Zhigang Lv

Dr. Zhigang Lv


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
17 March 2026 in ISA Transactions
Verified
Journal Article
Closed access
CWChu Wang
Chu Wang
SZShuang Zhang
Shuang Zhang
PWPeng Wang
Peng Wang
ZLZhongliang Li
Zhongliang Li
RORachid Outbib
Rachid Outbib
MDManfeng Dou
Manfeng Dou

2 February 2026 in Entropy
Verified
Journal Article
Open access
PDF file
RDRuohai Di
Ruohai Di
HFHao Fan
Hao Fan
HFHanxiao Feng
Hanxiao Feng
ZLZhigang Lv
Zhigang Lv
LSLei Shu
Lei Shu
RQRuoyu Qian
Ruoyu Qian
The detection of defects in Printed Circuit Boards (PCBs) is a critical yet challenging task in industrial quality control, characterized by the prevalence of small targets and complex backgrounds. While deep learning models like YOLOv5 have shown promise, they often lack the ability to quantify predictive uncertainty, leading to overconfident errors in challenging scenarios—a major source of false alarms and reduced reliability in automated manufacturing inspection lines. From a Bayesian perspective, this overconfidence signifies a failure in probabilistic calibration, which is crucial for trustworthy automated inspection. To address this, we propose MFE-YOLO, a Bayesian-enhanced detection framework built upon YOLOv5 that systematically integrates uncertainty-aware mechanisms to improve both accuracy and operational reliability in real-world settings. First, we construct a multi-background PCB defect dataset with diverse substrate colors and shapes, enhancing the model’s ability to generalize beyond the single-background bias of existing data. Second, we integrate the Convolutional Block Attention Module (CBAM), reinterpreted through a Bayesian lens as a feature-wise uncertainty weighting mechanism, to suppress background interference and amplify salient defect features. Third, we propose a novel FIoU loss function, redesigned within a probabilistic framework to improve bounding box regression accuracy and implicitly capture localization uncertainty, particularly for small defects. Extensive experiments demonstrate that MFE-YOLO achieves state-of-the-art performance, with mAP@0.5 and mAP@0.5:0.95 values of 93.9% and 59.6%, respectively, outperforming existing detectors, including YOLOv8 and EfficientDet. More importantly, the proposed framework yields better-calibrated confidence scores, significantly reducing false alarms and enabling more reliable human-in-the-loop verification. This work provides a deployable, uncertainty-aware solution for high-throughput PCB inspection, advancing toward trustworthy and efficient quality control in modern manufacturing environments.

1 September 2021 in Journal of Physics: Conference Series
Verified
Conference Paper
Open access
PDF file
MZMeng Zhou
Meng Zhou
ZLZhigang Lv
Zhigang Lv
YLYe Li
Ye Li
RDRuoHai Di
RuoHai Di
HZHongjie Zhu
Hongjie Zhu
KHKeqing Hao
Keqing Hao

24 September 2022 in Entropy
Verified
Journal Article
Open access
PDF file
ZLZhigang Lv
Zhigang Lv
YCYiwei Chen
Yiwei Chen
RDRuohai Di
Ruohai Di
HWHongxi WANG
Hongxi WANG
XSXiaojing Sun
Xiaojing Sun
XLXiaoyan Li
Xiaoyan Li
The Bayesian Network (BN) structure learning algorithm based on dynamic programming can obtain global optimal solutions. However, when the sample cannot fully contain the information of the real structure, especially when the sample size is small, the obtained structure is inaccurate. Therefore, this paper studies the planning mode and connotation of dynamic programming, restricts its process with edge and path constraints, and proposes a dynamic programming BN structure learning algorithm with double constraints under small sample conditions. The algorithm uses double constraints to limit the planning process of dynamic programming and reduces the planning space. Then, it uses double constraints to limit the selection of the optimal parent node to ensure that the optimal structure conforms to prior knowledge. Finally, the integrating prior-knowledge method and the non-integrating prior-knowledge method are simulated and compared. The simulation results verify the effectiveness of the method proposed and prove that the integrating prior knowledge can significantly improve the efficiency and accuracy of BN structure learning.

1 January 2023
Verified
Conference Paper
Open access
PDF file
JLJianheng Li
Jianheng Li
PWPeng Wang
Peng Wang
ZLZhigang Lv
Zhigang Lv
LLLiangliang Li
Liangliang Li
The retrieval of paper form data is large, and the storage cost is high and wasteful, and the digital reproduction of form data has become more and more common. Due to the immaturity of the early watchmaking technology of the mechanical craft card, there were problems such as dotted vertical frame lines, misplaced frame lines, and table spreads. First, a table document tilt correction algorithm based on straight line detection is used, which is convenient for better realization of table extraction and positioning and identification of effective information. Homogeneous, according to the characteristics of irregular table style, an irregular table extraction algorithm based on double histogram statistical clustering is designed to achieve precise positioning of table areas. Then, according to the sample characteristics of the cross-page table, a cross-page table splicing algorithm based on local characteristics is designed to ensure the integrity of the data. Finally, by extracting irregular table cell images and bi-histogram projection, a nonregular table cell extraction algorithm based on image statistics clustering is realized.