Haibo Fan’s research while affiliated with University of Science and Technology of China and other places

What is this page?

This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (1)

FIGURE 1. Samples of the processed returned ultrasonic signals. Different colors of points denote different ultrasound frequencies. (a) and (b) represent two normal types. (c) and (d) represent two different types of defects.
FIGURE 5. Classification accuracy(%) of different networks of the first 100 epochs during the training.
FIGURE 6. Comparison of classification accuracy(%) of multi-head structure and multi-branch structure. The structure of multi-head is identical to the structure of multi-branch when the number of heads equals one.
Complexity analysis of the classification task on the dataset IV-A.
Classification accuracy(%) of different branch numbers. For excluding the influence of the different quantities of parameters, there is a result of K = 1(double para.), which has double quantities of parameters compared to the single branch structure(K = 1) and the same quantities compared to the two branches structure(K = 2)

+1

Multi-Branch Global Graph Convolution Network for Point Clouds
  • Article
  • Full-text available

January 2021

·

90 Reads

IEEE Access

Haibo Fan

·

·

Yanfei Liu

·

Jianming Ye

For finding the defect inside of rails, we usually use the ultrasound to inspect the rails and obtain point clouds data of the returned ultrasonic signals after preprocessing. The quantity of points in each point cloud is not fixed, and the point clouds are disordered and unstructured. The points have heterogeneous attributes which include not only continuous coordinates but discrete frequencies of the ultrasound. For classifying the special point clouds, we propose a network architecture, named Multi-Branch Global Graph Convolution Network. For better utilizing the features of heterogeneous attributes of each point, we introduce the point channel-wise attention mechanism to weight the attributes. The backbone of the network is a multi-branch learn network based on the global graph convolution. By constructing the global graph and applying the graph convolution, the network possesses the ability to learn the local relative position information and global semantic information. Experimental results show the effectiveness of each component of the network and the state-of-the-art performance of the proposed network achieves on the classification task of defect recognition.

Download