Yan Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yan Liu has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Haiying Xia
Guangxi Key Laboratory of Brain-Inspired Computing and Intelligent Chips, School of Electronic and Information Engineering, Guangxi Normal University, Guilin, China
2 shared publications - Yunbai Qin
School of Electronic and Information Engineering, Guangxi Normal University, Guilin 541004, China
2 shared publications - Cong Wang
School of Electronic and Information Engineering, Guangxi Normal University, Guilin 541004, China
1 shared publication - Cong Wang
School of Electronic and Information Engineering, Guangxi Normal University, Guilin 541004, China
1 shared publication - Zhonglan Lin
School of Electronic and Information Engineering, Guangxi Normal University, Guilin 541004, China
1 shared publication - Zhonglan Lin
School of Electronic and Information Engineering, Guangxi Normal University, Guilin 541004, China
1 shared publication
Latest Publications
- Open Access
- 15 August 2024
- Published by MDPI AG in Electronics
- Vol. 13 (16) , 3241
Detection of Scratch Defects on Metal Surfaces Based on MSDD-UNet
In this work, we enhanced the U-shaped network and proposed a method for detecting scratches on metal surfaces based on the Metal Surface Defect Detection U-Net (MSDD-UNet). Initially, we integrated a downsampling approach using a Space-To-Depth module and a lightweight channel attention module to address the loss of contextual information in feature maps that results from multiple convolution and pooling operations. Building on this, we developed an improved attention module that utilizes image frequency decomposition and cross-channel self-attention mechanisms, as well as the strengths of convolutional encoders and self-attention blocks. Additionally, this attention module was integrated into the skip connections between the encoder and decoder. The purpose was to capture dense contextual information, highlight small and fine target areas, and assist in localizing micro and fine scratch defects. In response to the severe foreground–background class imbalance in scratch images, a hybrid loss function combining focal loss and loss was put forward to train the model for precise scratch segmentation. Finally, experiments were conducted on two surface defect datasets. The results reveal that our proposed method is more advantageous than other state-of-the-art scratch segmentation methods.- 15 August 2024
- Published by MDPI AG in Electronics
- Vol. 13 (16) , 3241
Open AccessDetection of Scratch Defects on Metal Surfaces Based on MSDD-UNet
In this work, we enhanced the U-shaped network and proposed a method for detecting scratches on metal surfaces based on the Metal Surface Defect Detection U-Net (MSDD-UNet). Initially, we integrated a downsampling approach using a Space-To-Depth module and a lightweight channel attention module to address the loss of contextual information in feature maps that results from multiple convolution and pooling operations. Building on this, we developed an improved attention module that utilizes image frequency decomposition and cross-channel self-attention mechanisms, as well as the strengths of convolutional encoders and self-attention blocks. Additionally, this attention module was integrated into the skip connections between the encoder and decoder. The purpose was to capture dense contextual information, highlight small and fine target areas, and assist in localizing micro and fine scratch defects. In response to the severe foreground–background class imbalance in scratch images, a hybrid loss function combining focal loss and loss was put forward to train the model for precise scratch segmentation. Finally, experiments were conducted on two surface defect datasets. The results reveal that our proposed method is more advantageous than other state-of-the-art scratch segmentation methods. - Open Access
- 7 August 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (16) , 6914
A Method for Enhancing the Accuracy of Pet Breeds Identification Model in Complex Environments
Most existing studies on pet breeds classification focus on images with simple backgrounds, leading to the unsatisfactory performance of models in practical applications. This paper investigates training pet breeds classification models using complex images and constructs a dataset for identifying breeds of pet cats and dogs. We use this dataset to fine-tune three SOTA models: ResNet34, DenseNet121, and Swin Transformer. Specifically, in terms of top-1 accuracy, the performance of DenseNet is improved from 89.10% to 89.19%, while that of the Swin Transformer is increased by 1.26%, marking the most significant enhancement. The results show that training with our dataset significantly enhances the models’ classification capabilities in complex environments. Additionally, we offer a lightweight pet breeds identification model based on PBI-EdgeNeXt (Pet Breeds Identification EdgeNeXt). We utilizes the PolyLoss function and Sophia optimizer for model training. Furthermore, we compare our model with five commonly used lightweight models and find that the proposed model achieves the highest top-1 accuracy of 87.12%. These results demonstrate that the model achieves high accuracy, reaching the SOTA level.
- 7 August 2024
- Published by MDPI AG in Applied Sciences
- Vol. 14 (16) , 6914
Open AccessA Method for Enhancing the Accuracy of Pet Breeds Identification Model in Complex Environments
Most existing studies on pet breeds classification focus on images with simple backgrounds, leading to the unsatisfactory performance of models in practical applications. This paper investigates training pet breeds classification models using complex images and constructs a dataset for identifying breeds of pet cats and dogs. We use this dataset to fine-tune three SOTA models: ResNet34, DenseNet121, and Swin Transformer. Specifically, in terms of top-1 accuracy, the performance of DenseNet is improved from 89.10% to 89.19%, while that of the Swin Transformer is increased by 1.26%, marking the most significant enhancement. The results show that training with our dataset significantly enhances the models’ classification capabilities in complex environments. Additionally, we offer a lightweight pet breeds identification model based on PBI-EdgeNeXt (Pet Breeds Identification EdgeNeXt). We utilizes the PolyLoss function and Sophia optimizer for model training. Furthermore, we compare our model with five commonly used lightweight models and find that the proposed model achieves the highest top-1 accuracy of 87.12%. These results demonstrate that the model achieves high accuracy, reaching the SOTA level.