Lingyun Kong

Lingyun Kong

Auto-generated profile, based on publicly available information

Lingyun Kong

School of Electronic Information, Xijing University, Xian, Shaanxi, China

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
25 April 2025 in IEEE Access
Verified
Journal Article
Open access
ZXZhipeng Xue
Zhipeng Xue
LKLingyun Kong
Lingyun Kong
HWHaiyang Wu
Haiyang Wu
JCJiale Chen
Jiale Chen
Fire and smoke detection is an important measure to ensure the safety of people’s lives and property, as well as a crucial link in maintaining ecological balance and supporting scientific research. Traditional object detection methods rely more on manually designed features and rules. Although they are relatively simple to implement, their performance is limited in complex and variable practical applications. In contrast, deep learning-based methods can automatically learn deep features in data and have higher accuracy and stronger generalization ability. However, complex backgrounds, large environmental changes, and data requirements pose great challenges to high-precision outdoor smoke detection. To address these issues, this paper proposes an improved model, YOLOV11-DH3, based on YOLOV11. In this paper, the core DCN2 (Deformable Convolutional Networks2) of the YOLOV11 Head is replaced with the DCN3 module to form a new detection head. In addition, the loss function CIOU in YOLOV11 is replaced with IOU to consider the irregular shape of fire and smoke and the problem of multi-scale targets. To evaluate the performance of the algorithm, comprehensive experiments were conducted on two distinct datasets: a public fire and smoke dataset provided by Baidu Paddle featuring close-range views and a public wildfire smoke dataset from the YOLO official website with distant outdoor perspectives. The experimental results show that on the Baidu Paddle dataset, the average accuracy of the model is improved by 1.4% compared to the original model, reaching 58%, the F1 score is improved by 2%, reaching 58%, with a precision of 91.6% and recall of 90%. Our cross-dataset analysis provides valuable insights into model performance across different detection scenarios. The proposed model demonstrates the ability to accurately detect fire and smoke in complex backgrounds, and this progress is of great significance for protecting people’s lives and maintaining ecological balance.

16 December 2024 in IEEE Access
Verified
Journal Article
Open access
HWHaiyang Wu
Haiyang Wu
LKLingyun Kong
Lingyun Kong
DLDenghui Liu
Denghui Liu
Road defect detection is vital for road maintenance but remains challenging due to the complexity of backgrounds, low resolution, and crack similarity. This paper introduces YOLOv8-VOS(VOS means ’vanillaNet+ODConv+SEAttention’), an enhanced road crack detection algorithm that incorporates an improved Vanilla Net backbone with Squeeze-and-Excitation (SE) attention and ODConv modules. The loss function is replaced with WIoU to better balance bounding box regression. Experiments on the RDD2022 dataset demonstrate a 2% improvement in average accuracy over the original YOLOv8, achieving 53.7%. The proposed model effectively identifies road cracks in complex traffic backgrounds, contributing to safer and more efficient road maintenance.

26 July 2021 in IEEE Access
Verified
Journal Article
Open access
YCYuezhe Chen
Yuezhe Chen
LKLingyun Kong
Lingyun Kong
YWYang Wang
Yang Wang
DKDezhi Kong
Dezhi Kong
Aspect-level sentiment classification aims to solve the problem, which is to judge the sentiment tendency of each aspect in a sentence with multiple aspects. Previous works mainly employed Long Short-Term Memory (LSTM) and Attention mechanisms to fuse information between aspects and sentences, or to improve large language models such as BERT to adapt aspect-level sentiment classification tasks. The former methods either did not integrate the interactive information of related aspects and sentences, or ignored the feature extraction of sentences. This paper proposes a novel multi-grained attention representation with ALBERT (MGAR-ALBERT). It can learn the representation that contains the relevant information of the sentence and the aspect, while integrating it into the process of sentence modeling with multi granularity, and finally get a comprehensive sentence representation. In Masked LM (MLM) task, in order to avoid the influence of aspect words being masked in the initial stage of the pre-training, the noise linear cosine decay is introduced into n-gram. We implemented a series of comparative experiments to verify the effectiveness of the method. The experimental results show that our model can achieve excellent results on Restaurant dataset with numerous number of parameters reduced, and it is not inferior to other models on Laptop dataset.

