Active Learning Algorithms for the Classification of Hyperspectral Sea Ice Images

Abstract

Sea ice is one of the most critical marine disasters, especially in the polar and high latitude regions. Hyperspectral image is suitable for monitoring the sea ice, which contains continuous spectrum information and has better ability of target recognition. The principal bottleneck for the classification of hyperspectral image is a large number of labeled training samples required. However, the collection of labeled samples is time consuming and costly. In order to solve this problem, we apply the active learning (AL) algorithm to hyperspectral sea ice detection which can select the most informative samples. Moreover, we propose a novel investigated AL algorithm based on the evaluation of two criteria: uncertainty and diversity. The uncertainty criterion is based on the difference between the probabilities of the two classes having the highest estimated probabilities, while the diversity criterion is based on a kernelk-means clustering technology. In the experiments of Baffin Bay in northwest Greenland on April 12, 2014, our proposed AL algorithm achieves the highest classification accuracy of 89.327% compared with other AL algorithms and random sampling, while achieving the same classification accuracy, the proposed AL algorithm needs less labeling cost.

Sea Ice Detection Based on an Improved Similarity Measurement Method Using Hyperspectral Data

“…Finally, subsequent bands were obtained using the LP method. Then, the SVM classifier model [ 20 ] was applied to classify the sea ice using the selected bands. Data preprocessing was required during the method implementation to determine the band range with good separability for sea ice, to remove poor bands and so on.…”

“…The traditional threshold segmentation method is not efficient for addressing high-dimensional data, and it is difficult to obtain the optimal threshold [ 26 ]. Because SVM shows outstanding performance in solving small-sample, high-dimension classification problems, we selected SVM as the benchmark classifier to classify sea ice [ 20 ].…”

A Cooperative Framework Based on Active and Semi-supervised Learning for Sea Ice Classification using EO-1 Hyperion Data

“…Next, the ECBD clustering algorithm works by applying kernel k-means clustering in the kernel space where the SVM separating hyperplane operates. 10,15) Because of the kernel mapping, the set of the most diverse samples in the original space may not be the most diverse in the kernel space. The samples are mapped to a high-dimensional feature space, as it is possible to eliminate the linear inseparability in the original space by solving linear separable problems there.…”

“…Among them, h 2 , h 1 , and q 1 are the parameters used in the sampling strategy, Q 1 . According to Han et al, 10) we set q 1 to 3 and h 2 to 12. According to Experiment 5 of Patra and Bruzzone, 19) we set h 2 ¼ 2h 1 , that is, h 1 ¼ 6.…”

“…2) During recent years, the data sources commonly used include synthetic aperture radar, [3][4][5] multispectral satellite images with medium or high spatial resolution (e.g., images obtained when using a imaging spectroradiometer), [6][7][8] Landsat, 9) and hyperspectral images. 10) Of these, the hyperspectral remote sensing-images contain more spectral information, and an abundant amount of spatial information and have a higher capability for target recognition, which can greatly improve the accuracy of target detection.…”