Siyu Wang

Dr. Siyu Wang

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
5 December 2025 in Remote Sensing
Verified
Journal Article
Open access
PDF file
ZZZhanhong Zhuo
Zhanhong Zhuo
RLRuitao Lu
Ruitao Lu
YYYongxiang Yao
Yongxiang Yao
SWSiyu Wang
Siyu Wang
Zhi ZhengZhi Zheng
Zhi Zheng
XYXiaogang Yang
Xiaogang Yang
Detecting small objects from UAV-captured aerial imagery is a critical yet challenging task, hindered by factors such as small object size, complex backgrounds, and subtle inter-class differences. Single-modal methods lack the robustness for all-weather operation, while existing multimodal solutions are often too computationally expensive for deployment on resource-constrained UAVs. To this end, we propose TAF-YOLO, a lightweight and efficient multimodal detection framework designed to balance accuracy and efficiency. First, we propose an early fusion module, the Two-branch Adaptive Fusion Network (TAFNet), which adaptively integrates visible and infrared information at both pixel and channel levels before the feature extractor, maximizing complementary data while minimizing redundancy. Second, we propose a Large Adaptive Selective Kernel (LASK) module that dynamically expands the receptive field using multi-scale convolutions and spatial attention, preserving crucial details of small objects during downsampling. Finally, we present an optimized feature neck architecture that replaces PANet’s bidirectional path with a more efficient top-down pathway. This is enhanced by a Dual-Stream Attention Bridge (DSAB) that injects high-level semantics into low-level features, improving localization without significant computational overhead. On the VEDAI benchmark, TAF-YOLO achieves 67.2% mAP50, outperforming the CFT model by 2.7% and demonstrating superior performance against seven other YOLO variants. Our work presents a practical and powerful solution that enables real-time, all-weather object detection on resource-constrained UAVs.

23 April 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
QLQingge Li
Qingge Li
XYxiaogang yang
xiaogang yang
JFJiwei Fan
Jiwei Fan
RLRuitao Lu
Ruitao Lu
BTBin Tang
Bin Tang
SSShuang Su
Shuang Su
Cross-view geolocalization of unmanned aerial vehicles (UAVs) is a challenging task due to the positional discrepancies and uncertainties in scale and distance between UAVs and satellite views. Existing transformer-based geolocalization methods mainly use encoders to mine image contextual information. However, these methods have some limitations when dealing with scale changes in cross-view images. Therefore, we present an effective transformer-based Siamese network tailored for UAV geolocalization, called GeoFormer. First, an efficient transformer feature extraction network was designed, which utilizes linear attention to reduce the computational complexity and improve the computational efficiency of the network. Among them, we designed an efficient separable perceptron module based on depthwise separable convolution, which can effectively reduce the computational cost while improving the feature representation of the network. Second, we proposed a multiscale feature aggregation module, which deeply fuses salient features at different scales through a feedforward neural network to generate global feature representations with rich semantics, which improves the model's ability to capture image details and represent robust features. Additionally, we designed a semantic-guided region segmentation module, which utilizes a k -modes clustering algorithm to divide the feature map into multiple regions with semantic consistency and performs feature recognition within each semantic region to improve the accuracy of image matching. Finally, we designed a hierarchical reinforcement rotation matching strategy to achieve accurate UAV geolocalization based on the retrieval results of UAV view query satellite images using SuperPoint keypoints extraction and LightGlue rotation matching. According to the experimental results, our method effectively achieves UAV geolocalization.

3 October 2024 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
ZZZhengjie Zhu
Zhengjie Zhu
XYxiaogang yang
xiaogang yang
RLRuitao Lu
Ruitao Lu
TSTong Shen
Tong Shen
TZTao Zhang
Tao Zhang
SWSiyu Wang
Siyu Wang
It is well known that the diverse causes of low-light images challenge the adaptability of enhancement algorithms in uncertain environments. Most deep learning-based algorithms only learn single illuminance estimation or mapping relationship, which inhibit the generalization ability of the model. To address this, we propose a novel multi-illumination estimation framework based on ghost imaging theory, dubbed Ghillie. Specifically, we consider low-light enhancement as a re-imaging process for objects in dark scenes. First, the light modulation network (LMN) is designed to modulate a series of estimated lights following a normal light distribution. These lights "illuminate" the low-light image and the enhanced illuminance image can be reconstructed by a differential ghost imaging algorithm. Then, a gradient-guided denoising network (GDN) is constructed to eliminate noise and enhance details. Finally, we employ the color adaption network (CAN) to restore the color degradation. Additionally, a novel mean structural similarity loss (AM-SSIM) is proposed to guide the model to address the uneven image illumination. The qualitative and quantitative experimental results show that our enhanced methods outperform state-of-the-art methods on the vast majority of publicly available datasets.

