Publications by authors named "Lanhui Fu"
Article Synopsis
- The study introduces a new detection method, YOLOv7-BiGS, for identifying XinHui citrus varieties, aimed at enhancing the production of dried tangerine peel.
- This method combines a BiFormer attention mechanism and a lightweight GSConv convolution to improve feature extraction and reduce model complexity while keeping performance high.
- Experimental results show YOLOv7-BiGS outperformed other models like YOLOv5, YOLOv7, and YOLOv8 in precision, mean Average Precision (mAP), and recall, especially in challenging environmental conditions with occluded citrus images.
This article aims to improve the deep-learning-based surface defect recognition. In actual manufacturing processes, there are issues such as data imbalance, insufficient diversity, and poor quality of augmented data in the collected image data for product defect recognition. A novel defect generation method with multiple loss functions, DG2GAN is presented in this paper.
View Article and Find Full Text PDFIn this study, a series of AgPO/graphene oxide (GO) films were dip-coated on a metal nickel foam. The immobilized catalysts were characterized by X-ray diffraction, scanning electron microscopy, X-ray photoelectron spectroscopy, ultraviolet-visible spectroscopy, Raman spectroscopy, high-resolution transmission electron microscopy and photoluminescence spectroscopy. The results show that AgPO/GO was successfully supported on a nickel foam.
View Article and Find Full Text PDF