Multispectral rock-type separation and classification

Paskaleva, Biliana; Hayat, Majeed M.; Moya, Mary M.; Fogler, R.J.

doi:10.1117/12.558845

Cited by 5 publications

(12 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The small size of the training set is a key challenge in our ability to reliably estimate the second-order statistics of the classes. 11,12 When the dimension of data is relatively high compared to the number of samples in the training set for each group, the estimate of the covariance matrix becomes singular and cannot be used in Bayesian classification (since it cannot be inverted). This fact forces us to purposefully increase the size of training set by perturbing the endmembers with different types of minerals, vegetation, soil and water.…”

Section: Definition Of Training and Testing Setsmentioning

confidence: 99%

“…Second, to asses the reliability of QDIP in the presence of noise, we compare the classification results obtained by applying the feature reduction algorithm with the results obtained by means of more traditional sensors such as the Multispectral Thermal Imager (MTI). 11,12 The rest of the paper is organized as follows. In Section 2 we present the general analytical model for the spectral sensor and the feature representation in its space.…”

Section: Introductionmentioning

confidence: 99%

“…11 Our study has two main objectives. First, comparison of the performance of the optimal directions that consider the noise variation and the directions selected in an ideal environment shows significant improvement in the classification results when the noise statistic are embedded in the algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Feature selection for spectral sensors with overlapping noisy spectral bands

Paskaleva¹,

Hayat²,

Tyo

et al. 2006

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

Quantum-dot infrared photodetectors (QDIPs) are emerging as a promising technology for midwave-and longwave-infrared remote sensing and spectral imaging. One of the key advantages that QDIPs offer is their bias-dependent spectral response, which is brought about by the asymmetric bandstructure of the dot-in-a-well (DWELL) configuration. Photocurrents of a single QDIP, driven by different operational biases can, therefore, be viewed as outputs of different bands. It has been shown that this property, combined with post-processing strategies (applied to the outputs of a single sensor operated at different biases), can be used to perform adaptive spectral tuning and matched filtering. However, unlike traditional sensors, bands of a QDIP exhibit significant spectral overlap, an attribute that calls for the development of novel methods for feature selection. Additionally, the presence of detector noise further complicates such feature selection. In this paper, the theoretical foundations for discriminant analysis, based on spectrally adaptive feature selection, are developed and applied to data obtained from QDIP sensors in the presence of noise. The approach is based on a generalized canonicalcorrelation-analysis framework that is used in conjunction with an optimization criterion for the selection of feature subspaces. The criterion ranks the best linear combinations of the overlapping bands, providing minimal energy norm (a generalized Euclidean norm) between the centers of classes and their respective reconstructions in the space spanned by sensor bands. Experiments using ASTER-based synthetic QDIP data are used to illustrate the performance of rock-type Bayesian classification according to the proposed feature-selection method.

show abstract

Section: Definition Of Training and Testing Setsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Feature selection for spectral sensors with overlapping noisy spectral bands

Paskaleva¹,

Hayat²,

Tyo

et al. 2006

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, the resulting covariance matrices remained singular and our second step was to increase the size of training set by perturbing the endmembers with different types of minerals, vegetation, soil and water as done in our earlier work 6 . 7 We used random mixture ratios ranging from 0% − 10%. We also created mixtures between fine and coarse size rocks and between coarse and fine size, according to their geological properties.…”

Section: Populating the Training Datamentioning

confidence: 99%

Optimized algorithm for spectral band selection for rock-type classification

Paskaleva

Hayat

2005

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

Efficient use of hyperspectral (HS) sensors that can selectively activate individual, narrow spectral bands requires the development of optimized band-selection strategies that are adapted to the needs of specific detection and classification problems. By removing superfluous components of the HS data, optimized band selection significantly reduces the computational burden and improves robustness in classification. In this paper, a new method for selection of a subset of HS bands is proposed that is tailored to the problem of recognition of classes of rocks and minerals. Based on the analysis of the distribution of the amplitudes of the principal components (PCs) for a given training data set, this method identifies subsets of HS bands that provide highest spectral contrast. Three criteria are considered in the band selection process. The first criterion is based on ranking the HS bands according to the minimum distance among their respective PCs' amplitudes. The second criterion is based on ranking the HS bands according to a variant of the Kullback-Liebler divergence between a uniform distribution and the distribution of the amplitudes of the PCs for each HS band. This criterion assigns a high number on HS-bands whose PC-amplitudes' distribution either exhibits a wide range or a strong similarity to the uniform distribution. The third criterion is based on ranking the HS bands according to the empirical relative variances of the inter-amplitude distances of successive amplitudes at each HS band; it ranks high those HSbands with small inter-amplitude-distance variance and large amplitude range. These band-selection strategies are applied to laboratory HS data of rocks and minerals yielding a subset of thirteen optimal multispectral bands. The classification performance with the reduced number of bands is compared with that of the 13-band Multispectral Thermal Imager showing a moderate improvement in classification.

show abstract

“…Supervised classification approaches have frequently been used for mapping alterations and rock types in the geological studies, for instance, by Kruse et al (2002), Kruse (2002), Rowan and Mars (2003), Paskaleva et al (2004), Favretto and Geletti (2004), Hewson et al (2005), Galvao et al (2005), Vaughan et al (2005), Hubbard and Crowley (2005), as well as Wang and Zhang (2006). The reasons for selecting a particular classification method, however, are rarely discussed.…”

Section: Introductionmentioning

confidence: 98%

A selective combined classification algorithm for mapping alterations on ASTER data

Oskouei

Busch

2012

Appl Geomat

View full text Add to dashboard Cite

We present a method for combined classification aiming to map alterations in a set of ASTER (Advanced Spaceborne Thermal Emission and reflection) data in the Erongo Complex, Namibia. Ten alterations detected by the matched filtering unmixing method on the Hyperion dataset of the area are therefore used as training classes. The separability of the classes was computed to evaluate the ability of ASTER data to spectrally discriminate between these classes. The outcome of this computation is satisfactory for the highprobability training dataset. In order to improve the accuracy of upcoming processes, classes with high similarity (low separability) were combined. The classification of ASTER scene is then performed with the use of both individual and combined classification classifiers. A new combined classification method (named selective combined classification (SCC)) was developed in this research to achieve the highest possible accuracy in the resultant classification map. An accuracy analysis has proven the advantages and capability of SCC among all classifiers tested in this study (both individual and combined).

show abstract

Multispectral rock-type separation and classification

Cited by 5 publications

References 0 publications

Feature selection for spectral sensors with overlapping noisy spectral bands

Feature selection for spectral sensors with overlapping noisy spectral bands

Optimized algorithm for spectral band selection for rock-type classification

A selective combined classification algorithm for mapping alterations on ASTER data

Contact Info

Product

Resources

About