JYJunfeng Yu

Junfeng 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


Junfeng Yu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • YCANet: Target Detection for Complex Traffic Scenes Based on Camera-LiDAR Fusion

    by Zhen Shen, Yunze He, Xu Du, Junfeng Yu, Hongjin Wang, Yaonan Wang

    In traffic scenes, target detection is impacted by factors such as complex backgrounds, illumination, and mutual occlusion of moving targets, all of which tend to cause the sensors to perform poorly and have a high false detection rate. To solve these challenges, this study proposes a multi-data source target detection network integrating target tracking based on camera and LiDAR fusion (YCANet), which utilizes the improved YOLOv7 and CenterPoint to detect the visible images and point clouds separately, and uses the Aggregated Euclidean Distances (AED) as the new metric in the data correlation module for tracking the detection results of the images and point clouds, which effectively improves the correlation robustness and reduces the tracking errors. An optimal matching fusion strategy is presented to merge the detection and tracking results of two sensors for decision matching. The fusion of camera and LiDAR improves the poor detection results, while the target tracking incorporated into the detection approach reduces the false detection rate. Homemade dataset and partial ONCE dataset are used for network training and testing, and comparing with the other seven algorithms, the experimental results show that our proposed approach better meets the accuracy requirements, and the mAP reaches 83.40% while maintaining the false detection rate of 18.19%.


  • Thermal Inspection of Subsurface Defects in Wind Turbine Blade Segments Under the Natural Solar Condition

    by Xiang Li, Yunze He, Hongjin Wang, Gaosen Sun, Junfeng Yu, Xu Du, Mingshuang Hu, Yaonan Wang

    This study explores the potential of unmanned aerial vehicles (UAV) thermography to inspect wind turbine blades using natural solar radiation. A comparative study of numerical analysis and experiments is presented. The simulation was performed using realistic solar boundary conditions. In parallel, outdoor thermal experiments were conducted utilizing a UAV on several customized specimens, including regular glass fiber reinforced polymer boards and a segment from a real wind turbine blade. The results demonstrated the detectability of solar thermography to inspect internal delaminations in blade structure. In particular, the results showed that the thermal patterns of the subsurface defects are dynamic throughout the daytime, which means the distinction between surface and interior defects could be attained. A comparison of two kinds of delaminations and surface defects is also provided.


  • An Infrared Image Stitching Method for Wind Turbine Blade Using UAV Flight Data and U-Net

    by Junfeng Yu, Yunze He, Fan Zhang, Gaosen Sun, Yuejun Hou, Hao Liu, Jie Li, Ruizhen Yang, Hongjin Wang

    Infrared image stitching of wind turbine blades is a challenging task due to weak texture and low contrast. To address this issue, an infrared image stitching method for wind turbine blades based on UAV flight data and U-Net is proposed in this paper. Remove complex background using U-Net semantic segmentation network and retain the complete area of the blades to be stitched. The rotation angle of blades is calculated to unify the orientation of the blades in all images to be horizontal, the increasing pixel between adjacent images is calculated by the velocity, diagonal field of view and object distance of UAV and scaling coefficient is calculated to align the upper and lower edges of blades. Based on the above parameters, the panoramic infrared images of wind turbine blades can be finally obtained. Through comparative analysis and quantitative evaluation with the widely used image stitching algorithms, it is shown that the proposed method can obtain better quality results.


  • An Improved U-Net Model for Infrared Image Segmentation of Wind Turbine Blade

    by Junfeng Yu, Yunze He, Hao Liu, Fan Zhang, Jie Li, Gaosen Sun, Xiaofei Zhang, Ruizhen Yang, Pan Wang, Hongjin Wang

    Infrared image segmentation of the wind turbine blade is an important link of wind turbine condition monitoring. Nevertheless, there remain plenty of challenges when the traditional semantic segmentation networks are applied directly to the infrared image segmentation of wind turbine blade, including the blurred boundaries caused by similar thermal radiation of background and blade, the irregularity of edge identification caused by uneven heating and the hub misidentified as blade. To address these issues, a semantic segmentation neural network based on the basic U-Net is improved for infrared image segmentation of the wind turbine blade. The hierarchical-split depthwise separable convolution block is integrated into the constructed network to have a higher segmentation accuracy. And the fusion of convolution layer and batch normalization layer is performed in inference to get a better speed accuracy trade-off. The experimental results show that the proposed approach outperforms all the comparing methods, which further demonstrates that the trained network can achieve superior runtime performance and accurate segmentation with excellent anti-interference performance.