Rethinking Data Leakage in Patch-Based Hyperspectral Image Classification with Traditional Deep Networks
Remote Sensing · 25.03.2026 · 18 (7) · 979
DOI: 10.3390/rs18070979 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 25.03.2026. Verfasst von Kaizhe Zhan und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Outstanding Youth Project of Jiangxi Natural Science Foundation, National Natural Science Foundation of China https://ror.org/01h0zpd94, Open Project of Henan Provincial Key Laboratory of Oracle Bone Information Processing.
Abstract (Originalsprache)
The application of hyperspectral image (HSI) processing techniques has become increasingly important in many fields such as agriculture, environmental detection, and mining. However, the number of annotated samples in existing hyperspectral datasets is limited, and most hyperspectral classification models typically use patch data for model training. There, the pixels to be classified are often taken as the center, and then a mean pixel length is calculated from the surrounding pixel neighborhood to form patch data. However, during model training, the researchers found that there was an area of overlapping pixels between most of the training data and the test data. This inevitably led to data leakage, resulting in excellent classification performance of the model. To solve this problem, we develop a method of replacing overlapping pixels (ROP) in patch data, which means the training pixel points of the same class are used to replace the test pixel points that appear in the overlapping region of the training patch data. Furthermore, a multiple feature extraction and fusion (MFEF) module is also proposed to enhance the capacity of the HSI model to extract spectral–spatial feature information from new patch data. The results on five publicly available HSI datasets demonstrate that the proposed resolving data leakage network (RDLNet) can provide competitive classification results on the patch data reconstructed with the ROP strategy, which outperforms existing state-of-the-art (SOTA) classification methods as well.
Förderung
- Outstanding Youth Project of Jiangxi Natural Science Foundation
- National Natural Science Foundation of China https://ror.org/01h0zpd94
- Open Project of Henan Provincial Key Laboratory of Oracle Bone Information Processing
- Jiangxi Science and Technology Normal University https://ror.org/04r1zkp10
- Department of Education of Jiangxi Province
Zitieren
APA
Zhan, K., Huo, Z., Yin, Z., & Yang, Z. (2026). Rethinking Data Leakage in Patch-Based Hyperspectral Image Classification with Traditional Deep Networks. Remote Sensing, 18(7), 979. https://doi.org/10.3390/rs18070979
Vancouver
Zhan K, Huo Z, Yin Z, Yang Z. Rethinking Data Leakage in Patch-Based Hyperspectral Image Classification with Traditional Deep Networks. Remote Sensing. 2026;18(7):979. doi:10.3390/rs18070979
BibTeX
@article{Zhan2026,
author = {Zhan, Kaizhe and Huo, Zijie and Yin, Zhijian and Yang, Zhen},
title = {Rethinking Data Leakage in Patch-Based Hyperspectral Image Classification with Traditional Deep Networks},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {7},
pages = {979},
doi = {10.3390/rs18070979}
}
EUVIMED