Performance Prediction of Hyperspectral Target Detection Algorithms via Importance Sampling

Loughlin, Cooper; Truslow, Eric; Manolakis, Dimitris G.; Weisner, Andrew; Bostick, Randall L.

doi:10.1109/jstars.2019.2916260

Cited by 2 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If it is assumed the outputs approximately follow a Gaussian distribution, a threshold can be selected for an expected false alarm rate (FAR). A threshold may also be selected using more accurate performance prediction techniques such as [18], [19]. Detected pixels are passed to the ID/FAM step.…”

Section: B Cluster Detectorsmentioning

confidence: 99%

Efficient Hyperspectral Target Detection and Identification With Large Spectral Libraries

Loughlin

Pieper

Manolakis

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

Numerous hyperspectral algorithms have been developed to detect both full and sub-pixel solid target materials. Target signatures are obtained from spectral libraries that contain both target and non-target materials. When the library is large and contains many potential targets, it is inefficient to run an individual detector for each material of interest. Additionally, such an approach produces numerous false alarms (i.e., multiple detections per pixel) due to spectral similarity among targets. In this paper, we present an efficient approach for detecting multiple targets within large spectral libraries while mitigating false alarms. We first group spectrally similar materials within the library into a hierarchy of clusters. From each cluster containing a target material, a single detector is obtained. Each detector represents multiple library spectra, so an identification step is needed for detected pixels. Detected pixels are modeled as a mixture between their local in-scene background and candidate library spectra. Candidates are chosen from adjacent library clusters. The candidate model providing the best fit is chosen to report. Use of local background spectra provides a physically meaningful mixing model that adapts to detected pixels. Clustering the library reduces the computational complexity of modeling detected pixels. We demonstrate detection and false alarm mitigation performance of our proposed algorithm with a real hyperspectral dataset.

show abstract

Section: B Cluster Detectorsmentioning

confidence: 99%

Efficient Hyperspectral Target Detection and Identification With Large Spectral Libraries

Loughlin

Pieper

Manolakis

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

Forest Ecological Diversity Change Prediction Discrete Dynamic Model

Zhou

Liao

et al. 2022

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

Forest ecosystem is the most important terrestrial ecosystem type of species diversity in the world. The protection and sustainable use of forest ecological diversity is of great significance to forest protection and sustainable management. In order to maintain biodiversity, the traditional concept of forest resources management must be changed. In recent years, the mining and application of big data have become the frontier content of international ecological diversity and macroecology research. By establishing the discrete dynamic model of forest ecological diversity change, this study finds a method that can keep the number of trees stable in a short time and ensure the sustainable development of forest resources. This article studies and constructs the forest information system and constructs an open-source WebGIS (network geographic information system) scheme for the field of spatial information and supporting OGC (occupationally generated content). At the same time, the grey theory GM (1,1) model is used to predict the development trend of forest resources in China. The results show that under the promotion of policies and research projects, China’s forestry development has realized the double growth of forest area and volume, and the forest coverage rate has reached a new level.

show abstract

Performance Prediction of Hyperspectral Target Detection Algorithms via Importance Sampling

Cited by 2 publications

References 17 publications

Efficient Hyperspectral Target Detection and Identification With Large Spectral Libraries

Efficient Hyperspectral Target Detection and Identification With Large Spectral Libraries

Forest Ecological Diversity Change Prediction Discrete Dynamic Model

Contact Info

Product

Resources

About