Abstract
We investigate band selection for hyperspectral image classification. Mutual information (MI) measures the statistical dependence between two random variables. By modeling the reference map as one of the two random variables, MI can, therefore, be used to select the bands that are more useful for image classification. A new method is proposed to estimate the MI using an optimally constructed reference map, reducing reliance on ground-truth information. To reduce the interferences from noise and clutters, the reference map is constructed by averaging a subset of spectral bands that are chosen with the best capability to approximate the ground truth. To automatically find these bands, we develop a searching strategy consisting of differentiable MI, gradient ascending algorithm, and random-start optimization. Experiments on AVIRIS 92AV3C dataset and Pavia University scene dataset show that the proposed method outperformed the benchmark methods. In AVIRIS 92AV3C dataset, up to 55% of bands can be removed without significant loss of classification accuracy, compared to the 40% from that using the reference map accompanied with the dataset. Meanwhile, its performance is much more robust to accuracy degradation when bands are cut off beyond 60%, revealing a better agreement in the MI calculation. In Pavia University scene dataset, using 45 bands achieved 86.18% classification accuracy, which is only 1.5% lower than that using all the 103 bands.
Highlights
Hyperspectral sensors simultaneously measure hundreds of contiguous spectral bands with a fine spectral resolution, e.g., 0.01 μm
To improve the applicability of the method, we propose a new band-selection scheme based on estimating a reference map, which is constructed by Parzen window approximation and optimization algorithms
The experiment was designed to assess the change of classification accuracy as spectral bands are progressively removed according to the ranked Mutual information (MI) values
Summary
Hyperspectral sensors simultaneously measure hundreds of contiguous spectral bands with a fine spectral resolution, e.g., 0.01 μm. We study the band selection in the context of data classification, where retaining raw data appearance and not losing the original physical meaning are desirable for the purpose of registration with other source images [e.g., synthetic aperture radar (SAR) imagery] In this case, the dimensionality-reduction techniques based on feature selection is attractive. To improve the applicability of the method, we propose a new band-selection scheme based on estimating a reference map, which is constructed by Parzen window approximation and optimization algorithms. Because the high-separability bands are likely to appear in a spectrum, where the light is absorbed by the constituent atoms or molecules,[22] bands in these characteristic regions are more useful to classification and they are contiguous naturally It is, desirable to make a continuous constraint for the bands used to estimate the reference map.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
