H
Haifeng Li
Researcher at Central South University
Publications - 196
Citations - 5068
Haifeng Li is an academic researcher from Central South University. The author has contributed to research in topics: Computer science & Feature extraction. The author has an hindex of 21, co-authored 167 publications receiving 1837 citations. Previous affiliations of Haifeng Li include Instituto Nacional de Técnica Aeroespacial & Harbin Institute of Technology.
Papers
More filters
Journal ArticleDOI
T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction
TL;DR: In this article, a novel neural network-based traffic forecasting method, the temporal graph convolutional network (T-GCN) model, which is combined with the graph convolutionsal network and the gated recurrent unit (GRU), is proposed.
Journal ArticleDOI
DASNet: Dual Attentive Fully Convolutional Siamese Networks for Change Detection in High-Resolution Satellite Images
TL;DR: The weighted double-margin contrastive loss is proposed to address the imbalanced sample is a serious problem in change detection, i.e., unchanged samples are much more abundant than changed samples, which is one of the main reasons for pseudochanges.
Journal ArticleDOI
Ensemble of differential evolution variants
Guohua Wu,Guohua Wu,Xin Shen,Haifeng Li,Huangke Chen,Anping Lin,Ponnuthurai Nagaratnam Suganthan +6 more
TL;DR: The success of EDEV reveals that, through an appropriate ensemble framework, different DE variants of different merits can support one another to cooperatively solve optimization problems.
Posted Content
Temporal Graph Convolutional Network for Urban Traffic Flow Prediction Method.
TL;DR: Experiments demonstrate that the T-GCN model can obtain the spatiotemporal correlation from traffic data and the predictions outperform state-of-art baselines on real-world traffic datasets.
Journal ArticleDOI
Automatic Pavement Crack Detection by Multi-Scale Image Fusion
TL;DR: This work develops a windowed minimal intensity path-based method to extract the candidate cracks in the image at each scale, and develops a crack evaluation model based on a multivariate statistical hypothesis test that outperforms all counterparts.