Most existing depth networks that perform hyperspectral anomaly detection (HAD) using reconstruction errors tend to fit anomalous pixels, resulting in small reconstruction errors for anomalies, which are not favorable for separating targets from hyperspectral images (HSIs). To achieve a superior background reconstruction network for HAD purposes, a self-supervised blind-block reconstruction network (termed BockNet) with a guard window is proposed. BockNet creates a blind-block (guard window) at the center of the network's receptive field, making it unable to see the information inside the guard window when reconstructing the central pixel. This process seamlessly embeds a sliding dual-window model into our BockNet, where the inner window is the guard window, and the outer window is the receptive field outside the guard window. Naturally, BockNet uses only the outer window information to predict/reconstruct the central pixel of the perceptive field. During the reconstruction of pixels inside anomalous targets of different sizes, the targets typically fall into the guard window, weakening the contribution of anomalies to the reconstruction results and allowing these reconstructed pixels to converge to the background distribution of the outer window region. Accordingly, the reconstructed HSI can be regarded as a pure background HSI, and the reconstruction error of anomalous pixels will be further enlarged, thus improving the discrimination ability of the proposed network for anomalies. Extensive experiments on two datasets show the competitive performance of our BockNet compared to state-of-the-art detectors.
The very high spectral resolution of Hyperspectral Images (HSIs) enables the identification of materials with subtle differences and the extraction subpixel information. However, the increasing of spectral resolution often implies an increasing in the noise linked with the image formation process. This degradation mechanism limits the quality of extracted information and its potential applications. Since HSIs represent natural scenes and their spectral channels are highly correlated, they are characterized by a high level of self-similarity and are well approximated by low-rank representations. These characteristic underlies the state-of-the-art in HSI denoising. However, in presence of rare pixels, the denoising performance of those methods is not optimal and, in addition, it may compromise the future detection of those pixels. To address these hurdles, we introduce RhyDe (Robust hyperspectral Denoising), a powerful HSI denoiser, which implements explicit low-rank representation, promotes self-similarity, and, by using a form of collaborative sparsity, preserves rare pixels. The denoising and detection effectiveness of the proposed robust HSI denoiser is illustrated using semi-real data.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.