scenes, etc., change in practice. When such a distribution mismatch occurs, it will cause a significant performance drop. Domain adaptation methods provide an effective way to address this problem by transferring knowledge from the source domain (training data) to the target domain (test data). In this article, we proposed an unsupervised faster R-CNN SAR target detection framework based on domain adaptation, which can improve SAR target detection performance in the unlabeled target domain by borrowing the knowledge of the labeled source domain. Our approach is composed of the following three stages: pixel-domain adaptation (PDA), multilevel feature domain adaptation (MFDA), and iterative pseudolabeling (IPL). By generating transition domain using generative adversarial networks, the PDA stage can reduce the appearance differences of SAR images. At the MFDA stage, the detector can not only learn the domain-invariant global features and instance-level regional features via multilevel adversarial learning in the common feature space but also reweight the low-level global features according to their relative importance to the target domain. At the IPL stage, we design an iterative pseudo labeling strategy that can select pseudo-labels on instance level and image level to encourage the detector to learn more discriminative features of the target domain directly. We evaluate our method using miniSAR and FARADSAR datasets. The experimental results demonstrate the effectiveness of the proposed unsupervised domain adaptation target detection approach.

Unsupervised Domain Adaptation for Human Activity Recognition in Radar

Source-Free Domain Adaptation for Millimeter Wave Radar Based Human Activity Recognition.

Domain Adaptation for Infrared-Radar Cross-Scene Multimodal Detection

Eliminate Aspect Angle Variations for Human Activity Recognition using Unsupervised Deep Adaptation Network

Unsupervised Domain Adaptation across FMCW Radar Configurations Using Margin Disparity Discrepancy

Unsupervised Aircraft Detection in SAR Images with Image-Level Domain Adaption from Optical Images

Multi-domains Based Human Activity Classification in Radar

Unsupervised Domain Adaptation for Activity Recognition Across Heterogeneous Datasets

Unsupervised Domain Adaptation for SAR Target Detection

Unsupervised Domain Adaptation for Micro-Doppler Human Motion Classification Via Feature Fusion

$\Mathtt {radar}$: Adversarial Driving Style Representation Learning with Data Augmentation

Unsupervised Domain Adaptation for Device-free Gesture Recognition

Unsupervised Domain Adaptation in Activity Recognition: A GAN-Based Approach

Self-Training Based Adversarial Domain Adaptation for Radio Signal Recognition

Unsupervised Domain Adaptation for RF-based Gesture Recognition

Human Activity Recognition Based on Deep Learning and Micro-Doppler Radar Data

MF-Match: A Semi-Supervised Model for Human Action Recognition

Human Activity Recognition Based on Multidomain Fusion Network for LFMCW Radar

Indoor 3D Human Pose Estimation Using Single Millimeter-wave Radar and Conditional Generative Adversarial Network

Omnidirectional Spectrogram Generation for Radar-based Omnidirectional Human Activity Recognition

Residual Neural Network Driven Human Activity Recognition by Exploiting FMCW Radar