SYShuaiying Yu

Shuaiying 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

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Shuaiying Yu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:

  • Guoliang Yang

    School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou, China

    3 shared publications
  • Guoliang Yang

    School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, China

    3 shared publications
  • Hao Yang

    School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou, China

    3 shared publications
  • Hao Yang

    School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, China

    3 shared publications
  • Jixiang Wang

    School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, China

    3 shared publications
  • Ziling Nie

    School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, China

    3 shared publications

Latest Publications

  • Open Access

    Author Correction: Attention and feature transfer based knowledge distillation

    by Guoliang Yang, Shuaiying Yu, Yangyang Sheng, Hao Yang

  • Open Access

    Attention and feature transfer based knowledge distillation

    by Guoliang Yang, Shuaiying Yu, Yangyang Sheng, Hao Yang

    Existing knowledge distillation (KD) methods are mainly based on features, logic, or attention, where features and logic represent the results of reasoning at different stages of a convolutional neural network, and attention maps symbolize the reasoning process. Because of the continuity of the two in time, transferring only one of them to the student network will lead to unsatisfactory results. We study the knowledge transfer between the teacher-student network to different degrees, revealing the importance of simultaneously transferring knowledge related to the reasoning process and reasoning results to the student network, providing a new perspective for the study of KD. On this basis, we proposed the knowledge distillation method based on attention and feature transfer (AFT-KD). First, we use transformation structures to transform intermediate features into attentional and feature block (AFB) that contain both inference process information and inference outcome information, and force students to learn the knowledge in AFBs. To save computation in the learning process, we use block operations to align the teacher-student network. In addition, in order to balance the attenuation ratio between different losses, we design an adaptive loss function based on the loss optimization rate. Experiments have shown that AFT-KD achieves state-of-the-art performance in multiple benchmark tests.


  • Open Access

    HMC: Hybrid model compression method based on layer sensitivity grouping

    by Guoliang Yang, Shuaiying Yu, Hao Yang, Ziling Nie, Jixiang Wang

    Previous studies have shown that deep models are often over-parameterized, and this parameter redundancy makes deep compression possible. The redundancy of model weight is often manifested as low rank and sparsity. Ignoring any part of the two or the different distributions of these two characteristics in the model will lead to low accuracy and a low compression rate of deep compression. To make full use of the difference between low-rank and sparsity, a unified framework combining low-rank tensor decomposition and structured pruning is proposed: a hybrid model compression method based on sensitivity grouping (HMC). This framework unifies the existing additive hybrid compression method (AHC) and the non-additive hybrid compression method (NaHC) proposed by us into one model. The latter group the network according to the sensitivity difference of the convolutional layer to different compression methods, which can better integrate the low rank and sparsity of the model compared with the former. Experiments show that our approach achieves a better trade-off between test accuracy and compression ratio when compressing the ResNet family of models than other recent compression methods using a single strategy or additive hybrid compression.


  • Open Access

    MSREA-Net: An Efficient Skin Disease Segmentation Method Based on Multi-Level Resolution Receptive Field

    by Guoliang Yang, Ziling Nie, Jixiang Wang, Hao Yang, Shuaiying Yu

    Aiming at the low contrast of skin lesion image and inaccurate segmentation of lesion boundary, a skin lesion segmentation method based on multi-level split receptive field and attention is proposed. Firstly, the depth feature extraction module and multi-level splitting receptive field module are used to extract image feature information; secondly, the hybrid pooling module is used to build long-term and short-term dependencies and integrate global information and local information. Finally, the reverse residual external attention module is introduced to construct the decoding part, which can mine the potential relationship between data sets and improve the network segmentation ability. Experiments on ISBI2017 and ISIC2018 data sets show that the Dice similarity coefficient and Jaccard index reach 88.67% and 91.84%, 79.25% and 81.48%, respectively, and the accuracy reaches 93.89% and 96.16%. The segmentation method is superior to the existing algorithms as a whole. Simulation experiments show that the network has a good effect on skin lesion image segmentation and provides a new method for skin disease diagnosis.


  • Top Cited Papers
    Open Access

    A Lightweight YOLOv8 Tomato Detection Algorithm Combining Feature Enhancement and Attention

    by Guoliang Yang, Jixiang Wang, Ziling Nie, Hao Yang, Shuaiying Yu

    A tomato automatic detection method based on an improved YOLOv8s model is proposed to address the low automation level in tomato harvesting in agriculture. The proposed method provides technical support for the automatic harvesting and classification of tomatoes in agricultural production activities. The proposed method has three key components. Firstly, the depthwise separable convolution (DSConv) technique replaces the ordinary convolution, which reduces the computational complexity by generating a large number of feature maps with a small amount of calculation. Secondly, the dual-path attention gate module (DPAG) is designed to improve the model’s detection precision in complex environments by enhancing the network’s ability to distinguish between tomatoes and the background. Thirdly, the feature enhancement module (FEM) is added to highlight the target details, prevent the loss of effective features, and improve detection precision. We built, trained, and tested the tomato dataset, which included 3098 images and 3 classes. The proposed algorithm’s performance was evaluated by comparison with the SSD, faster R-CNN, YOLOv4, YOLOv5, and YOLOv7 algorithms. Precision, recall rate, and mAP (mean average precision) were used for evaluation. The test results show that the improved YOLOv8s network has a lower loss and 93.4% mAP on this dataset. This improvement is a 1.5% increase compared to before the improvement. The precision increased by 2%, and the recall rate increased by 0.8%. Moreover, the proposed algorithm significantly reduced the model size from 22 M to 16 M, while achieving a detection speed of 138.8 FPS, which satisfies the real-time detection requirement. The proposed method strikes a balance between model size and detection precision, enabling it to meet agriculture’s tomato detection requirements. The research model in this paper will provide technical support for a tomato picking robot to ensure the fast and accurate operation of the picking robot.


  • Open Access

    A Multi-Scale Dehazing Network with Dark Channel Priors

    by Guoliang Yang, Hao Yang, Shuaiying Yu, Jixiang Wang, Ziling Nie

    Image dehazing based on convolutional neural networks has achieved significant success; however, there are still some problems, such as incomplete dehazing, color deviation, and loss of detailed information. To address these issues, in this study, we propose a multi-scale dehazing network with dark channel priors (MSDN-DCP). First, we introduce a feature extraction module (FEM), which effectively enhances the ability of feature extraction and correlation through a two-branch residual structure. Second, a feature fusion module (FFM) is devised to combine multi-scale features adaptively at different stages. Finally, we propose a dark channel refinement module (DCRM) that implements the dark channel prior theory to guide the network in learning the features of the hazy region, ultimately refining the feature map that the network extracted. We conduct experiments using the Haze4K dataset, and the achieved results include a peak signal-to-noise ratio of 29.57 dB and a structural similarity of 98.1%. The experimental results show that the MSDN-DCP can achieve superior dehazing compared to other algorithms in terms of objective metrics and visual perception.