Qian Ma

Qian Ma

Auto-generated profile, based on publicly available information

Qian Ma

Dalian Maritime University

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
21 November 2025 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
XWXite Wang
Xite Wang
YWYinglong Wang
Yinglong Wang
MBMei Bai
Mei Bai
QMQian Ma
Qian Ma
In the field of remote sensing image processing, remote sensing image object detection is a crucial undertaking. However, the existing target detection algorithms have a considerable number of model parameters, which results in a slow detection speed that is not conducive to application deployment and real-time detection. Additionally, due to the background complexity of remote sensing images and the large number of small target objects, the performance of existing algorithmic models applied directly with remote sensing images is unsatisfactory. To address the aforementioned issues, this paper proposes an efficient and lightweight architecture based on YOLOv8: AUD-YOLO. In particular, a convolution module, GEConv, based on improved efficient multi-scale attention (EMA), is proposed to replace the original ordinary convolution at the end of the backbone part. This can enhance the accuracy of target detection in remote sensing images while maintaining a lightweight operation. The utilization of dynamic upsampling operators serves to perform upsampling operations with greater accuracy and to more effectively extract image features pertaining to small target objects. Finally, the issue of sample imbalance is addressed by utilising Wise-iou as the position coordinate loss function of the neural network. In the experiments, we validate the effectiveness and robustness of the method using the public remote sensing datasets NWPU VHR-10 and DIOR. The experimental results demonstrate that the AUD-YOLO model proposed in this paper achieves mAP values of 91.1% and 82.7%, which are 1.6% and 1.5% higher than the mAP indexes of the baseline model, and the number of parameters is reduced by 4.7% compared with the original model. We comprehensively consider the number of model parameters and model detection accuracy to make the model more adept at detecting remote sensing images.

28 January 2025 in IEEE Transactions on Consumer Electronics
Verified
Journal Article
Closed access
XLXinyue Long
Xinyue Long
SGShikai Guo
Shikai Guo
YCYu Chai
Yu Chai
Hui LiHui Li
Hui Li
SJSumaira Ameer Jan
Sumaira Ameer Jan
NQNing Qiao
Ning Qiao
Software vulnerabilities pose significant security threats to modern systems, particularly those involving complex execution sequences and intricate call relationships across multiple execution points.For instance, in a scenario where a software system integrates legacy code in a low-resource programming language like PHP, detecting vulnerabilities becomes challenging due to data scarcity and the complexity of temporal relationships among code fragments. This scarcity hampers the ability to capture critical temporal features essential for identifying vulnerabilities spanning multiple execution points.Consequently, existing approaches face major limitations, including neglecting temporal information in code fragments and lacking sufficient data to enable effective generalization for models in low-resource languages.To address these challenges, we introduce, a novel approach that enhances vulnerability detection in low-resource languages by extracting complex temporal features from code fragments and employing parameter-efficient transfer learning to leverage shared knowledge from resource-rich languages. comprises two key components: 1) Code Vulnerability Detection Component: This component models temporal dependencies by leveraging execution paths extracted from Abstract Syntax Trees (ASTs), capturing both short-term variations and long-term dependencies among code fragments. This enables comprehensive extraction of complex temporal features, significantly enhancing the accuracy of vulnerability detection. 2) Cross-Lingual Transfer Component: This component learns generalizable features from resource-rich languages and efficiently transfers them to low-resource languages. By updating a small number of downstream parameters, it enhances model generalization and achieves precise vulnerability detection. We evaluated using a diverse set of programming languages from publicly available GitHub repositories, employing C as the resource-rich source language and Java, Python, and PHP as relatively low-resource target languages. Experimental results demonstrate that outperforms four state-of-the-art approaches across multiple low-resource languages. Specifically, achieves average improvements of 14.63% in Accuracy, 30.59% in Precision, 37.32% in Recall, and 33.65% in F1-Score score over the best baseline approaches.

20 August 2025 in IEEE Transactions on Consumer Electronics
Verified
Journal Article
Closed access
ZWZhaoqin Wang
Zhaoqin Wang
QMQian Ma
Qian Ma
DCDefu Cui
Defu Cui
SGShikai Guo
Shikai Guo
HLHui Li
Hui Li
QNQiao Ning
Qiao Ning
The missing values pervasively exist in IoT multivariate time series, which severely impairs the interpretability of the data and poses a major problem for advanced analysis and pattern recognition tasks. Existing IoT time series imputation methods typically address missing values through deletion, statistical estimation, or deep-learning-based imputation However, these methods often fail to sufficiently explore temporal information or are multi-stage. In this paper, we introduce a novel model for IoT time series imputation, called AttnWGAIN, which utilizes an attention mechanism to capture both temporal and feature correlations in the data and imputes missing values (MVs) in a generative manner. Specifically, AttnWGAIN is a GAN-based model where the generator leverages a number of diagonal masked attention blocks to reconstruct an IoT complete time series, and the discriminator attempts to distinguish between real (observed) and fake (imputed) elements to enhance imputation accuracy. Furthermore, by training the model with a joint optimization strategy, AttnWGAIN demonstrates superior performance in IoT time series imputation. We conduct extensive experiments on four real-world IoT datasets, and the results quantitatively and qualitatively show that AttnWGAIN outperforms state-of-the-art methods in IoT time series imputation tasks. The source code is available at https://github.com/wzq-come-on/AttnWGAIN.git.