1 January 2025 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
SWSiyu Wang
Siyu Wang
XYxiaogang yang
xiaogang yang
RLRuitao Lu
Ruitao Lu
DZDingwen Zhang
Dingwen Zhang
WXWeiying Xie
Weiying Xie
ZZZhenyu Zhang
Zhenyu Zhang

13 April 2023 in Drones
Verified
Journal Article
Open access
PDF file
QLQingge Li
Qingge Li
XYxiaogang yang
xiaogang yang
RLRuitao Lu
Ruitao Lu
JFJiwei Fan
Jiwei Fan
SWSiyu Wang
Siyu Wang
ZQZhen Qin
Zhen Qin
Post-disaster search and rescue is critical to disaster response and recovery efforts and is often conducted in hazardous and challenging environments. However, the existing post-disaster search and rescue operations have problems such as low efficiency, limited search range, difficulty in identifying the nature of the target, and wrong target location. Therefore, this study develops an air–ground integrated intelligent cognition visual enhancement system based on a UAV (VisionICE). The technique combines a portable AR display device, a camera-equipped helmet, and a quadcopter UAV for efficient patrols over a wide area. First, the system utilizes wireless image sensors on the UAV and helmet to capture images from the air and ground views. Using the YOLOv7 algorithm, the cloud server calculates and analyzes these visual data to accurately identify and detect targets. Lastly, the AR display device obtains real-time intelligent cognitive results. The system allows personnel to simultaneously acquire air and ground dual views and achieve brilliant cognitive results and immersive visual experiences in real time. The findings indicate that the system demonstrates significant recognition accuracy and mobility. In contrast to conventional post-disaster search and rescue operations, the system can autonomously identify and track targets of interest, addressing the difficulty of a person needing help to conduct field inspections in particular environments. At the same time, the system can issue potential threat or anomaly alerts to searchers, significantly enhancing their situational awareness capabilities.

6 November 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
ZZZhanhong Zhuo
Zhanhong Zhuo
RLRuitao Lu
Ruitao Lu
YYYongxiang Yao
Yongxiang Yao
SWSiyu Wang
Siyu Wang
Zhi ZhengZhi Zheng
Zhi Zheng
XYXiaogang Yang
Xiaogang Yang
Detecting small objects from UAV-captured aerial imagery is a critical yet challenging task, hindered by factors such as small object size, complex backgrounds, and subtle inter-class differences. Single-modal methods lack the robustness for all-weather operation, while existing multimodal solutions are often too computationally expensive for deployment on resource-constrained UAVs. To this end, we propose TAF-YOLO, a lightweight and efficient multimodal detection framework designed to balance accuracy and efficiency. First, an early fusion module, the Two-branch Adaptive Fusion Network (TAFNet), which adaptively integrates visible and infrared information at both pixel and channel levels before the feature extractor, maximizing complementary data while minimizing redundancy. Second, a Large Adaptive Selective Kernel (LASK) module that dynamically expands the receptive field using multi-scale convolutions and spatial attention, preserving crucial details of small objects during downsampling. Finally, an optimized feature neck architecture that replaces PANet's bidirectional path with a more efficient top-down pathway. This is enhanced by a Dual-Stream Attention Bridge (DSAB) that injects high-level semantics into low-level features, improving localization without significant computational overhead. On the VEDAI benchmark, TAF-YOLO achieves 67.2% mAP50, outperforming the CFT model by 2.7% and demonstrating superior performance against seven other YOLO variants. Our work presents a practical and powerful solution that enables real-time, all-weather object detection on resource-constrained UAVs.