YYYang Yu

Yang Yu

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yang Yu has a total of 33 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Traffic Image Defogging Method under Hazy Environments Based on Frequency Domain Fusion and Attention Mechanism

    by Si-Jia Jiang, Ming-Jian Huang, Yang Yu, Hai-Bin Huang, Dong Liang

    Clear visual data is a fundamental prerequisite for real-time detection of and decision-making about abnormal traffic conditions in highway intelligent traffic systems (ITS). However, as a frequent weather phenomenon, haze can significantly reduce image contrast and obscure critical visual features, posing a serious threat to the effectiveness of highway ITS. This paper proposes an image defogging method based on spatial-frequency domain context joint convolutional network, aiming to address the limitations of existing methods in capturing frequency domain correlations and reducing spatial redundancy. Firstly, the method decomposes the haze characteristics from the spatial-frequency domain perspective, establishes a frequency domain defogging model more suitable for hazy scenes, and fully utilizes multi-scale frequency features to represent the correlation between the fog layer and the clear background. Secondly, based on the scale-space transformation and frequency-division feature mapping strategy, the feature correlations of low-frequency and high-frequency components are effectively decomposed, and redundancy is eliminated to improve computational efficiency. Then, a multi-layer channel attention mechanism is introduced, enabling the model to flexibly extract the weight information of haze distributions at different layers based on the frequency-domain feature decomposition, strengthening the attention to key features, and gradually stripping the complex overlapping fog layers through a sequential iterative approach. Finally, a defogging dataset is constructed to verify the effectiveness of the proposed method. Experiment results demonstrate that the proposed method achieves a peak signal-to-noise ratio of 30.3 dB and a structural similarity index measure of 0.9, outperforming existing methods. These results confirm the effectiveness of our method in restoring image clarity and preserving structural details, offering practical support for highway ITS under adverse weather conditions.


  • Pose-Skeleton Guided Cross-Attention Representation Fusion for Occluded Pedestrian Re-Identification

    by Shuze Geng, Yifan Liu, Zijin Wang, Gang Yan, Yang Yu, Yingchun Guo

    Most methods address occluded pedestrian Re-Identification (Re-ID) by employing external auxiliary models in the feature output stage of the backbone network to locate visible appearance areas. Nevertheless, these approaches suffer from issues such as occlusion information diffusion and imprecise masks generated by external models, indicating the need for further exploration in the decoupling of pedestrian features from occlusion information. In light of these challenges, we propose an innovative algorithm called Pose-Skeleton guided Cross-attention Representation fusion (PSCR) method. Firstly, we introduce the Visible Appearance Region Attention (VARA) model designed to leverage pose information for guiding the backbone network in effectively distinguishing between occlusion information and pedestrian features at the intermediate layer. By employing a suppression strategy, the model is able to effectively suppress occlusion interference and alleviate the diffusion of occlusion information. Next, to achieve precise localization of pedestrian-specific semantic regions, a groundbreaking Skeletal Area Modeling (SAM) is proposed. Leveraging the principles of mathematical modeling and capitalizing on the efficacy of human keypoint confidence, this module generates finely-grained masks for local skeleton regions and extracts an exhaustive set of local features. Lastly, under the constraints imposed by spatial attention masks, a cross-attention mechanism is employed to fuse the features acquired from the previous two steps with local features. This fusion process results in the generation of enhanced local features that seamlessly integrate aligning high-level semantic information. Extensive experimentation demonstrates that the proposed algorithm exhibits notable performance advancements when compared to existing methodologies.


  • Open Access

    Perception and Structure Association Network for Micro-Expression Recognition

    by Zhihao Wang, Yuanyuan Zhang, Bofan Yu, Yang Yu

    Micro-expression recognition (MER) poses a formidable challenge due to the fleeting and subtle nature of micro-expressions (MEs). Current approaches predominantly focus on either motion features or facial structure, yet fail to effectively integrate both perspectives. This paper introduces a novel Perception and Structure Association Network (PSA-Net) for MER, aiming to effectively extract key features in MEs by performing multi-scale perception of motion features and constructing adaptive facial structure association. Specifically, to perceive fine-grained ME features and achieve multi-scale key feature extraction, we have designed the Multi-scale Micro-motion Perception module. This module extracts motion features by dividing windows and performing self-attention calculations within each window. The motion features of each window are then fused layer by layer using shifted window attention and spatial attention. Simultaneously, to effectively construct the structure association of facial ME regions, we have designed the Adaptive Facial Structure Association module. This module employs graph attention mechanisms and global adaptive attention maps to achieve the adaptive structure association of facial regions of interest, enabling the constructed facial graph structure to effectively extract global ME information and the relationships between nodes. Additionally, to effectively fuse these two key ME features, we propose the Fusion Loss, which can effectively learn the similarities between them, promoting the model to learn more discriminative feature representations, thus achieving effective feature fusion. Extensive experiments conducted on four public ME datasets (CASME, CASME II, SAMM, CAS(ME)3) have achieved recognition accuracy/F1 scores of 81.87%/79.26%, 87.80%/88.08%, 82.35%/76.76%, and 63.10%/49.47% respectively, surpassing most existing methods, which indicates that our method can effectively improve MER performance.