21 May 2025 in IEEE Transactions on Consumer Electronics
Verified
Journal Article
Closed access
ZLZhengdao Li
Zhengdao Li
SGShikai Guo
Shikai Guo
LZLehuan Zhang
Lehuan Zhang
Hui LiHui Li
Hui Li
ZYZhiguo Yang
Zhiguo Yang
QMQian Ma
Qian Ma
Defect prediction tools are designed to assist practitioners in effectively prioritizing the risk files that are most likely to manifest software defects after release, with limited software quality assurance resources. It plays a vital role in code inspection activities. Currently, numerous automated defect prediction methods have been proposed to enhance the efficiency of code review and software quality. However, most previous methods often rely on a substantial number of labeled samples for training. Additionally, when confronted with tasks involving a broad domain span, these methods frequently struggle to effectively identify defective files and lines. To address this issue, we propose a model named DEF-Hunter for rapidly adapting to new file-level and line-level defect predictions when encountering new technical domains. DEF-Hunter consists of two components: defect identification component and transfer learning component. The defect identification component employs a hierarchical attention network to learn the hierarchical structure of the source code, capturing the surrounding code tokens and neighboring lines. It utilizes attention mechanisms to calculate risk scores for code tokens contributing to the prediction of defective files. The transfer learning component employs meta-learning to enhance the model’s cross-domain generalization ability, achieving few-shot line-level defect prediction. The experimental results demonstrate that DEF-Hunter outperforms the state-of-the-art methods in both performance, cost-effectiveness and across metrics such as Balance Accuracy, AUC, MCC, and Recall@Top20%LOC, e.g., compared to state-of-the-art methods, DEF-Hunter improves the Balance Accuracy score by 29%, 29%, 17.49%, 7.32%, and 2.38%, respectively.

27 April 2026 in ACM Transactions on Internet Technology
Verified
Journal Article
Open access
WDWenhui Du
Wenhui Du
CLChenchen Li
Chenchen Li
SYShouyu Yin
Shouyu Yin
HLHaorui Lin
Haorui Lin
HLHui Li
Hui Li
QMQian Ma
Qian Ma
Software fault localization aims to identify faulty elements in a program by analyzing program information and the execution data of test cases. This process plays a crucial role in improving development efficiency, reducing debugging costs, and ensuring reliable software operation. However, in practical scenarios, due to the large scale of programs and the relatively small proportion of faulty statements, the number of failing test cases in a test suite is lower than that of passing test cases to some extent, leading to an imbalance in defect knowledge. Additionally, mutual interference and masking among multiple faults in a program can introduce characterizing noise into the test suite, further increasing the difficulty of fault localization. To address these limitations, we propose a novel method called SA-BCL, which aims to tackle the issues of defect knowledge imbalance and characterizing noise. SA-BCL comprises four key components: the data processing component, the defect knowledge balancing component, the confident learning denoising component, and the program spectrum reducing component. Specifically, the data processing component constructs the program spectrum by analyzing program behaviors and test results during the execution of test cases. The defect knowledge balancing component mitigates the imbalance of defect knowledge by employing boundary identification and oversampling techniques to augment the spectrum corresponding to failing test cases located at the boundary of the test suite. The confident learning denoising component leverages confident learning techniques to identify and eliminate characterizing noise in the program spectrum. The program spectrum reducing component iteratively computes the suspiciousness scores of program elements and reduces the spectrum to facilitate fault localization. We conduct a series of experiments on both the synthetic multi-fault dataset and the real multi-fault dataset, comparing SA-BCL with baseline approaches. The experimental results clearly demonstrate that SA-BCL outperforms the state-of-the-art methods, achieving improvements of up to 37.5% in average wasted effort, 29.2% in precision, and 22.5% in recall. In addition, SA-BCL maintains comparable time costs within the same order of magnitude as baseline methods, and most of the improvements are statistically significant, demonstrating its robustness.

15 December 2025 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
WSWeihao Sun
Weihao Sun
SGShikai Guo
Shikai Guo
SWSiwen Wang
Siwen Wang
QMQian Ma
Qian Ma
Hui LiHui Li
Hui Li
YWYongpeng Weng
Yongpeng Weng
The automation of logic circuit design enhances chip performance, energy efficiency, and reliability, and is widely applied in the field of Electronic Design Automation (EDA). And-Inverter Graphs (AIGs) efficiently represent, optimize, and verify the functional characteristics of digital circuits, enhancing the efficiency of EDA development. Due to the complex structure and large scale of nodes in real-world AIGs, accurate modeling is challenging, leading to existing work lacking the ability to jointly model functional and structural characteristics, as well as insufficient dynamic information propagation capability. To address the aforementioned challenges, we propose AIGer, with the aim to enhance the expression of AIGs and thereby improve the efficiency of EDA development. Specifically, AIGer consists of two components: 1) Node logic feature initialization embedding component and 2) AIGs feature learning network component. The node logic feature initialization embedding component projects logic nodes, such as AND and NOT, into independent semantic spaces, to enable effective node embedding for subsequent processing. Building upon this, the AIGs feature learning network component employs a heterogeneous graph convolutional network, designing dynamic relationship weight matrices and differentiated information aggregation approaches to better represent the original structure and information of AIGs. The combination of these two components enhances AIGer’s ability to jointly model functional and structural characteristics and improves its message passing capability, thereby strengthening its expressive power for AIGs and enhancing the development efficiency of logic circuits. Experimental results indicate that AIGer outperforms the current best models in the Signal Probability Prediction (SPP) task, improving MAE and MSE by 18.95% and 44.44%, respectively. In the Truth Table Distance Prediction (TTDP) task, AIGer achieves improvements of 33.57% and 14.79% in MAE and MSE, respectively, compared to the best-performing models1.