Huaizhang Liao

Huaizhang Liao

Auto-generated profile, based on publicly available information

Huaizhang Liao

College of Electronic Science National University of Defense Technology,Changsha,China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 September 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
HLHuaizhang Liao
Huaizhang Liao
JXJingyuan Xia
Jingyuan Xia
ZYZhixiong Yang
Zhixiong Yang
FPFulin Pan
Fulin Pan
ZLZhen Liu
Zhen Liu
YLYongxiang Liu
Yongxiang Liu
This paper focuses on generating Inverse Synthetic Aperture Radar (ISAR) images from optical images, in particular, for orbit space targets. ISAR images are widely applied in space target observation and classification tasks, whereas, limited to the expensive cost of ISAR sample collection, training deep learning-based ISAR image classifiers with insufficient samples and generating ISAR samples from emulation optical images via image translation techniques have attracted increasing attention. Image translation has highlighted significant success and popularity in computer vision, remote sensing and data generation societies. However, most of the existing methods are implemented under the discipline of extracting the explicit pixel-level features and do not perform effectively while entailing translation to domains with specific implicit features, such as ISAR image does. We propose a meta-learning based domain prior to implicit feature modelling and apply it to CycleGAN and UNIT models to realize effective translations between the ISAR and optical domains. Two representative implicit features, ISAR scattering distribution feature from the physical domain and the classification identifying feature from the task domain, are elaborately formulated with explicit modelling in statistic form. A meta-learning based training scheme is introduced to leverage the mutual knowledge of domain priors across different samples, and thus allows few-shot learning capacity with dramatically reduced training samples. Extensive simulations validate that the obtained ISAR images have better visible-authenticity and training-effectiveness than the existing image translation approaches on various synthetic datasets. Source codes are available at.

23 April 2025 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
YZYuxin Zhao
Yuxin Zhao
HLHuaizhang Liao
Huaizhang Liao
DKDerong Kong
Derong Kong
ZYZhixiong Yang
Zhixiong Yang
JXJingyuan Xia
Jingyuan Xia
This paper addresses the challenge of generating high-fidelity Inverse Synthetic Aperture Radar (ISAR) images from optical images, particularly for space targets. We propose a framework for the generation of ISAR images incorporating component refinement, which attains high-fidelity ISAR scattering characteristics through the integration of an advanced generation model predicated on semantic segmentation, designated as Semantic-Aware ISAR Generation (SAIG). SAIG renders ISAR images from optical equivalents by learning mutual semantic segmentation maps. Extensive simulations demonstrate its effectiveness and robustness, outperforming state-of-the-art methods by over 8% across key evaluation metrics.

17 November 2023
Verified
Conference Paper
Closed access
HLHuaizhang Liao
Huaizhang Liao
JXJingyuan Xia
Jingyuan Xia
ZYZhixiong Yang
Zhixiong Yang
ZLZhen Liu
Zhen Liu
YLYongxiang Liu
Yongxiang Liu
Despite the deep learning has shown revolutionary success in classification tasks, the performance is determined by the training data. Therefore, image augmentation and generation are necessary for the tasks that are lack of sufficient training samples. This paper focuses on generating Inverse Synthetic Aperture Radar (ISAR) images from the optical counterparts, in particular, for the classification of the satellite tasks. We propose a hierarchical domain prior CycleGAN (HDP-CycleGAN) to achieve an image translation between the ISAR and optical satellite images, as such providing synthetic training samples for deep classification models. The proposed HDP-CycleGAN constitutes a physical feature projection consistency for ISAR scattering distribution feature extraction and a classification-consistency to integrate the identifying feature with the generative model. Extensive simulations validate that the obtained ISAR images have better visible-authenticity and training-effectiveness than the existing image translation alternatives.

27 July 2023
Verified
Conference Paper
Closed access
MCMingyao Chen
Mingyao Chen
ZYZhixiong Yang
Zhixiong Yang
HLHuaizhang Liao
Huaizhang Liao
JXJingyuan Xia
Jingyuan Xia
TLTianpeng Liu
Tianpeng Liu
LLLi Liu
Li Liu
In inverse synthetic aperture radar (ISAR) imaging applications, the lack of image clarity often limits their usage in down-streaming tasks such as recognition and component judgment. The super-resolution of ISAR images after imaging needs to consider the uncertainty of the blur kernel and cannot simply use traditional super-resolution (SR) methods. Moreover, ISAR images are scarce, expensive to obtain, and pre-training based on a large number of images is challenging. Therefore, this paper proposes a Single-Image Super-Resolution (SISR) method with generalization ability for different blur kernels, which does not rely on a large amount of image data training for ISAR image super-resolution. A large number of experiments show that the proposed method is superior to the traditional Bicubic algorithm and learning-based DIP, Double-DIP methods, and achieves significantly better performance with less pre-training time.

21 October 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
HLHuaizhang Liao
Huaizhang Liao
YLYun Liu
Yun Liu
JXJingyuan Xia
Jingyuan Xia
YLYongxiang Liu
Yongxiang Liu
With the increasing complexity of radar task scenarios and target characteristics, conventional monostatic High Resolution Range Profile recognition methods struggle to meet the demands for high-performance and robust classification, underscoring the critical need for multistation radar fusion frameworks. However, current multistation HRRP fusion approaches predominantly rely on implicit feature learning from the holistic HRRP series, overlooking the differential contributions of local discriminative segments and globally informative frequency-domain representations that capture physical scattering characteristics. These limitations lead to significant performance degradation under noisy conditions. To address these challenges, this paper proposes a Triple Stream Information Extraction (TSIE) framework that comprehensively exploits time-domain, frequency-domain, and time-frequency-domain characteristics for multistation HRRP recognition. Our framework introduces three dedicated components: (1) a Time Segmentation Encoding module that employs a stochastic Monte Carlo sampling process to adaptively identify and emphasize critical local discriminative segments; (2) a Global Frequency Embedding module that utilizes an energy-guided attention mechanism to adaptively extract globally informative and physically meaningful features from the spectrum; and (3) a Time-Frequency Analysis module based on Continuous Wavelet Transform to capture localized spectral dynamics and transient patterns. The complementary features from these three streams are integrated through a Multi-Contrastive Learning strategy that enforces feature consistency across stations while enhancing inter-class discriminability. Extensive experiments on both simulated and measured HRRP datasets demonstrate that TSIE achieves the best recognition accuracies of 97.57% and 96.21%, respectively, outperforming baseline methods. Crucially, under the challenging noise-injection (SNR=0dB) condition, TSIE maintains robust performance with 81.27% accuracy, demonstrating superior performance over baseline approaches.