GYGaoming Yang

Gaoming Yang

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Gaoming Yang has a total of 72 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • An Efficient Tag Reading Protocol Based on Distance in Commodity RFID Systems

    by Xiujun Wang, Xianglong Hua, Tao Tao, Gaoming Yang, Luyu Gui

  • Open Access

    Patch-Divided Flexible and Diverse Super-Resolution Style Transfer

    by Guoren Yao, Gaoming Yang, Xintian Liu

    Recently, super-resolution transmission methods have been introduced on limited GPU. However, due to the stereotyped style exchange process, the GPU consumption and diversity of super-resolution style transfer are still challenging. In addition, existing methods can inadvertently lead to content leakage and uneven stroke size distribution, resulting in less attractive results. Hence, we introduce a fast and diverse super-resolution transfer (FDST) model, which can realize more flexible super-resolution multi-style transfer by mapping noise and information from another style in the style encoder. In addition, we propose two loss functions within the existing framework to support the preservation of content structure: Patch Content-Consistent Patch Loss (Patch-CCPL) and Patch contrastive loss. The proposed method effectively and elaborately integrates colors and texture structures. The key idea of FDST is to divide the super-resolution image into small patches, and then perform diverse style conversions on each small patch by injecting subtle noise or sub-style images. We implemented theoretical analyses and extensive results to qualitatively and quantitatively evaluate our method and compare it with the state-of-the-art algorithm. Extensive experiments on 4K content images demonstrate that FDST achieves a user preference score of 0.207, SSIM of 0.492, and LPIPS of 0.526, outperforming existing methods in content preservation while requiring only 2.573 GB model storage. Ablation studies confirm the contribution of each component, and a discussion of security applications including watermarking and forensic analysis is provided.

  • Robust Features Learning Under the Guidance of Generalization Features

    by Xianjin Fang, Fanmao Zeng, Zhongyi Fang, Gaoming Yang, Ji Zhang

    Deep neural networks (DNNs) are vulnerable to adversarial attacks, where attackers can easily deceive these models by introducing slight perturbations to input examples that are imperceptible to humans, leading to incorrect predictions. This vulnerability significantly undermines the deployment of deep learning models in security-critical applications. Adversarial training is among the most practical methods to improve the robustness of DNNs. However, state-of-the-art adversarial training methods exhibit a trade-off: They improve model robustness on training data but result in a significant gap in generalization, with high robustness during training but lower robustness during testing. In this paper, we propose a novel adversarial training framework to enhance the robust generalization of DNNs. This framework introduces two key innovations: perturbing features and generalized feature guidance. Specifically, we designed a feature perturbation strategy that introduces perturbations to all features that could influence the model's predictions. The goal is to enable the model to learn more robust features while enhancing the robustness of its existing features. Additionally, we leverage the generalized features from a clean model to guide the training of a robust model, with the aim of mitigating natural accuracy degradation. We evaluate the effectiveness of the proposed method through extensive experiments on popular benchmark datasets, such as CIFAR and the Street View House Numbers dataset. We also test its robustness against state-of-the-art adversarial attacks, including projected gradient descent and AutoAttack. The results demonstrate that our approach significantly improves the robust generalization of DNNs.

  • Open Access

    A lightweight fabric defect detection with parallel dilated convolution and dual attention mechanism

    by Zheqing Zhang, Kezhong Lu, Gaoming Yang

    Detecting defects in fabrics is essential to quality control in the manufacturing process of textile productions. To increase detection efficiency, a variety of automatic fabric defect detections have been developed. However, most of these methods rely on complex model with heavy parameters, leading to high computational costs that hinder their adaptation to real-time detection environments. To overcome these obstacles, we proposed a lightweight fabric defect detection (Light-FDD), building upon the You Only Look Once v8 Nano (YOLOv8n) framework with further optimizations. Specifically, the backbone employed an improved FasterNet architecture for feature extraction. In order to capture multi-scale contextual information, we designed a parallel dilated convolution downsampling (PDCD) block to replace the conventional downsampling block in the backbone. In addition, a novel dual attention mechanism, called the global context and receptive-filed (GCRF) attention, was presented to help the model focus on key regions. Furthermore, a lightweight cross-stage partial (CSP) layer was deployed by dual convolution for feature fusion, reducing redundant parameters to further lighten the model. Results from extensive experiments on public fabric defect datasets showed that Light-FDD outperforms existing state-of-the-art lightweight models in terms of detection accuracy while requiring low computational cost. The present study suggests that the performance and effectiveness of detection models can be balanced through the adoption of reasonable strategies.


