Haijian Zhang

• Ph.D
• Professor (Associate) at Wuhan University

Objects in aerial images tend to be densely scattered and appear in arbitrary orientations, making the annotation process quite costly. To reduce the annotation cost, existing methods propose randomly annotating a proportion of images or objects for aerial object detection with fewer label usage. These approaches, however, can lead to redundancy in...

Tiny objects, with their limited spatial resolution, often resemble point-like distributions. As a result, bounding box prediction using point-level supervision emerges as a natural and cost-effective alternative to traditional box-level supervision. However, the small scale and lack of distinctive features of tiny objects make point annotations pr...

Unmanned Aerial Vehicle (UAV) Cross-View Geo-Localization (CVGL) presents significant challenges due to the view discrepancy between oblique UAV images and overhead satellite images. Existing methods heavily rely on the supervision of labeled datasets to extract viewpoint-invariant features for cross-view retrieval. However, these methods have expe...

Most artificial lights exhibit subtle fluctuations in intensity and frequency in response to the influence of the grid's alternating current, providing the potential to estimate the Electric Network Frequency (ENF) from conventional frame-based videos. Nevertheless, the performance of Video-based ENF (V-ENF) estimation largely relies on the imaging...

This paper explores the important role of blind source separation (BSS) techniques in separating M mixtures including N sources using a dual-sensor array, i.e., M=2, and proposes an efficient two-stage underdetermined BSS (UBSS) algorithm to estimate the mixing matrix and achieve source recovery by exploiting time–frequency (TF) sparsity. First, we...

The stereo event-intensity camera setup is widely applied to leverage the advantages of both event cameras with low latency and intensity cameras that capture accurate brightness and texture information. However, such a setup commonly encounters cross-modality parallax that is difficult to be eliminated solely with stereo rectification especially f...

The stereo event-intensity camera setup is widely applied to leverage the advantages of both event cameras with low latency and intensity cameras that capture accurate brightness and texture information. However, such a setup commonly encounters cross-modality parallax that is difficult to be eliminated solely with stereo rectification especially f...

Most of the artificial lights fluctuate in response to the grid's alternating current and exhibit subtle variations in terms of both intensity and spectrum, providing the potential to estimate the Electric Network Frequency (ENF) from conventional frame-based videos. Nevertheless, the performance of Video-based ENF (V-ENF) estimation largely relies...

Beamforming weights prediction via deep neural networks has been one of the mainstreams in multi-channel speech enhancement tasks. The spectral-spatial cues are crucial in beamforming weights estimation, however, many existing works fails to optimally predict the beamforming weights with an absence of adequate spectral-spatial information learning....

Beamforming weights prediction via deep neural networks has been one of the main methods in multi-channel speech enhancement tasks. The spectral-spatial cues are crucial in beamforming weights estimation , however, many existing works fail to optimally predict the beamforming weights with an absence of adequate spectral-spatial information learning...

Super-Resolution from a single motion Blurred image (SRB) is a severely ill-posed problem due to the joint degradation of motion blurs and low spatial resolution. In this paper, we employ events to alleviate the burden of SRB and propose an Event-enhanced SRB (E-SRB) algorithm, which can generate a sequence of sharp and clear images with High Resol...

VIS-NIR face recognition remains a challenging task due to the distinction between spectral components of two modalities and insufficient paired training data. Inspired by the CycleGAN, this paper presents a method aiming to translate VIS face images into fake NIR images whose distributions are intended to approximate those of true NIR images, whic...

italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S uper- R esolution from a single motion B lurred image (SRB) is a severely ill-posed problem due to the joint degradation of motion blurs and low spatial resolution. In this paper, we employ events to alleviate the burden of SRB and propose an E...

This study addresses the problem of finding the optimal correspondence for a given synthetic aperture radar (SAR) image patch from a large collection of optical reference patches, which is crucial for various applications, including remote sensing, place recognition, and aircraft navigation. However, achieving one-to-one SAR-Optical Patch Correspon...

The design of high-resolution and cross-term (CT) free time-frequency distributions (TFDs) has been an open problem. Classical kernel based methods are limited by the trade-off between resolution and CT suppression, even under optimally derived parameters. To break the current limitation, we propose a data-driven model directly based on Wigner-Vill...

In recent years, object detection in high-resolution synthetic aperture radar (SAR) images has made significant progress, especially after the introduction of deep learning. However, objects such as dense oil tanks, which are compactly arranged in SAR images, are still challenging to recognize due to the unique imaging mechanism of SAR. Inspired by...

