Validating Hyperspectral Image Segmentation

Abstract

Hyperspectral satellite imaging attracts enormous research attention in the remote sensing community, hence automated approaches for precise segmentation of such imagery are being rapidly developed. In this letter, we share our observations on the strategy for validating hyperspectral image segmentation algorithms currently followed in the literature, and show that it can lead to over-optimistic experimental insights. We introduce a new routine for generating segmentation benchmarks, and use it to elaborate ready-to-use hyperspectral training-test data partitions. They can be utilized for fair validation of new and existing algorithms without any training-test data leakage.

arXiv commentSubmitted to IEEE Geoscience and Remote Sensing Letters

References in corpus (3)

• Adam: A Method for Stochastic Optimization
• Going Deeper with Contextual CNN for Hyperspectral Image Classification
• BASS Net: Band-Adaptive Spectral-Spatial Feature Learning Neural Network for Hyperspectral Image Classification

Cited by in corpus (11)

• A survey on computational spectral reconstruction methods from RGB to hyperspectral imaging
• Robust deep learning-based semantic organ segmentation in hyperspectral images
• Unsupervised Segmentation of Hyperspectral Images Using 3D Convolutional Autoencoders
• AeroRIT: A New Scene for Hyperspectral Image Analysis
• Transfer Learning for Segmenting Dimensionally-Reduced Hyperspectral Images
• Band Selection from Hyperspectral Images Using Attention-based Convolutional Neural Networks
• Triplet-Watershed for Hyperspectral Image Classification
• Hyperspectral Data Augmentation
• AMBER -- Advanced SegFormer for Multi-Band Image Segmentation: an application to Hyperspectral Imaging
• Deep Amended Gradient Descent for Efficient Spectral Reconstruction from Single RGB Images
• Segmenting Hyperspectral Images Using Spectral-Spatial Convolutional Neural Networks With Training-Time Data Augmentation