1 January 2021 in Complexity
Verified
Journal Article
Open access
PDF file
YWYang Wang
Yang Wang
ZWZhen Wang
Zhen Wang
LKLingyun Kong
Lingyun Kong
For synchronization of a class of chaotic systems in the presence of nonvanishing uncertainties, a novel time‐varying gain observer‐based sliding mode control is proposed. First, a novel time‐varying gain disturbance observer (TVGDO) is developed to estimate the uncertainties. Then, by using the output of TVGDO to modify sliding mode control (SMC), a new TVGDO‐based SMC scheme is developed. Although the observation and control precision of conventional fixed gain disturbance observer‐based control (FGDOC) for chaotic systems can be guaranteed by a high observer gain, the undesirable spike problem may be caused by the high gain if the initial values of estimate and true states are not equal. The most attractive feature of this work is that the newly proposed TVGDO can eliminate the spike problem by developing a time‐varying gain scheme. Finally, the effectiveness of the proposed method is demonstrated by the numerical simulation.

12 January 2026 in IEEE Access
Verified
Journal Article
Open access
ZXZhipeng Xue
Zhipeng Xue
LKLingyun Kong
Lingyun Kong
HWHaiyang Wu
Haiyang Wu
Unmanned aerial vehicle (UAV) target detection has become increasingly critical across diverse applications including surveillance, search and rescue operations, and autonomous navigation. While the YOLO series models have demonstrated exceptional efficiency and real-time performance in object detection tasks, existing YOLOv11 models still exhibit limitations in complex UAV scenarios. To address these challenges, this paper proposes YOLOv11-HNU, an enhanced detection model incorporating two key innovations. First, we integrate the HFFBlock (Hierarchical Feature Fusion Block) module into the neck of YOLOv11 to enhance multi-scale feature fusion capabilities. Second, we implement the UIOU (Unified Intersection over Union) loss function to optimize training dynamics. Experimental results on the Roboflow UAVs dataset show that the HFFBlock module alone improves mAP@0.5 from 0.975 to 0.978, while the complete YOLOv11-HNU model maintains this optimal performance. On the more challenging Roboflow Drone-Kaggle-Data dataset, the HFFBlock module increases mAP@0.5 from 0.944 to 0.950, while the UIOU loss function alone achieves the best single improvement to 0.957. The final YOLOv11-HNU model, combining both innovations, achieves 0.951 mAP@0.5, representing a solid 0.7 percentage point improvement over the baseline. Comprehensive ablation studies reveal that the improvement strategies exhibit dataset-dependent characteristics, with HFFBlock providing consistent gains across datasets and UIOU showing particular effectiveness in complex detection scenarios. The proposed YOLOv11-HNU effectively enhances UAV target detection accuracy, providing a valuable technical solution for real-world UAV detection applications.

20 November 2020 in Mathematical Problems in Engineering
Verified
Journal Article
Open access
PDF file
YWYang Wang
Yang Wang
ZWZhen Wang
Zhen Wang
DKDezhi Kong
Dezhi Kong
LKLingyun Kong
Lingyun Kong
YQYukun Qiao
Yukun Qiao
The Euler dynamical equation which describes the attitude motion of a rigid body will exhibit very complex dynamic behaviors under the action of different external torques. Many special types of new chaotic attractors are presented, including hidden attractors, double-body-double-core chaotic attractors, and single-body-three-core-tree-wing chaotic attractors. The position of equilibrium points in several typical cases of the Euler dynamic equation is solved, and the stability of linearized equation at each equilibrium point and its influence on the formation of the chaotic attractor are analyzed. An improved nonlinear relay control law based on Euler angle feedback is developed to stabilize a new chaotic spacecraft attitude motion to an appointed equilibrium point or a periodic orbit.