The electric network frequency (ENF) is an important and extensively researched forensic criterion to authenticate digital recordings, but currently it is still challenging to extract reliable ENF traces from recordings in uncontrollable environments. In this paper, we present a framework for robust ENF extraction from real-world audio recordings,...

Time-frequency representation (TFR) plays a significant role in interpreting and analyzing nonstationary multimode signals. However, currently, it is still challenging to handle multi-mode signals with closely-spaced or spectrally-overlapped instantaneous frequencies (IFs), especially under low signal-to-noise ratio (SNR) conditions. To address thi...

The constant Q transform (CQT) has been shown to be one of the most effective speech signal pre-transforms to facilitate synthetic speech detection, followed by either hand-crafted (subband) constant Q cepstral coefficient (CQCC) feature extraction and a back-end binary classifier, or a deep neural network (DNN) directly for further feature extract...

The constant Q transform (CQT) has been shown to be one of the most effective speech signal pre-transforms to facilitate synthetic speech detection, followed by either hand-crafted (subband) constant Q cepstral coefficient (CQCC) feature extraction and a back-end binary classifier, or a deep neural network (DNN) directly for further feature extract...

The detection of the electric network frequency (ENF) in digital recordings is an essential step before the subsequent ENF extraction and forensic analysis. In this letter, we extend the state-of-the-art single-tone time-frequency (TF) domain ENF detector to the multi-tone scenario and propose a multi-harmonic combining (MHC) method, exploiting ENF...

We present a framework for robust electric network frequency (ENF) extraction from real-world audio recordings, featuring multi-tone ENF harmonic enhancement and graph-based optimal harmonic selection. Specifically, We first extend the recently developed single-tone ENF signal enhancement method to the multi-tone scenario and propose a harmonic rob...

Power system frequency could be captured by digital recordings and extracted to compare with a reference database for forensic time-stamp verification. It is known as the electric network frequency (ENF) criterion, enabled by the properties of random fluctuation and intra-grid consistency. In essence, this is a task of matching a short random seque...

Time-frequency representation (TFR) allowing for mode reconstruction plays a significant role in interpreting and analyzing the nonstationary signal constituted of various modes. However, it is difficult for most previous methods to handle signal modes with closely-spaced or spectrally-overlapped instantaneous frequencies (IFs) especially in advers...

Blind separation of multipath fading signals with impulsive interference and Gaussian noise is a very challenging issue due to multipath effects, which are often encountered in practical scenarios. Since the strong coherence among multipath signals leads to the extreme superposition in time-frequency (TF) domain, this paper proposes an iterative th...

One fundamental problem in Earth Vision is to accurately find the locations and identify the categories of the interesting objects in the aerial images, for which oriented bounding boxes (OBBs) are usually employed to depict better the objects emerging with arbitrary orientations. However, the regression of the OBBs always suffers from the ambiguou...

Recently, it has been discovered that the electric network frequency (ENF) could be captured by digital audio, video, or even image files, and could further be exploited in forensic investigations. However, the existence of the ENF in multimedia content is not a sure thing, and if the ENF is not present, ENF-based forensic analysis would become use...

Kernel functions in quadratic time-frequency distributions (TFDs) are viewed as low-pass filters in the ambiguity function (AF) domain to suppress interfering cross-terms (CTs). Traditional kernel design methods are signal-dependent or have manually selected parameters, which impose restrictions on eliminating CTs and achieving high-resolution TFDs...

Model-driven algorithms on distributed compressive sensing with multiple measurement vectors (MMVs) have been generally based on the assumption that the vectors in the signal matrix are jointly sparse. However, the signal matrix in many practical scenarios violates the above assumption, since there might exist unknown dependency between vectors. It...

Time-frequency distributions (TFDs) play a vital role in providing descriptive analysis of non-stationary signals involved in realistic scenarios. It is well known that low time-frequency (TF) resolution and the emergency of cross-terms (CTs) are two main issues, which make it difficult to analyze and interpret practical signals using TFDs. In orde...

Future wireless communication systems are facing with many challenges due to their complexity and diversification. Orthogonal frequency division multiplexing (OFDM) in 4G cannot meet the requirements in future scenarios, thus alternative multicarrier modulation (MCM) candidates for future physical layer have been extensively studied in the academic...

Single-channel speech separation in time domain and frequency domain has been widely studied for voice-driven applications over the past few years. Most of previous works assume known number of speakers in advance, however, which is not easily accessible through monaural mixture in practice. In this paper, we propose a novel model of single-channel...

