Shuai Zhao

Dr. Shuai Zhao

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
19 February 2025 in Remote Sensing
Verified
Journal Article
Open access
PDF file
SZSicheng Zhou
Sicheng Zhou
YLYang Lei
Yang Lei
HLHuiting Liu
Huiting Liu
CZChongging Zhou
Chongging Zhou
嘉刘嘉诚 刘
嘉诚 刘
KWKeyi Wang
Keyi Wang
The timely and accurate detection of unidentified drones is vital for public safety. However, the unique characteristics of drones in complex environments and the varied postures they may adopt during approach present significant challenges. Additionally, deep learning algorithms often require large models and substantial computational resources, limiting their use on low-capacity platforms. To address these challenges, we propose LAMS-YOLO, a lightweight drone detection method based on linear attention mechanisms and adaptive downsampling. The model’s lightweight design, inspired by CPU optimization, reduces parameters using depthwise separable convolutions and efficient activation functions. A novel linear attention mechanism, incorporating an LSTM-like gating system, enhances semantic extraction efficiency, improving detection performance in complex scenarios. Building on insights from dynamic convolution and multi-scale fusion, a new adaptive downsampling module is developed. This module efficiently compresses features while retaining critical information. Additionally, an improved bounding box loss function is introduced to enhance localization accuracy. Experimental results demonstrate that LAMS-YOLO outperforms YOLOv11n, achieving a 3.89% increase in mAP and a 9.35% reduction in parameters. The model also exhibits strong cross-dataset generalization, striking a balance between accuracy and efficiency. These advancements provide robust technical support for real-time drone monitoring.

10 April 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
SZSicheng Zhou
Sicheng Zhou
LYLei Yang
Lei Yang
HLHuiting Liu
Huiting Liu
CZChongqin Zhou
Chongqin Zhou
JLJiacheng Liu
Jiacheng Liu
KWKeyi Wang
Keyi Wang
The timely and accurate detection of unidentified drones is crucial for public safety. However, challenges arise due to background noise in complex environments and limited feature representation of small, distant targets. Additionally, deep learning algorithms often demand substantial computational resources, limiting their use on low-capacity platforms. To address these issues, we propose LMWP-YOLO, a lightweight drone detection method that incorporates a multidimensional collaborative attention mechanism and multi-scale fusion. Inspired by ARM CPU efficiency optimizations, the model uses depthwise separable convolutions and efficient activation functions to reduce parameter size. The neck structure is enhanced with a collaborative attention mechanism and multi-scale fusion, improving feature representation. An optimized loss function refines bounding box matching for small targets, while a pruning strategy removes redundant filters, boosting computational efficiency. Experimental results show that LMWP-YOLO outperforms YOLO11n, with a 22.07% increase in mAP and a 52.51% reduction in parameters. The model demonstrates strong cross-dataset generalization, balancing accuracy and efficiency. These findings contribute to advancements in small drone target detection.

9 September 2021 in IEEE Journal of Quantum Electronics
Verified
Journal Article
Closed access
YYYu Yang
Yu Yang
SZShuai Zhao
Shuai Zhao
YSYuan Shen
Yuan Shen
LMLingjun Meng
Lingjun Meng
TCTianci Chen
Tianci Chen
KWKeyi Wang
Keyi Wang
The nonlinear effects in optical resonators give a promise for the generation of new laser sources. Here, a whispering gallery mode(WGM) micro-rod resonator is fabricated and the mechanism of transmission between Kerr comb to Raman comb is investigated by experimental results and amended Lugiato-Lefever equation(LLE) model. We show that, numerically and experimentally, the WGM resonator-based optical frequency comb can be selectively accessed from Kerr comb state to Kerr-Raman state via pump and detuning controlling. When the detuning is initially set at red detuning, by simply increasing the pump power, the comb state can lock to the stable Raman soliton comb. This indicates the Raman soliton comb is easier to generate compared to the Kerr soliton comb in silica-based WGM resonator. We wish the experimental results demonstrated here can be used for broadband comb applications.

9 February 2023 in Applied Optics
Verified
Journal Article
Closed access
LZLei Zhang
Lei Zhang
YYYu Yang
Yu Yang
SZShuai Zhao
Shuai Zhao
TCTianci Chen
Tianci Chen
KYKaiyang Yu
Kaiyang Yu
Bolin CaiBolin Cai
Bolin Cai
The hyperspectral full polarization imaging system (HFPIS) based on spatial modulation and liquid crystal tunable filter (LCTF) can modulate the polarization information of the target into the interferogram by means of interference imaging. It has the advantage of rapid imaging of the hyperspectral full polarization information from the target, and has good real time imaging and practicality. Through the spectroscopic imaging mode of a Savart prism, the corresponding interference optical path and imaging system are designed, including a beam expander, spatial modulator, LCTF, focusing system, and imaging sensor. This system can extract the different information from the target and demodulate it so as to obtain the hyperspectral polarization image. The experiment shows that the HFPIS can reveal the texture, contour, and other details of the target in the fog, and has obvious advantages over the traditional intensity imaging methods.

30 May 2023 in Journal of the Optical Society of America A
Verified
Journal Article
Closed access
TCTianci Chen
Tianci Chen
ZKZhaofeng Kang
Zhaofeng Kang
YYYu Yang
Yu Yang
SZShuai Zhao
Shuai Zhao
JZJun Zhang
Jun Zhang
KWKeyi Wang
Keyi Wang
Optical microresonators confine light to small volumes through resonant circulation. Herein, whispering gallery mode (WGM) microresonators have high Q factors among these microresonators, which have significant research value in the fields of fundamental physics research and optoelectronic devices. However, maintaining a very high surface finish on the side of the microresonator is necessary, as is keeping a coupling distance of tens of nanometers between the microresonator and the coupling waveguide. Thus, this makes the fabrication, coupling, and packaging of the microresonator very difficult and seriously hinders the practical application of the microresonator. In this study, the concept of gradient refractive index (GRIN) microresonator is proposed, and the radial GRIN is introduced to change the light direction and form a closed optical path within the microresonator. Herein, the mode field position of the GRIN microresonator is derived from the light transmission equation, and the theoretical result is proved by finite difference time domain (FDTD) simulation. Hence, there are several advantages to using this novel optical microresonator, including its high Q factor, strong coupling stability, and ease of integration.

2 July 2021 in Optics Express
Verified
Journal Article
Open access
PDF file
SZShuai Zhao
Shuai Zhao
YYYu Yang
Yu Yang
YSYuan Shen
Yuan Shen
GCGuangyu Cheng
Guangyu Cheng
YWYushan Wang
Yushan Wang
KWKeyi Wang
Keyi Wang
A microfocus X-ray grating interferometer (MFXGI) is proposed to measure the profile of the X-ray wavefront and slope error of X-ray optical elements. This device consists of a phase grating G1 to modulate the incoming wavefront and an absorption grating G2 as a transmission mask for the position-sensitive detector. The wavefront distortions caused by the deformable mirror were analyzed under operating conditions for in situ investigation of X-ray optical elements. The MFXGI can obtain direct and reflected beams in one recorded image at the same time through a microfocus X-ray source. The direct beam can be used to calculate the parameter errors and spherical shape for error compensation and retrieve the aspherical shape of the height profile. This instrument is expected to be a valuable tool for further technical progress in X-ray adaptive optics and X-ray mirror manufacturing and mounting.