XLXiaofei Lu

Xiaofei Lu

*This is an algorithmically generated author record based on publicly available information.

Xiaofei Lu has a total of 59 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Target Localization of a Quadrotor UAV with Multi-Level Coordinate System Transformation Based on Monocular Camera Position Compensation

    by Zhefu Zheng, Haoting Liu, Zhipeng Ye, Mengmeng Wang, Haiguang Li, Xiaofei Lu, Qing Li

    In recent years, unmanned aerial vehicle (UAV) technology has been increasingly widely used in natural disaster rescue. To enable fast and accurate localization of rescue targets in disaster environments, this paper proposes a multi-level coordinate system transformation method for quadrotor UAVs based on monocular camera position compensation. First, the preprocessed image object is transformed from pixel coordinates to camera coordinates. Second, to address the issue that coupling errors between the camera and UAV coordinate systems degrade the accuracy of coordinate conversion and target positioning, a Static–Dynamic Compensation Model (SDCM) for UAV camera position error is established. This model leverages a UAV attitude-based compensation mechanism to enable accurate conversion of camera coordinates to UAV coordinates and north-east-down (NED) coordinates. Finally, according to the Earth model, a multi-level continuous conversion chain from the target coordinates to the Earth-centered–Earth-fixed (ECEF) coordinates and the world-geodetic-system 1984 (WGS84) coordinates is constructed. Extensive experimental results show that the accuracy of the overall positioning method is improved by approximately 23.8% after completing our camera position compensation, which effectively enhances the positioning performance under the basic method of coordinate transformation, and provides technical support for the rapid rescue in the post-disaster phase.


  • Open Access

    Interactive Maintenance of Space Station Devices Using Scene Semantic Segmentation

    by Haoting Liu, Chuanxin Liao, Xikang Li, Zhen Tian, Mengmeng Wang, Haiguang Li, Xiaofei Lu, Zhenhui Guo, Qing Li

    A novel interactive maintenance method for space station in-orbit devices using scene semantic segmentation technology is proposed. First, a wearable and handheld system is designed to capture images from the astronaut in the space station’s front view scene and play these images on a handheld terminal in real-time. Second, the proposed system quantitatively evaluates the environmental lighting condition in the scene by calculating image quality evaluation parameters. If the lighting condition is not proper, a prompt message will be given to the astronaut to remind him or her to adjust the environment illumination. Third, our system adopts an improved DeepLabV3+ network for semantic segmentation of these astronauts’ forward view scene images. Regarding the improved network, the original backbone network is replaced with a lightweight convolutional neural network, i.e., the MobileNetV2, with a smaller model scale and computational complexity. The convolutional block attention module (CBAM) is introduced to improve the network’s feature perception ability. The atrous spatial pyramid pooling (ASPP) module is also considered to enable an accurate calculation of encoding multi-scale information. Extensive simulation experiment results indicate that the accuracy, precision, and average intersection over the union of the proposed algorithm can be better than 95.0%, 96.0%, and 89.0%, respectively. And the ground application experiments have also shown that our proposed technique can effectively shorten the working time of the system user.

  • Open Access

    Small Scale Multi-Object Segmentation in Mid-Infrared Image Using the Image Timing Features–Gaussian Mixture Model and Convolutional-UNet

    by Meng Lv, Haoting Liu, Mengmeng Wang, Dongyang Wang, Haiguang Li, Xiaofei Lu, Zhenhui Guo, Qing Li

    The application of intelligent video monitoring for natural resource protection and management has become increasingly common in recent years. To enhance safety monitoring during the grazing prohibition and rest period of grassland, this paper proposes a multi-object segmentation algorithm based on mid-infrared images for all-weather surveillance. The approach integrates the Image Timing Features–Gaussian Mixture Model (ITF-GMM) and Convolutional-UNet (Con-UNet) to improve the accuracy of target detection. First, a robust background modelling, i.e., the ITF-GMM, is proposed. Unlike the basic Gaussian Mixture Model (GMM), the proposed model dynamically adjusts the learning rate according to the content difference between adjacent frames and optimizes the number of Gaussian distributions through time series histogram analysis of pixels. Second, a segmentation framework based on Con-UNet is developed to improve the feature extraction ability of UNet. In this model, the maximum pooling layer is replaced with a convolutional layer, addressing the challenge of limited training data and improving the network’s ability to preserve spatial features. Finally, an integrated computation strategy is designed to combine the outputs of ITF-GMM and Con-UNet at the pixel level, and morphological operations are performed to refine the segmentation results and suppress noises, ensuring clearer object boundaries. The experimental results show the effectiveness of proposed approach, achieving a precision of 96.92%, an accuracy of 99.87%, an intersection over union (IOU) of 94.81%, and a recall of 97.75%. Furthermore, the proposed algorithm meets real-time processing requirements, confirming its capability to enhance small-target detection in complex outdoor environments and supporting the automation of grassland monitoring and enforcement.