Existing audio watermarking methods usually treat the host audio signals of a function of time or frequency individually, while considering them in the joint time-frequency (TF) domain has received less attention. This paper proposes an audio watermarking framework from the perspective of TF analysis. The proposed framework treats the host audio si...

In electric network frequency (ENF) based audio forensics, the ENF signal captured in a questioned audio recording is estimated and analyzed for authentication purposes. However, the captured ENF signal is usually contaminated by very strong noise and interference. In this paper, we propose a robust filtering algorithm (RFA) for ENF signal enhancem...

Recently, the paradigm of unfolding iterative algorithms into finite-length feed-forward neural networks has achieved a great success in the area of sparse recovery. Benefit from available training data, the learned networks have achieved state-of-the-art performance in respect of both speed and accuracy. However, the structure behind sparsity, imp...

Power line detection plays an important role in an automated UAV-based electricity inspection system, which is crucial for real-time motion planning and navigation along power lines. Previous methods which adopt traditional filters and gradients may fail to capture complete power lines due to noisy backgrounds. To overcome this, we develop an accur...

VIS-NIR face recognition remains a challenging task due to the distinction between spectral components of two modalities and insufficient paired training data. Inspired by the CycleGAN, this paper presents a method aiming to translate VIS face images into fake NIR images whose distributions are intended to approximate those of true NIR images, whic...

In this paper, we propose an energy-efficient resource allocation (RA) algorithm in cognitive radio-enabled 5th generation (5G) systems, where the scenario including one primary system and multiple secondary cells is considered. Because of the high spectrum leakage of traditional orthogonal frequency division multiplexing (OFDM), alternative modula...

As an alternative to the traditional sampling theory, compressed sensing allows acquiring much smaller amount of data, still estimating the spectra of frequency-sparse signals accurately. However, compressed sensing usually requires random sampling in data acquisition, which is difficult to implement in hardware. In this paper, we propose a determi...

Recently, a comment paper on “A New Elliptical Model for Device-Free Localization” (Sensors 2016, 16, 577) has been presented, and the authors have provided a modified model. However, there are still some misunderstandings. In this reply, we further explain the proposed elliptical model in (Sensors 2016, 16, 577) to make it more understandable.

The classical solution to an underdetermined system of linear equations mainly has two opposite directions, which lead to either a large ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> -norm sparse solution or a non-sparse minimum ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xl...

Daily acquisition of large amounts of aerial and satellite images has facilitated subsequent automatic interpretations of these images. One such interpretation is object detection. Despite the great progress made in this domain, the detection of multi-scale objects, especially small objects in high resolution satellite (HRS) images, has not been ad...

Device-free localization (DFL), which can detect and locate a person by measuring the changes in received signals, is one of the primary techniques in wireless sensor networks (WSNs). Recently, research on fingerprint-based localization in changing environments has been receiving increasing attention. However, when the environment changes due to fu...

Noise suppression and the estimation of the number of sources are two practical issues in applications of underdetermined blind source separation (UBSS). This paper proposes a noise-robust instantaneous UBSS algorithm for highly overlapped speech sources in the short-time Fourier transform (STFT) domain. The proposed algorithm firstly estimates the...

In some applications of frequency estimation, it is challenging to sample at as high as Nyquist rate due to hardware limitations. The compressed sensing theory asserts that one can recover sparse signals using undersampled measurements. Recently the methods based on atomic norm techniques deal with continuous-valued frequencies and completely elimi...

Based on Time-Frequency (TF) analysis and a-contrario theory, this paper presents a new approach for extraction of linear arranged power transmission tower series in Polarimetric Synthetic Aperture Radar (PolSAR) images. Firstly, the PolSAR multidimensional information is analyzed using a linear TF decomposition approach. The stationarity of each p...

Automatic modulation classification (AMC) plays a key role in cognitive radar, cognitive radio and some other civilian and military fields to identify the type of modulation. In this paper, a deep learning based modulation classification method is developed for discriminating digital modulated signals. This proposed method uses a stacked sparse aut...

In this paper, we study the resource allocation problem of two-cell cognitive radio systems, each cell includes multiple secondary users (SUs). The aim is to maximize the sum capacity of the two cells with constraints. For comparison, two types of multi-carrier modulation, conventional orthogonal frequency division multiplexing (OFDM) and filter ba...

Device-free localization (DFL) is expected to detect and locate a person by measuring the changes of received signals in wireless sensor networks without the need of any device. Fingerprint-based DFL in changeable environments has attracted wide attenuation in recent years. However, the accuracy of fingerprint-based localization could be improved f...

