Data Leakage in Patch-Based Hyperspectral Image Classification: Quantifying the Impact of Spatial Overlap

Key points

• This paper investigates same-class train-test spatial overlap in patch-based HSI classification using two measures: overlap percentage (OP), which quantifies the global amount of overlapped testing patch pixels, and average overlap ratio (AOR), which measures the local severity among affected testing patches.1
• Experiments on the Pavia University dataset compare random and non-random spatial sampling using SVM, MLP, 2D-CNN, 3D-CNN, ViT, and MorpMamba.1
• The results show that deep patch-based models achieve high accuracy under random sampling, with 3D-CNN reaching 96.17% Overall Accuracy (OA), but drop substantially under non-random spatial sampling, where 3D-CNN decreases to 55.20% and ViT and 2D-CNN drop by 40.71 and 38.81 percentage points (PP), respectively.1
• These findings demonstrate that random patch-based evaluation can substantially inflate classification performance, especially for models that strongly exploit spatial context.1

Sources (1)

• [1]Data Leakage in Patch-Based Hyperspectral Image Classification: Quantifying the Impact of Spatial Overlap
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 05:55 PM

  Patch-based learning improves hyperspectral image (HSI) classification by exploiting local spectral-spatial information, but random train-test sampling from the same image can cause spatial patch overlap, leading to data leakage and optimistic performance estimates.

  This paper investigates same-class train-test spatial overlap in patch-based HSI classification using two measures: overlap percentage (OP), which quantifies the global amount of overlapped testing patch pixels, and average overlap ratio (AOR), which measures the local severity among affected testing patches.

Extractive summary: sentences quoted from the sources.