XSXu Sun

Xu Sun

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xu Sun has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Unified diffusion‐based object detection in multi‐modal and low‐light remote sensing images

    by Xu Sun, Yinhui Yu, Qing Cheng

    Remote sensing object detection remains a challenge under complex conditions such as low light, adverse weather, modality attacks or losses. Previous approaches typically alleviate this problem by enhancing visible images or leveraging multi‐modal fusion technologies. In view of this, the authors propose a unified framework based on YOLO‐World that combines the advantages of both schemes, achieving more adaptable and robust remote sensing object detection in complex real‐world scenarios. This framework introduces a unified modality modelling strategy, allowing the model to learn abundant object features from multiple remote sensing datasets. Additionally, a U‐fusion neck based on the diffusion method is designed to effectively remove modality‐specific noise and generate missing complementary features. Extensive experiments were conducted on four remote sensing image datasets: Multimodal VEDAI, DroneVehicle, unimodal VisDrone and UAVDT. This approach achieves average precision scores of 50.5, 55.3, 25.1, and 20.7, which outperforms advanced multimodal remote sensing object detection methods and low‐light image enhancement techniques.


  • Low-Rank Multimodal Remote Sensing Object Detection With Frequency Filtering Experts

    by Xu Sun, Yinhui Yu, Qing Cheng

    Visible-infrared object detection for remote sensing images plays an important role in the unmanned aerial vehicle (UAV) around-the-clock application. Most of the existing work focuses on designing complex network architectures to fuse complementary features, while few methods consider computational complexity and susceptibility against modality attacks, limiting the deployment of state-of-the-art frameworks. In this paper, we present a low-rank multimodal object detection approach with frequency filtering experts, called LF-MDet, which is based on the advanced DINO (detection transformer with improved denoising anchor boxes) framework. This approach achieves more accurate detection with fewer computational resources. In particular, when the specific modality is attacked or missing, our method still maintains higher robustness towards such pervasive perturbations. Specifically, we propose a low-rank enhancement technology and a dynamic illumination-aware mask module to enable a single backbone network in the form of batch formulation to unbiasedly and compatibly extract multimodal features. Furthermore, we design a lightweight frequency expert encoder from the frequency domain perspective to efficiently fuse complementary features by filtering out amplitude noise components and mixing feature tokens. Extensive experiments are conducted on the multimodal remote sensing object detection datasets, VEDAI and DroneVehicle. The results demonstrate the superiority of the proposed approach over advanced multimodal remote sensing object detectors. Compared to the baseline method, our LF-MDet effectively reduces the floating point of operations (FLOPs) by approximately 65% while improving detection accuracy. The code is available at https://github.com/cq100/LF-MDet.


  • LED recognition method based on deep learning in UAV optical camera communication

    by Xu Sun, YinHui Yu, Qing Cheng

    Optical camera communication (OCC) is a potential technology in unmanned aerial vehicle (UAV) communication scenes, which can be used as a complementary scheme for radio frequency (RF) technology to effectively mitigate electromagnetic interference. The UAV OCC system still has the problems of low reliability, poor robustness, and long latency. In this paper, we propose the n -ary image that contains binary image information with different thresholds for the LED area. In addition, the multi-spectrum fast recognition (MFR) algorithm based on deep learning is designed. The MFR algorithm combines the frequency channel attention (FCA) and the involution convolution operator, which take the n -ary image as input to accurately identify the LED state. The experiment results show that the proposed method can reduce the bit error rate (BER) and latency of the UAV OCC system.


  • Open Access

    An LED Detection and Recognition Method Based on Deep Learning in Vehicle Optical Camera Communication

    by Xu Sun, Wenxiao Shi, Qing Cheng, Wei Liu, Zhuo Wang, Jiadong Zhang

    In the Vehicle to Vehicle (V2V) communication based on Optical Camera Communication (OCC), optical signals are transmitted using LED arrays and received employing cameras. In a complex scene, how to accurately detect and recognize LEDs in real time remains a problem. To solve this problem, this paper designs an end-to-end network based on You Only Look Once version 5 (YOLOv5) object detection model, which can precisely detect the LED array position in real time and alleviate motion blur simultaneously. Further, we propose an LED segmentation recognition method, which is beneficial to more reliable LED status recognition. It allows more light sources to be used for communication, which can effectively improve data rate in the vehicle OCC system. The effectiveness of our method is demonstrated by theoretical analysis and experiments in real traffic scenes. Our code is available at https://github.com/cq100/D2Net.


  • Design, analysis and testing of a novel linear collaborative pump

    by Li Yang, Jia Yingmin, Sun Xu, Jiao Zongxia

    In order to eliminate severe internal friction and leakage in a conventional axial piston pump, a linear piston pump is designed. Certain spool valve is introduced to rectify the flow of the pump, and a modular collaborative rectification structure is proposed. The kinematic flow is analyzed to verify the basic design principle of the pump. The preliminary mechanical structure is designed for the first-generation prototype. The dynamic characteristics of a functional pump module are modelled, simulated and analyzed. The experiment on the prototype validated the feasibility and controllability of the pump.