Frequency estimation of multiple sinusoids is significant in both theory and application. In some application scenarios, only sub-Nyquist samples are available to estimate the frequencies. A conventional approach is to sample the signals at several lower rates. In this paper, we propose a novel method based on subspace techniques using three-channe...

Frequency estimation of multiple sinusoids is significant in both theory and application. In some application scenarios, only sub-Nyquist samples are available to estimate the frequencies. A conventional approach is to sample the signals at several lower rates. In this paper, we address frequency estimation of the signals in the time domain through...

Spectrum sensing has been identified as an essential enabling functionality for cognitive radio (CR) systems to guarantee that CR users could share the spectrum resource with licensed users on a non-interfering basis. Recently, simultaneous sensing of multi-band licensed user activity has been attracting more and more research interest. Generally,...

Recent development of compressive sensing technology has greatly benefited radar imaging problems. In this paper, we investigate the problem of obtaining enhanced targets of interest such as ships and airplanes, where targets often exhibit structured sparsity in the imaging scene. A novel structured sparsity-driven autofocus algorithm is proposed b...

As an alternative to the traditional sampling theory, compressed sensing allows acquiring much smaller amount of data, still estimating the spectra of frequency-sparse signals accurately. However, compressed sensing usually requires random sampling in data acquisition, which is difficult to implement in hardware. In this paper, we propose a determi...

As an important extension to sparsity, Block Structured Sparsity (BSS) has been widely investigated and a large set of algorithms have been designed to recover signals with BSS. In this paper, a dynamic system is proposed to recover signals with BSS, namely, D-BSS. Particularly, D-BSS turns to exploiting the dynamic systems governed by an norm cons...

Device-free localization (DFL) based on wireless sensor networks (WSNs) is expected to detect and locate a person without the need for any wireless devices. Radio tomographic imaging (RTI) has attracted wide attention from researchers as an emerging important technology in WSNs. However, there is much room for improvement in localization estimation...

Device-free localization (DFL) based on wireless sensor networks (WSNs) is expected to detect and locate a person without the need for any wireless devices. Radio tomographic imaging (RTI) has attracted wide attention from researchers as an emerging important technology in WSNs. However, there is much room for improvement in localization estimation...

In this work, we propose an adaptive set-membership (SM) reduced-rank filtering algorithm using the constrained constant modulus criterion for beamforming. We develop a stochastic gradient type algorithm based on the concept of SM techniques for adaptive beamforming. The filterweights are updated only if the bounded constraint cannot be satisfied....

In this paper, a class of deterministic sensing matrices are constructed by selecting rows from Fourier matrices. These matrices have better performance in sparse recovery than random partial Fourier matrices. The coherence and restricted isometry property of these matrices are given to evaluate their capacity as compressive sensing matrices. In ge...

Direction of arrival (DOA) estimation methods based on joint sparsity are attractive due to their superiority of high resolution with a limited number of snapshots. However, the common assumption that signals from different directions share the spectral band is inappropriate when they occupy different bands. To flexibly deal with this situation, a...

In this letter, a sinusoidal time-frequency distribution based filtering (STFD-F) algorithm is proposed and analysed for estimating mono-component stationary sinusoidal signals embedded in strong noise. An initial frequency estimation of the sinusoidal signal is required in the STFD-F algorithm. We theoretically derive the closed-form expressions o...

In some applications of frequency estimation, the frequencies of multiple sinusoids are required to be estimated from sub-Nyquist sampling sequences. In this paper, we propose a novel method based on subspace techniques to estimate the frequencies by using under-sampled samples. We analyze the impact of under-sampling and demonstrate that three sub...

In some applications of frequency estimation, it is challenging to sample at as high as Nyquist rate due to hardware limitations. Based on conventional ESPRIT, a new method for frequency estimation with double-channel sub-Nyquist sampling is proposed. When the two undersampled multiples are relatively prime integers, partial samples are selected to...

In many applications of frequency estimation, the frequencies of the signals are so high that the data sampled at Nyquist rate are hard to acquire due to hardware limitation. In this paper, we propose a novel method based on subspace techniques to estimate the frequencies by using two sub-Nyquist sample sequences, provided that the two under-sample...

The robust detection of ships is one of the key techniques in coastal and marine applications of synthetic aperture radar (SAR). Conventional SAR ship detectors involved multiple parameters, which need to be estimated or determined very carefully. In this paper, we propose a new ship detection approach based on multi-scale heterogeneities under the...

Frequency hopping (FH) signals have been widely employed in wireless communication networks to combat interference and avoid collision. This paper considers the blind FH signal estimation problem in antenna array systems, where the direction-of-arrivals, hopping time and frequency are all unknown to the users. A hierarchical sparsity-aware techniqu...

