%20methods%20use%20the%20low-rank%20representation%20(LRR)%20model%20to%20separate%20the%20background%20and%20anomaly%20components%2C%20where%20the%20anomaly%20component%20is%20optimized%20by%20handcrafted%20sparse%20priors%20(e.g.%2C%20%24ell_%7B2%2C1%7D%24-norm).%20However%2C%20this%20may%20not%20be%20ideal%20since%20they%20overlook%20the%20spatial%20structure%20present%20in%20anomalies%20and%20make%20the%20detection%20result%20largely%20dependent%20on%20manually%20set%20sparsity.%20To%20tackle%20these%20problems%2C%20we%20redefine%20the%20optimization%20criterion%20for%20the%20anomaly%20component%20in%20the%20LRR%20model%20with%20a%20self-supervised%20network%20called%20self-supervised%20anomaly%20prior%20(SAP).%20This%20prior%20is%20obtained%20by%20the%20pretext%20task%20of%20self-supervised%20learning%2C%20which%20is%20customized%20to%20learn%20the%20characteristics%20of%20hyperspectral%20anomalies.%20Specifically%2C%20this%20pretext%20task%20is%20a%20classification%20task%20to%20distinguish%20the%20original%20hyperspectral%20image%20(HSI)%20and%20the%20pseudo-anomaly%20HSI%2C%20where%20the%20pseudo-anomaly%20is%20generated%20from%20the%20original%20HSI%20and%20designed%20as%20a%20prism%20with%20arbitrary%20polygon%20bases%20and%20arbitrary%20spectral%20bands.%20In%20addition%2C%20a%20dual-purified%20strategy%20is%20proposed%20to%20provide%20a%20more%20refined%20background%20representation%20with%20an%20enriched%20background%20dictionary%2C%20facilitating%20the%20separation%20of%20anomalies%20from%20complex%20backgrounds.%20Extensive%20experiments%20on%20various%20hyperspectral%20datasets%20demonstrate%20that%20the%20proposed%20SAP%20offers%20a%20more%20accurate%20and%20interpretable%20solution%20than%20other%20advanced%20HAD%20methods.)
![[논문 리뷰] Spectral-Spatial Extraction through Layered Tensor ...](https://moonlight-paper-snapshot.s3.ap-northeast-2.amazonaws.com/arxiv/spectral-spatial-extraction-through-layered-tensor-decomposition-for-hyperspectral-anomaly-detection-4.png)
Elevate your brand with our commercial figure 1 from spectral-spatial anomaly detection in hyperspectral gallery featuring extensive collections of business-ready images. designed for business applications featuring play, doll, and game. ideal for corporate communications and branding. Our figure 1 from spectral-spatial anomaly detection in hyperspectral collection features high-quality images with excellent detail and clarity. Suitable for various applications including web design, social media, personal projects, and digital content creation All figure 1 from spectral-spatial anomaly detection in hyperspectral images are available in high resolution with professional-grade quality, optimized for both digital and print applications, and include comprehensive metadata for easy organization and usage. Explore the versatility of our figure 1 from spectral-spatial anomaly detection in hyperspectral collection for various creative and professional projects. Time-saving browsing features help users locate ideal figure 1 from spectral-spatial anomaly detection in hyperspectral images quickly. Regular updates keep the figure 1 from spectral-spatial anomaly detection in hyperspectral collection current with contemporary trends and styles. Professional licensing options accommodate both commercial and educational usage requirements. Comprehensive tagging systems facilitate quick discovery of relevant figure 1 from spectral-spatial anomaly detection in hyperspectral content. The figure 1 from spectral-spatial anomaly detection in hyperspectral archive serves professionals, educators, and creatives across diverse industries. Diverse style options within the figure 1 from spectral-spatial anomaly detection in hyperspectral collection suit various aesthetic preferences.