  • Teacher-Student Prototype Enhancement Network for a Few-Shot Remote Sensing Scene Classification

    by Ye Zhu, Shanying Yang, Yang Yu

    Few-shot remote sensing scene classification identifies new classes from limited labeled samples where the great challenges are intraclass diversity, interclass similarity, and limited supervision. To alleviate these problems, a teacher-student prototype enhancement network is proposed for a few-shot remote sensing scene classification. Instead of introducing an attentional mechanism in mainstream studies, a prototype enhancement module is recommended to adaptively select high-confidence query samples, which can enhance the support prototype representations to emphasize intraclass and interclass relationships. The construction of a few-shot teacher model generates more discriminative predictive representations with inputs from many labeled samples, thus providing a strong supervisory signal to the student model and encouraging the network to achieve accurate classification with a limited number of labeled samples. Extensive experiments of four public datasets, including NWPU-remote sens ing image scene classification (NWPU-RESISC45), aerial image dataset (AID), UC Merced, and WHU-RS19, demonstrate that this method achieves superior competitive performance than the state-of-the-art methods on five-way, one-shot, and five-shot classifications.

  • Progressive Mask Transformer With Edge Enhancement for Image Manipulation Localization

    by Ye Zhu, Jian Liu, Yang Yu, Yingchun Guo, Xiaoke Hao

    Recent developments in image editing techniques have given rise to serious challenges to the credibility of multimedia data. Although some deep learning methods have achieved impressive results, they often fail to detect subtle edge artefacts, and current mainstream methods focus mainly on the foreground content and ignore the background content, which also contains abundant information related to manipulation. To address this issue, this paper proposes a progressive mask transformer with an edge enhancement network for image manipulation localization. Specifically, an edge enhancement flow is introduced to detect subtle manipulated edge artefacts and guide the localization of manipulated regions. Then, the manipulated, genuine and global features are progressively refined using a progressive mask transformer module. We perform extensive experiments on NIST16, Coverage, CASIA and IMD20 datasets to verify the effectiveness of our method, and the results demonstrate that the proposed method outperforms state-of-the-art methods by a wide margin based on on commonly used evaluation metrics.


  • Open Access

    The Heterogeneous Effect and Transmission Paths of Air Pollution on Housing Prices: Evidence from 30 Large- and Medium-Sized Cities in China

    by Guangping Liu, Guo Zhang, Yanfei Li, Yang Yu

    Utilizing panel data from 30 large- and medium-sized cities in China from 2003 to 2018, this study employs the bootstrap method to investigate the heterogeneous impact of air pollution on housing prices and its underlying mechanisms. The research findings indicate that an increase in the concentration of various air pollutants has a restraining effect on housing prices, with the primary components of haze exerting a greater influence. Air pollution indirectly affects housing prices by influencing urban economic development and the development of the tertiary industry, with population density and the development of the secondary industry not serving as mediators in this relationship. Finally, policy recommendations are provided, including enhancing mandatory standards for energy consumption and waste gas emissions by enterprises, establishing a robust environmental monitoring system, and promoting industrial transformation and upgrading, aiming to provide insights for the stable development of the real estate industry.


  • Open Access

    Edge-Guided Parallel Network for VHR Remote Sensing Image Change Detection

    by Ye Zhu, Kaikai Lv, Yang Yu, Wenjia Xu

    Change Detection (CD) is an important research topic in the remote sensing field, and it has a wide range of applications, including resource monitoring, disaster assessment, urban planning, etc. Recently, Deep Learning (DL) has shown its advantages in CD. However, most existing DL-based methods can not capture the complementary information between bi-temporal and difference features. This paper proposes an Edge-Guided Parallel Network (EGPNet) to solve this problem. First, our EGPNet extracts bi-temporal and difference features simultaneously through a parallel encoding framework. During parallel encoding, we design an Supplementary Mechanism (SM) to enrich the difference features with bi-temporal features. Second, we fuse bi-temporal and difference features at each feature level to sufficiently exploit their complementarity. Finally, the Edge-Aware Module (EAM) and Edge-Guidance Feature Module (EFM) are introduced to enhance the edge representation for improving blurred edges of detection results. Benefiting from the rich change-related information in difference features and detailed information in bi-temporal features, our EGPNet can detect change regions entirely and accurately. Experimental results on the LEVIR-CD, SYSU-CD and CDD datasets demonstrate that the proposed method outperforms several state-of-the-art approaches. Especially our EGPNet can detect more precise and sharper edges than other methods. We open source the Pytorch implementation and the pre-trained model at https://github.com/Lvkyky/EGPNet


  • Part-Based Representation Enhancement for Occluded Person Re-Identification

    by Gang Yan, Zijin Wang, Shuze Geng, Yang Yu, Yingchun Guo

    Retrieving an occluded pedestrian remains a challenging problem in person re-identification (re-id). Most existing methods utilize external detectors to disentangle the visible body parts. However, these methods are unstable due to domain bias and consume numerous computing resources. In this paper, we propose a novel and lightweight Part-based Representation Enhancement (PRE) network for occluded re-id that takes full advantages of the local correlations to aggregate distinctive information for local features without relying on auxiliary detectors. First, according to the information qualities of different body parts, we design a reasonable partition strategy to obtain the local features. Next, a Partial Relationship Aggregation (PRA) module is developed to self-mine the visibility of the body and construct a correlation matrix for collecting the information related to pre-defined classes. Following this, we propose an Inter-part Omnibearing Fusion (IOF) module that leverages the occlusion-suppressed class features to enhance the distinctiveness of the local features via feature completion and reverse fusion strategies. During the testing phase, the global and reconstructed local features are concatenated together for re-id without a complex visible region matching algorithm. Extensive experiments on occluded, partial, and holistic re-id benchmarks demonstrate the superiority of PRE over state-of-the-art methods in terms of accuracy and model complexity.