Target Identification and Detection Using LWIR Hyperpectral Signature Transformation of Multiple Missions without Registration

Mayer, Rulon; Priest, R. G.

doi:10.21236/ada389830

Cited by 2 publications

(19 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Approximating the cross covariance matrix avoids time-consuming and computationally difficult subpixel registration between the two image sets. This study shows how the previously studied registration-free algorithms, specifically, diagonal, [1], [2] target CV [1], [2], and whitening/dewhitening [3], [4] transforms, fall within a subset of transformations having integer exponent parameter ( , respectively). The general equation for the transformation describes a continuum of transformations, differentiated by variable exponent .…”

Section: Introductionmentioning

confidence: 95%

“…Changes in atmosphere, sensor response, temperature, and solar illumination can impede target detection during extended time intervals. One approach starts with a known target spectral signature from a given time, alters the target signature to account for changes in the scene and uses that altered signature in a spectral matched filter to help identify objects at the different time [1]- [4]. Previous efforts [1]- [4] Manuscript received August 29, 2003; revised November 8, 2003.…”

Section: Introductionmentioning

confidence: 99%

“…Specifically, registration-free techniques such as the diagonal and target CV methods [1], [2] depend on approximating the cross covariance for two imaging sessions of the same scene by assuming that radiation collected from each pixel can be related through scaling, independent of pixel or position within the scene. The whitening/dewhitening transform [3], [4], however, approximates the whitened version of the cross covariance as a unit matrix.…”

Section: Introductionmentioning

confidence: 99%

“…The whitening/dewhitening transform has also been called covariance equalization [5]. By approximating the cross-covariance, targets can be successfully detected with low false-alarm rates, even for targets lying underneath camouflage netting [1]- [4]. Approximating the cross covariance matrix avoids time-consuming and computationally difficult subpixel registration between the two image sets.…”

Section: Introductionmentioning

confidence: 99%

“…Such tests examine the transforms resilience to displacement errors found in registration and airborne global positioning and inertial measurement systems. Previously [1]- [4], random pixel sampling within the overlap regions tested the sensitivity of registration-free transforms to coverage and showed that 10% or greater coverage can adequately transform spectral signatures.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Family of Spectral Target Signature Transforms: Relationship to the Past, New Transforms, and Sensitivity Tests

Mayer

Bucholtz

Scribner

et al. 2004

IEEE Geosci. Remote Sensing Lett.

View full text Add to dashboard Cite

Previous studies introduced, examined, and tested a variety of registration-free transformations, specifically the diagonal, whitening/dewhitening, and target CV (covariance) transformations, that temporally evolve spectral signatures under varying conditions. Target spectral signatures were transformed using imagery of spatially overlapping regions from datasets collected at different times. This letter shows that these previously studied registration-free transformations can be described using a single general equation and form a subset of a family of valid transforms. Further, this letter finds only a small reduction in target detection results from using nonoverlapping but similar regions to transform target signatures in matched filter searches. The diagonal, whitening/dewhitening, and target CV transformations appear in this formulation as having integral exponent parameter ( = 0 1 2, respectively). These transformed target signatures, used in matched filter searches, were tested on images taken from two very different data collects using different sensors, targets, and backgrounds. In one dataset, the transform was applied to multispectral images taken from an airborne longwave infrared sensor binned to 30 bands. The other dataset used images of a variety of targets that were collected using broad-band, bore-sighted, staring array sensors sensitive to visible, shortwave infrared, midwave infrared, and longwave infrared wavelengths. Transformation performance was tested using target-to-clutter ratio (TCR) and receiver operator characteristic (ROC) curve. Transforms with exponents between 0 and 2 yielded the largest TCR and remained relatively constant. Optimal exponents for the transformation were target dependent for both datasets. Further, only minimal reduction in TCR was observed for transforms ( 2) that had no shared overlapping regions, although the areas used to transform the spectral signatures shared similar content (trees, roads, etc.).

show abstract

Section: Introductionmentioning

confidence: 95%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Family of Spectral Target Signature Transforms: Relationship to the Past, New Transforms, and Sensitivity Tests

Mayer

Bucholtz

Scribner

et al. 2004

IEEE Geosci. Remote Sensing Lett.

View full text Add to dashboard Cite

show abstract

A metric of background candidate assessment for spectral target signature transforms

Mayer

Bucholtz²,

Allman³

et al. 2005

IEEE Geosci. Remote Sensing Lett.

View full text Add to dashboard Cite

Previous studies examined and tested a number of statistics-based registration-free transforms to find targets amid cluttered backgrounds. These transforms temporally evolve spectral target signatures under global, varying conditions using collected imagery of regions of similar objects and content distribution from datasets gathered at two different times. The transformed target signature is then inserted into the matched filter to search for targets. Although critical for transforming spectral target signatures, finding two suitable candidate regions is often difficult, computationally intensive, and may require the aid of an image analyst. This is the first study to examine a metric to help identify suitable areas for spectral target transformation. Specifically, this study examines and finds that the average correlation coefficient between the corrected histograms of the multispectral image cube collected at two times can help assess the similarity of the areas and indicate the target-to-clutter ratio, a metric shown to predict target detection performance in matched filter searches for targets.

show abstract

Target Identification and Detection Using LWIR Hyperpectral Signature Transformation of Multiple Missions without Registration

Cited by 2 publications

References 9 publications

A Family of Spectral Target Signature Transforms: Relationship to the Past, New Transforms, and Sensitivity Tests

A Family of Spectral Target Signature Transforms: Relationship to the Past, New Transforms, and Sensitivity Tests

A metric of background candidate assessment for spectral target signature transforms

Contact Info

Product

Resources

About