Hyperspectral Image Classification Promotion Using Clustering Inspired Active Learning
Chen Ding,Mengmeng Zheng,Feixiong Chen,Yuankun Zhang,Xusi Zhuang,Enquan Fan,Dushi Wen,Lei Zhang,Wei Wei,Yanning Zhang
DOI: https://doi.org/10.3390/rs14030596
IF: 5
2022-01-26
Remote Sensing
Abstract:Deep neural networks (DNNs) have promoted much of the recent progress in hyperspectral image (HSI) classification, which depends on extensive labeled samples and deep network structure and has achieved surprisingly good generalization capacity. However, due to the expensive labeling cost, the labeled samples are scarce in most practice cases, which causes these DNN-based methods to be prone to over-fitting and influences the classification result. To mitigate this problem, we present a clustering-inspired active learning method for enhancing the HSI classification result, which mainly contributes to two aspects. On one hand, the modified clustering by fast search and find of peaks clustering method is utilized to select highly informative and diverse samples from unlabeled samples in the candidate set for manual labeling, which empowers us to appropriately augment the limited training set (i.e., labeled samples) and thus improves the generalization capacity of the baseline DNN model. On the other hand, another K-means clustering-based pseudo-labeling scheme is utilized to pre-train the DNN model with all samples in the candidate set. By doing this, the pre-trained model can be effectively generalized to unlabeled samples in the testing set after being fine tuned-based on the augmented training set. The experiment accuracies on two benchmark HSI datasets show the effectiveness of the proposed method.
environmental sciences,imaging science & photographic technology,remote sensing,geosciences, multidisciplinary