This paper considers the problem of estimating multiple frequency hopping signals with unknown hopping pattern. By segmenting the received signals into overlapped measurements and leveraging the property that frequency content at each time instant is intrinsically parsimonious, a sparsity-inspired high-resolution time-frequency representation (TFR)...

Conventional time-varying filtering is inefficient for severely spoiled signals because it requires discernible spectral content of the signal in time–frequency domain. Motivated by the time–frequency peak filtering (TFPF) algorithm, a robust time-varying filtering (RTVF) algorithm is proposed in this paper for the objectives of filtering and separ...

In multi-hop WLANs, it is more efficient to let some clients relay traffic for the clients whose distances are a bit far away from the AP, which increases the transmission efficiency. The existing solutions for multi-hop WLANs try to choose one or two fixed best relays to forward packets. However, the links between faraway clients and the AP are in...

Frequency hopping signals have been widely employed in wireless networks due to its robustness in anti-jamming and interference. In the scenario of coexistent networks, however , each user inevitably receives multiple unknown frequency hopping signals from different networks. In wireless networks with energy constraint, conventional re-transmission...

Pedestrian detection is one of the most important task for video analytics of an intelligent surveillance system. In this paper, we propose a framework to improve the detection performance of a generic pedestrian detector for highly crowded scenes. The generic offline-trained pedestrian detectors usually cannot handle the problem of detecting pedes...

In this paper, a speech signal recovery algorithm is presented for a personalized voice command automatic recognition system in vehicle and restaurant environments. This novel algorithm is able to separate a mixed speech source from multiple speakers, detect presence/absence of speakers by tracking the higher magnitude portion of speech power spect...

Automatic modulation classification (AMC) has been a significant research topic in communication systems especially cognitive radio systems. The development of AMC algorithms is still at an immature stage for practical applications. In this paper, a supervised modulation classification scheme is proposed for automatic recognition of different types...

In this paper, an automatic speech recognition (ASR) system under ubiquitous environment is proposed, which is successfully implemented in a personalized voice command system under vehicle and living room environment. The proposed ASR system describes a novel scheme of separating speech sources from multi-speakers, detecting speech presence/absence...

The estimation of multicomponent radar signals with overlapped instantaneous frequencies (IFs) in low SNR environments has been a challenging research topic. This paper proposes an IF estimation algorithm for spectrally overlapped radar signals which contain continuous and stepped IF laws. Firstly, we design a gradient image via the time-frequency...

Motivated by the existing time-frequency peak filtering (TFPF) algorithm, herein a robust time-varying filtering (RTVF) algorithm is proposed for filtering and separating multicomponent frequency modulation (FM) signals. The performance of the TFPF based on windowed Wigner-Ville distribution is limited by the linear constraint on the waveform of th...

This letter considers the multiplicative perturbation problem in compressive sensing, which has become an increasingly important issue on obtaining robust performance for practical applications. The problem is formulated in a probabilistic model and an auto-calibration sparse Bayesian learning algorithm is proposed. In this algorithm, signal and pe...

The estimation of the mixing matrix as well as the number of sources in blind source separation are two challenging problems. This paper proposes an effective estimation method to solve these two problems for underdetermined blind separation of overlapped sources in short-time Fourier transform (STFT) domain. Our study considers the blind estimatio...

Most existing classification methods cannot work in low signal-to-noise ratio (SNR) environments. This limitation motivates the signal filtering before the classification process. In this paper, a general framework that links the time-frequency peak filtering (TFPF) and traditional feature-based signal classification is explored. As the name sugges...

Cognitive radio (CR) has been proposed to improve spectral efficiency while avoiding interference with licensed users. In this paper, we propose a resource allocation (RA) algorithm to perform uplink frequency allocation and power allocation among noncooperative multicells with multiuser per cell in CR systems. The maximization of the total informa...

Cognitive radio (CR) is proposed to automatically detect and exploit unused spectrum while avoiding harmful interference to the incumbent system. In this paper, we emphasize the channel capacity comparison of a CR network using two types of multicarrier communications: conventional Orthogonal Frequency Division Multiplexing (OFDM) with Cyclic Prefi...

Cognitive Radio (CR) is a fully reconfigurable radio that can intelligently change its communication variables in response to network and user demands. The ultimate goal of CR is to allow the Secondary User (SU) to utilize the available spectrum resource on a non-interfering basis to the Primary User (PU) by sensing the existence of spectrum holes....