XYXiaoyun Yan

Xiaoyun Yan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xiaoyun Yan has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open-Vocabulary RGB-Thermal Semantic Segmentation

    by Guoqiang Zhao, Junjie Huang, Xiaoyun Yan, Zhaojing Wang, Junwei Tang, Yangjun Ou, Xinrong Hu, Tao Peng


  • Industrial Process Soft Sensing Based on Bidirectional Optimization Learning of Data Augmentation and Prediction Models Under Limited Data

    by He Li, Zhaojing Wang, Xiaoyun Yan, Xinrong Hu, Lijun Li

    Soft sensing techniques are crucial for predicting key quality indicators in industrial processes. Despite the widespread application of deep learning in the soft sensing domain, challenges such as limited sampling and the complex nonlinear relationships among process variables limit the accuracy and adaptability of soft sensing models. Consequently, this study develops a bidirectional optimization learning of data augmentation and prediction modeling framework (BOL-DAPM). Considering that the generated samples must adhere to specific distribution characteristics and maintain the relationship between feature and target variables, a regression-constrained autoencoder (R-CAE) is developed that is capable of generating higher-quality new samples. To address the lack of consideration for maintaining intervariable correlation during the feature extraction process of soft sensing models, a nonlinear correlation index-constrained stacked target-related autoencoder (NC-STAE) is established, enhancing the accuracy of the predictive model. Considering the strong dependency between data generation and predictive models, a bidirectional optimization strategy is implemented through the loss function flow between the two models. This approach further improves the predictive accuracy of soft sensing with limited data. Experimental validation on datasets from the debutanizer column and concrete compressive strength confirmed that the proposed methods surpass recent comparative approaches in reducing prediction error, improving the coefficient of determination ( $R^{2}$ ) and lowering the mean absolute error (MAE), with an average precision performance increase of 35%.


  • DTDroid: Adversarial Packed Android Malware Detection Based on Traffic and Dynamic Behavioral

    by Junwei Tang, Sijie Zhou, Tao Peng, Xiaoyun Yan, Xinrong Hu, Wenlong Tian

    Android has occupied an important share of the operating system of intelligent terminal devices in the Internet of Things (IoT), and the malicious applications of Android have increased rapidly, posing a serious threat to the security of IoT. Machine learning has advanced significantly in the detection of android malware. In order to protect intellectual property, Android developers have begun to use packing techniques to enhance the security of their applications. However, attackers can also pack their malware, which may make feature extraction ineffective and interfere with the prediction results of learning-based classifiers. For this issue, we have designed and implemented a tool by dynamically loading the original DEX using a shell DexClassLoader to generate a batch of packed Android applications. And we have verified that several existing methods fail when faced with packed samples. Therefore, we propose a novel malware detection method called DTDroid that can resist code packing. DTDroid automatically captures network traffic characteristics of target samples based on fuzzy testing and network traffic packet extraction. At the same time, the dynamic behavior characteristics of the target application can be obtained by monitoring the corresponding runtime function calls and system status. The extracted two types of features are contextually spliced and converted into grayscale images, and then detected based on deep learning model. Experimental results show that the detection accuracy of our method reaches 94.22% and 95.14%, respectively, on two kinds of packed datasets, indicating that DTDroid has better robustness for packed samples than the existing methods.


  • Industrial Process Soft Sensing Based on Random Forest-Optimized Mallow's Cp Variable Selection

    by He Li, Zhaojing Wang, Xinrong Hu, Xiaoyun Yan, Fei Yang, Xiaoli Ruan

    Soft sensing has been extensively used in industrial processes to predict key quality variables. To establish an accurate sensor model, it is important to select appropriate auxiliary variables. In this paper, Random Forest optimized Mallow's Cp variable selection scheme is proposed for accurate soft sensing. Firstly, the method employs the Random Forest algorithm to rank the importance of variables. Then, it gradually selects variable subsets and calculates the Cp value for each subset to determine the optimal subset. Once the optimal subset is confirmed, we utilize it to train a Support Vector Regression model and evaluate the model's performance. The method considers both individual and combinations of multiple auxiliary variables on the target variable, which can address the problem of excessive auxiliary variables and redundant data in industrial production processes and reduce computation complexity. Finally, the penicillin fermentation process verifies the performance of the proposed method.

  • Exposure correction by deep curve estimation

    by Hang Luo, Jinxing Liang, Xiaoyun Yan


  • Nonlinear Dynamic Process Robust Monitoring Based on Spatiotemporal Preserving Projections With Rank-Order Distance

    by Zhaojing Wang, Xiaoyun Yan, Zhenwei Li, Xiong Pan, Yang Wang

    The prevalence of nonlinear dynamic characteristics and the interference of outliers are common occurrence in modern industrial systems. However, previous studies have inadequately considered the impact of these factors on process monitoring. This paper introduces a hybrid framework, referred to as spatiotemporal preserving projections (STPP), for fault detection in complex processes. Firstly, the conventional global and local preserving projections (GLPP) has been modified to address the constraints of hyperparameters and improve resilience. This is accomplished by proposing a strategy based on sparse representation to automatically determine local neighborhood in the global and local weight matrices. Additionally, a rank-order distance (ROD) based index is formulated to cope with outliers in the local weight matrix construction. Secondly, the modified GLPP is combined with slow feature analysis (SFA) to form the STPP method. This enables simultaneous preservation of spatial and temporal data structures in dimension reduction, further enhancing monitoring accuracy. Finally, a numerical case and wastewater treatment process demonstrate that the proposed STPP-based approach outperforms other relevant methods in fault detection.