Yunsheng Qian

Prof. Yunsheng Qian


Research Focus & Metrics

Academic Career Length

22Years (2005-Present)

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

Representative Publications

View All Publications
2 November 2024 in Scientific Reports
Verified
Journal Article
Open access
PDF file
WPWenwen Pan
Wenwen Pan
XSXiaofei Sun
Xiaofei Sun
YWYilun Wang
Yilun Wang
YCYang Cao
Yang Cao
YLYizheng Lang
Yizheng Lang
YQYunsheng Qian
Yunsheng Qian
Detecting defects on photovoltaic panels using electroluminescence images can significantly enhance the production quality of these panels. Nonetheless, in the process of defect detection, there often arise instances of missed detections and false alarms due to the close resemblance between embedded defect features and the intricate background information. To tackle this challenge, we propose an Adaptive Complementary Fusion (ACF) module designed to intelligently integrate spatial and channel information. This module is seamlessly integrated into YOLOv5 for detecting defects on photovoltaic panels, aiming primarily to enhance model detection performance, achieve model lightweighting, and accelerate detection speed. In order to validate the efficacy of the proposed module, we conducted experiments using a dataset comprising 4500 electroluminescence images of photovoltaic panels. Compared to the cutting-edge detection capability of YOLOv8, our YOLO-ACF method exhibits enhancements of 5.2, 0.8, and 2.3 percentage points in R, mAP50, and mAP50-95, respectively. In contrast to the lightest and fastest YOLOv5, YOLO-ACF achieves reductions of 12.9%, 12.4%, and 4.2% in parameters, weight, and time, respectively, while simultaneously boosting FPS by 5%. Through qualitative and quantitative comparisons with various alternative methods, we demonstrate that our YOLO-ACF strikes a good balance between detection performance, model complexity, and detection speed for defect detection on photovoltaic panels. Moreover, it demonstrates remarkable versatility across a spectrum of defect types.

17 October 2018 in IEEE Photonics Journal
Verified
Journal Article
Open access
YWYan Wang
Yan Wang
YQYunsheng Qian
Yunsheng Qian
XKXiangyu Kong
Xiangyu Kong
For the detection of weak signals, photon counting technology has attracted much attention. In this paper, a method based on photon counting is experimentally demonstrated to realize corona discharge detection at solar-blind ultraviolet (UV) wavelength. In our setup, a solar-blind UV optical filter and a Te-Cs cathode are used to eliminate the effects of light with undesired wavelengths, i.e., outside of the solar-blind band. A 2-stage microchannel plate (MCP) with a high voltage can achieve a gain of up to 106. A phosphor screen coupled with a CMOS can capture complete images without image distortion. By using an ICMOS, the setup can detect single-photon events. Then, applying photon counting statistics, a theoretical probability model of the photon numbers is obtained. Meanwhile, through analyzing the structure of the photon events considering both pixel and temporal resolution, two photon counting algorithms are proposed. Through experiments with an UV light source, the algorithm based on temporal resolution is proved to be more accurate. Finally, through an experiment with a corona discharge device, a solar-blind photon counting image is captured, and the photon number is calculated through the proposed counting algorithm. The probability curve of the practical photon numbers is acquired via several experiments carried out at the same conditions, which proves the accuracy of the setup and algorithm when compared with the theoretical probability model. Therefore, the proposed method can provide a scientific evaluation for corona detection.

5 November 2020 in Journal of Applied Physics
Verified
Journal Article
Closed access
PDF file
YZYijun Zhang
Yijun Zhang
KZKaimin Zhang
Kaimin Zhang
SLShiman Li
Shiman Li
SLShan Li
Shan Li
YQYunsheng Qian
Yunsheng Qian
YZYugang Zeng
Yugang Zeng
The surface Cs–O activation process directly determines quantum efficiency and stability of negative-electron-affinity photocathodes. To investigate the effects of excessive Cs and O supply on activation and to explore a more effective Cs–O activation recipe, Cs–O activation experiments of GaAs(100) photocathodes are carried out based on the current-driven solid Cs and O dispensers. By a comparison of differences in activation photocurrent, quantum efficiency, and photocurrent decay, it is found that the recipe of excessive O and non-excessive Cs is not suitable for activating GaAs photocathodes, while the recipe of continuous and completely excessive Cs along with intermittent and non-excessive O can achieve the most excellent photoemission performance, including the highest quantum efficiency in the long-wave threshold region and best stability under intense light irradiation after activation. Furthermore, this improved activation recipe with the least Cs–O alternating cycles is easier to operate. Combined with density functional calculations and dipole layer model, it is found that the activation recipe of completely excessive Cs and non-excessive O can form effective dipoles to the greatest extent, and avoid the direct interaction between As atoms and O atoms to form As–O–Ga oxides on the GaAs(100) reconstructed surface.

3 May 2019 in Journal of Real-Time Image Processing
Verified
Journal Article
Closed access
PDF file
XTXiaodong Tang
Xiaodong Tang
YQYunsheng Qian
Yunsheng Qian
XKXiangyu Kong
Xiangyu Kong
HWHonggang Wang
Honggang Wang

5 December 2022 in Journal of the Optical Society of America A
Verified
Journal Article
Closed access
YLYi-Zheng Lang
Yi-Zheng Lang
YQYunsheng Qian
Yunsheng Qian
XKXiang-Yu Kong
Xiang-Yu Kong
JZJing-Zhi Zhang
Jing-Zhi Zhang
YWYi-Lun Wang
Yi-Lun Wang
YCYang Cao
Yang Cao
Aiming to solve the problem of low-light-level (LLL) images with dim overall brightness, uneven gray distribution, and low contrast, in this paper, we propose an effective LLL image enhancement method based on the guided filter and multi-scale fusion for contrast enhancement and detail preservation. First, a base image and detail image(s) are obtained by using the guided filter. After this procedure, the base image is processed by a maximum entropy-based Gamma correction to stretch the gray level distribution. Unlike the existing methods, we enhance the detail image(s) based on the guided filter kernel, which reflects the image area information. Finally, a new method is proposed to generate a sequence of artificial images to adjust the brightness of the output, which has a better performance in image detail preservation compared with other single-input algorithms. Experiments show that the proposed method can provide a more significant performance in enhancing contrast, preserving details, and maintaining the natural feeling of the image than the state of the art.

2 September 2011 in physica status solidi (c)
Verified
Journal Article
Closed access
PDF file
JQJianliang Qiao
Jianliang Qiao
BCBenkang Chang
Benkang Chang
YQYunsheng Qian
Yunsheng Qian
XWXiaohui Wang
Xiaohui Wang
BLBiao Li
Biao Li
XFXiaoqian Fu
Xiaoqian Fu