  • 2-D Joint High-Resolution ISAR Imaging With Random Missing Observations via Cyclic Displacement Decomposition-Based Efficient SBL

    by Yuanyuan Wang, Fengzhou Dai, Qian Liu, Xiaofei Lu

    In this paper, the efficient sparse Bayesian learning (SBL) based two-dimensional (2D) joint high-resolution inverse synthetic aperture radar (ISAR) imaging approach with random missing observations in both the range-frequency and slow-time domains caused by frequency agility and pulse repetition interval (PRI) jitter is proposed. Firstly, the target return signal model containing the translational motion caused envelope migration and phase error and rotation caused range spatial-variant phase error (RSVPE) with 2D randomly missing observations is established. Next, considering the rotation caused RSVPE need to be estimated and compensated for each range cell individually in range domain, the modified conditional mean estimator based missing observations recover method and its fast implementation method based on fast Fourier transform (FFT) is designed. Following, the SBL based cyclic iteration approach is proposed to recover missing observations, estimate target motion parameters and compensate for the translational motion and RSVPE, and achieve the focused and cross-range scaled ISAR image. In addition, to alleviate the computational complexity increasing caused by data filling for SBL, a novel low cyclic displacement rank decomposition (LCDRD) for the Toeplitz-block-Toeplitz (TBT) structured covariance matrix of the completed observations in SBL is proposed and applied to design efficient SBL based ISAR imaging algorithms for two different types of observations. Finally, the effectiveness of the proposed algorithms is validated using both simulated and measured data.


  • Fast SBL for 2-D Joint Super-Resolution ISAR Imaging With Multidwell Observation Based on LC Decomposition of Fourth-Order Toeplitz Tensor

    by Yuanyuan Wang, Fengzhou Dai, Qian Liu, Yi Liang, Xiaofei Lu

    For multifunctional phased array radar systems, inverse synthetic aperture radar (ISAR) imaging typically requires multi-dwell coherent processing to achieve adequate cross-range resolution. In such cases, the traditional fast Fourier Transform (FFT)-based Doppler processing method can result in grating lobes in the cross-range dimension. Additionally, the range resolution is insufficient when the radar transmission bandwidth is relatively narrow. To address these two issues, this paper proposes a fast sparse Bayesian learning (SBL)/iterative Wiener filter (IWF)-based two-dimensional (2D) joint super-resolution ISAR imaging method, which enhances the resolution in both range and cross-range dimensions while suppressing grating lobes. The proposed fast SBL and IWF algorithms incorporate the lower-tri-Toeplitz-cyclic (LC) decomposition of the unfolded fourth-order Toeplitz tensor of the covariance matrix. Except for the LC decomposition, all operations utilize the FFT, allowing for efficient, precise, and memory-efficient processing of two types of multi-dwell observations. Furthermore, during image reconstruction, the target’s rotational velocity is estimated using the minimum entropy criterion, thereby achieving range-variant autofocus and 2D super-resolution ISAR imaging. Finally, simulated and measured data are employed to validate the effectiveness of the proposed algorithms.


  • HRRPs Reconstruction of Multiple High-Speed Targets With Large Time-Bandwidth Product LFM Waveform Based on Efficient SBL

    by Hang Dong, Fengzhou Dai, Juan Zhang, Xiaofei Lu

    High resolution range profiles (HRRPs) based ballistic target recognition (BTR) and intercept kill assessment (IKA) are critical missions in ballistic missile defense (BMD). However, the high speed motion brings non-negligible in-pulse Doppler effect, which makes the HRRP distorted, especially in transmitting large time-bandwidth product (LTBP) waveform. Meanwhile, there are multiple high-speed targets before and after the moment of interception in different flight states. In the article, the method for HRRPs reconstruction of multiple high-speed targets is proposed based on minimum entropy FFT-CLEAN and efficient sparse Bayesian learning (ESBL). First, multiple high-speed targets are not suitable for simultaneous processing in the HRRPs reconstruction process. A minimum entropy FFT-CLEAN algorithm is proposed, which can detect and separate the echo of each target, and obtain the coarse velocity estimate. Second, the HRRPs reconstruction and accurate velocity estimation are unified into the sparse Bayesian framework. To improve the reconstruction efficiency, Gohberg-Semencul (GS) factorization and preconditioned conjugate gradient (PCG) method are used in the SBL inference process. Finally, simulation and measurement results demonstrate that the proposed algorithm achieves better HRRP reconstruction performance and higher computational efficiency.


