Meta Transfer Learning for Few-Shot Hyperspectral Image Classification

We propose a novel meta-learning approach for few-shot hyperspectral image (HSI) classification, which learns to distil transferable prior knowledge from a base dataset with sufficient labeled samples and generalize the knowledge to an unseen dataset with extremely limited labeled samples for performance improvement. Specifically, we first construct a backbone classification model using an embedding module and a linear classifier. Then, we sample extensive synthetic few-shot tasks from the base dataset, each of which consists of a support set with limited labeled samples and a query set with some unlabeled test samples. Given these tasks, we propose to optimize the embedding module using an episode learning scheme where for each task we train the linear classier based on an initialized embedding module using the support set and ultimately optimize the embedding module based on the test error on the query set until the test error on all tasks is minimized. By doing this, the resultant embedding module is able to appropriately generalize to an unseen few-shot classification task and lead to good performance with the linear classifier. Experiments on two standard classification benchmarks under different few-shot settings demonstrate the efficacy of the proposed method.

1. Introduction

Hyperspectral image (HSI) is an 3-dimensional image with abundant spectral and spatial information, which has been widely utilzied in a wide range of applications, including precise agriculture, land-use planning, environmental monitoring[1] etc. In those applications, one of the fundamental tasks is HSI classification which aims at assign a specific class label to each pixels of the image based on some labeled training samples. Early HSI classification methods mainly include support vector machine (SVM), k-mean clustering [2], random forest[3] etc. Due to the limited expressive capacity resulted by their shallow structure, these methods fail to produce effective feature representation for spectral pixels in an HSI and their classification performance can thus be further improved.