Education, Science, Technology, Innovation and Life
Open Access
Sign In

Multi-Scale Graph Wavelet Convolution for Hyperspectral-LiDAR Urban Scene Classification

Download as PDF

DOI: 10.23977/acss.2025.090311 | Downloads: 10 | Views: 214

Author(s)

Junhua Ku 1,2, Jie Zhao 3

Affiliation(s)

1 School of Information Science and Technology, Qiongtai Normal University, Haikou, Hainan, 571127, China
2 Institute of Educational Big Data and Artificial Intelligence, Qiongtai Normal University, Haikou, Hainan, 571127, China
3 School of Science, Qiongtai Normal University, Haikou, Hainan, 571127, China

Corresponding Author

Junhua Ku

ABSTRACT

The Houston 2013 benchmark conducted in Houston describes a methodology that constructs a sparse pixel graph exclusively over labeled pixels. This approach employs learnable multi-scale spectral filtering using Chebyshev-approximated graph wavelets and incorporates a lightweight Multi-Layer Perceptron (MLP) head for classification purposes. The training procedure integrates feature MixUp at labeled nodes, a composite loss function combining focal loss and label smoothing, exponential moving average (EMA) weight tracking, an AdamW optimizer with warm-up and cosine scheduling, and mild graph augmentation techniques such as random edge drop and addition with re-normalization. Implementation executed precisely in accordance with the original methodology, with five independent runs producing an overall accuracy (OA) of 90.99% ± 0.41%, average accuracy (AA) of 92.11% ± 0.35%, and a Kappa coefficient (κ) of 0.9022 ± 0.0044. These results demonstrate not only high accuracy but also minimal variability, providing reassurance of the robustness of the methodology.

KEYWORDS

Hyperspectral imaging; LiDAR; graph neural networks; graph wavelets; multi-sensor fusion; urban land cover; Chebyshev approximation; focal loss; label smoothing

CITE THIS PAPER

Junhua Ku, Jie Zhao, Multi-Scale Graph Wavelet Convolution for Hyperspectral-LiDAR Urban Scene Classification. Advances in Computer, Signals and Systems (2025) Vol. 9: 88-97. DOI: http://dx.doi.org/10.23977/acss.2025.090311.

REFERENCES

[1] J. A. Palmason, J. A. Benediktsson, J. R. Sveinsson and J. Chanussot. Classification of hyperspectral data from urban areas using morphological preprocessing and independent component analysis, 2005 IEEE International Geoscience and Remote Sensing Symposium, 2005. pp. 4, doi: 10.1109/IGARSS.2005.1526133.
[2] M. Pedergnana, P. R. Marpu, M. Dalla Mura, J. A. Benediktsson, and L. Bruzzone, "Classification of remote sensing optical and LiDAR data using extended attribute profiles," IEEE J. Sel. Top. Signal Process., vol. 6, no. 7, pp. 856–865, Nov. 2012, doi: 10.1109/JSTSP.2012.2208177.
[3] Huang, W.; Zhao, Z.; Sun, L.; Ju, M. Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification. Remote Sens. 2022, 14, 6158. https://doi.org/10.3390/rs14236158
[4] Ding, Y.; Guo, Y.; Chong, Y.; Pan, S.; Feng, J. Global Consistent Graph Convolutional Network for Hyperspectral Image Classification. IEEE Trans. Instrum. Meas. 2021, 70, 5501516. https://doi.org/10.1109/TIM.2021.3056750 .
[5] Zhang, X.; Chen, S.; Zhu, P.; Tang, X.; Feng, J.; Jiao, L. Spatial Pooling Graph Convolutional Network for Hyperspectral Image Classification. IEEE Trans. Geosci. Remote Sens. 2022, 60, 5521315. https://doi.org/10.1109/TGRS.2022.3147891.
[6] Chen, Y.; Lu, X.; Zhang, Z.; Xue, Z.; Zhang, L.; Zhang, Z.; Wang, L. Joint Classification of Hyperspectral and LiDAR Data Using a Hierarchical CNN and Transformer. IEEE Trans. Geosci. Remote Sens. 2023, 61(1), 123–135. https://doi.org/10.1109/TGRS.2023.3314616.
[7] B. Rasti, B. Koirala, P. Scheunders and J. Chanussot, "MiSiCNet: Minimum Simplex Convolutional Network for Deep Hyperspectral Unmixing," in IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-15, 2022, Art no. 5522815, doi: 10.1109/TGRS.2022.3146904. 
[8] Liao W , Coillie F V , Gao L ,et al. Deep Learning for Fusion of APEX Hyperspectral and Full-waveform LiDAR Remote Sensing Data for Tree Species Mapping[J].IEEE Access, 2018, vol. 6, PP. 68716-68729. DOI:10.1109/ACCESS.2018.2880083.

Downloads: 38831
Visits: 727282

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.