  • Open Access

    Sparse SAR Imaging Method for Ground Moving Target via GMTSI-Net

    by Luwei Chen, Jiacheng Ni, Ying Luo, Qifang He, Xiaofei Lu

    Ground moving targets (GMT), due to the existence of velocity in range and azimuth direction, will lead to the deviation from their true position and defocus in the azimuth direction during the synthetic aperture radar (SAR) imaging process. To address this problem and compress the amount of echo data, a sparse SAR imaging method for ground moving targets is proposed. Specifically, we first constructed a two-dimensional sparse observation model of the GMT based on matched filter operators. Then, the observation model was solved by a deep network, GMT sparse imaging network (GMTSI-Net), which was mainly obtained by unfolding an iterative soft threshold algorithm (ISTA)-based iterative solution. Furthermore, we designed an adaptive unfolding module in the imaging network to improve the adaptability of the network to the input of echo data with different sampling ratios. The proposed imaging network can achieve faster and more accurate SAR images of ground moving targets under a low sampling ratio and signal-to-noise ratio (SNR). Simulated and measured data experiments were conducted to demonstrate the performance of imaging quality of the proposed method.


  • Ambiguity Function based High-Order Translational Motion Compensation

    by Zhenyu Zhuo, Lan Du, Xiaofei Lu, Ke Ren, Lu Li

    Micro-Doppler feature plays an important role in cone-shaped target recognition and parameter estimation. However, translational motion causes shifting, tilting, and aliasing in micro-Doppler spectrum and influences the extraction of micro-Doppler features. Most existing methods do not consider higher-order translation compensation. In this article, a high-order translational motion compensation method is proposed. First, the high-order ambiguity function (HAF) of a periodical polynomial phase signal exhibits periodicity along the lag axis and reaches the peak concentration when the lag is equal to the micro-motion period. The micro-motion period can be obtained by searching the peak concentration of echo's HAF. Then, the translational jerk and acceleration are respectively compensated by searching the peaks of the second- and third-order HAFs. Lastly, convolution is operated on the frequency spectrum to find the symmetric center located at the Doppler shift frequency, and the residual translational velocity can be estimated. Experimental results based on electromagnetic computation data show that the proposed method can accurately achieve high-order translational compensation.


  • Precession Parameter Estimation of Warhead With Fins Based on Micro-Doppler Effect and Radar Network

    by Rongzheng Zhang, Yong Wang, Chunmao Yeh, Xiaofei Lu

    Precession is a typical micro-motion of a ballistic missile warhead in the middle of the trajectory. The modulation of the narrow-band radar echo by the precession warhead is mainly reflected in the micro-Doppler effect, which is manifested as the Doppler frequency of the echo periodically changing with time. In the paper, an estimation method of precession parameters for warhead with fins using time-frequency analysis and dual radars is proposed. The estimated parameters include precession angle, spinning angular velocity and coning angular velocity. The geometric and signal model of the precession warhead are established and the analytical expression of the Doppler frequency is deduced. Then, the precession parameters are obtained using the dual view angles of dual radars in this paper. Simulation experimental results illustrate the effectiveness of the proposed method.


  • Open Access

    A Novel Optimal Time Window Determination Approach for ISAR Imaging of Ship Targets

    by Rui Cao, Yong Wang, Chunmao Yeh, Yun Zhang, Xiaofei Lu

    A well-focused and high-resolution inverse synthetic aperture radar (ISAR) image of ship target is appropriate to the target classification and recognition, especially for the ISAR image presenting the ship's top or side view. For this purpose, a novel optimal time window determination (OTWD) approach based on the imaging projection plane (IPP) selection is proposed in this article. First, the proposed OTWD approach determines the view of ship target in the ISAR image via the IPP selection. Then, the Doppler frequency is estimated via the time-frequency distribution (TFD) calculation for the azimuth signal of one scatterer. For the case of one range bin with multiple scatterers, a novel TFD extraction approach based on the clustering analysis is proposed. Afterward, the Doppler frequency is analyzed with sharpness to determine the optimal imaging time window, which can ensure the high image quality and optimal azimuth resolution. The proposed OTWD method is verified as accurate and effective with the experiments on simulated and real measured data.