  • High-Dimensional Data Release With Local Differential Privacy Under IoT Architecture

    by Xinxin Ye, Gaoming Yang, Hai Deng, Pan Jie, Rongshi Wu, Hui Jiang

    Local differential privacy (LDP) mechanisms are widely used to collect data generated by IoT sensor devices to protect sensitive information. However, it easily leads to low data utility and high data computing cost due to complex structure and high dimensionality of IoT data. To alleviate this problem, we propose a High-dimensional Data Publishing method using Random responses based on Markov network (HDPRM). This method efficiently conducts the collection and analysis of high-dimensional data under the IoT architecture and satisfies LDP. In particular, it uses the expectation maximization (EM) algorithm to reconstruct the joint distribution of high-dimensional data attributes. Specifically, to improve the effectiveness of data release, we calculate the correlation between attributes and construct corresponding Markov network on the server. Additionally, we cluster high-dimensional attributes through the junction tree algorithm, filter out the joint probability that meets the requirements, and then use this probability to generate synthetic data for publication from the sampled data. Extensive experiments are conducted to comprehensively evaluate the performance of the HDPRM on three real-world datasets. The results show that the method achieves higher data utility under LDP guarantee compared to state-of-the-art methods.


  • Open Access

    Aspect-based Sentiment Analysis for COVID-19: A Heterogeneous Graph Convolutional Network Approach

    by Linlin Hou, Wenhui Tu, Ting Yu, Ting Jiang, Mohamed Bah, Zenghui Xu, Yu Zhang, Gaoming Yang, Ji Zhang

    The epidemic of infectious diseases has a significant impact on society, the economy, and people’s lives. Social media, with its high user participation and rapid information dissemination, plays a crucial role in shaping public opinion. Fine-grained sentiment analysis of public opinion on infectious diseases can provide valuable insights for improving the quality of public services. However, there are few relevant studies on Chinese data due to language complexity and low resources. Moreover, most of the existing approaches utilize the Graph Neural Network (GCN) method by syntactic dependency trees to construct graphs of text, which ignore the potential link relationships between aspects and words. Therefore, to address this limitation, in this article, we propose a new method based on GCN using aspect-specific heterogeneous graphs, named ASHGCN, which combines BiLSTM, heterogeneous graphs, GCN, the mask and the attention mechanism. We mine social media posts related to COVID-19 for aspect-based sentiment analysis task (ABSA) for ten aspect entity types in both Chinese and English data. The heterogeneous graph is designed with two node types (aspect nodes and non-aspect nodes) and four edge connection types, including various relationships between aspect entities, and between aspect entities and non-aspect entities. In addition, we release a Chinese dataset and an English dataset that include medical and named entities, along with corresponding sentiment labels. Experiments on our datasets, as well as two public datasets, demonstrate that our method greatly improves performance in the ABSA task. Ablation experiments and case studies further support the effectiveness of the proposed approach.


  • Open Access

    Coal and gas outburst prediction based on data augmentation and neuroevolution

    by Wenbing Shi, Ji Huang, Gaoming Yang, Shuzhi Su, Shexiang Jiang

    Coal and gas outburst (CGO) is a complicated natural disaster in underground coal mine production. In constructing smart mines, predicting CGO risks efficiently and accurately is necessary. This paper proposes a CGO risk prediction method based on data augmentation and a neuroevolution algorithm, denoted as ANEAT. First, sample features are applied to the transfer function using a pointwise intensity transformation to obtain new feature samples. It solves the problems of imbalanced data samples and insufficient diversity. Second, the feature importance score sorting and Sparse PCA dimensionality reduction are performed on the data-augmented samples. It provides the initial genome code for the evolutionary neural network. Finally, an evolutionary neural network for CGO prediction is constructed through population initialization, fitness evaluation, species differentiation, genome mutation, and recombination. The optimal phenotype is obtained in the evolutionary generations. In the experiment, we verify the effectiveness of ANEAT from multiple aspects such as data augmentation effectiveness analysis, deep learning model comparison, swarm intelligence optimization algorithm comparison, and other method comparisons. The results show that the MAE, RMSE, and EVAR indexes of ANEAT on the test set are 0.0816, 0.1322, and 0.8972, respectively. It has the optimal CGO prediction effect. ANEAT realizes the high-precision mapping of feature parameters and outburst risk with a lightweight network architecture, which can be well applied to CGO prediction.