Technical note The application of selective principal components analysis (SPCA) to a Thematic Mapper (TM) image for the recognition of geomorphologic features configuration

Siljeström, Patricia; Moreno, Adela; Vikgren, K.; Caceres, L. M.

doi:10.1080/014311697216658

Cited by 34 publications

(14 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PCA has traditionally been used in remote sensing as a means of data compaction since it is common to find that the first 2 or 3 components are able to explain the majority of the variability in data values, while later components tend to be dominated by noise effects. The rejection of these later components reduces the volume of data with no appreciable loss of information (Siljestrom et al, 1997). Standardized principal components analysis (Eastman and Fulk, 1993) is applied (data per month is centered with mean 0.0 and standard deviation 1.0) and so each image is not weighed according to its variance.…”

Section: Methodsmentioning

confidence: 99%

Spatial and seasonal patterns of precipitation in Greece: the terrain segmentation approach

2014

Global NEST: The International Journal

View full text Add to dashboard Cite

The aim of the study is to investigate the spatial and seasonal variation of precipitation in Greece using multi-temporal data analysis techniques. Mean monthly precipitation grids at 1 km 2 resolution for the period 1950-2000 were used in the analysis. Cross correlation quantified spatio-temporal patterns which are summarized as follows a) the absolute correlations of precipitation versus elevation and longitude are minimized during the winter period, b) the latitude dependency of precipitation presents a seasonal shift where winter precipitation tends to be higher in northern Greece, while summer precipitation tends to be higher in southern Greece. Principal components analysis indicated that the first two components account for the 92.8% of variance in the spatio-temporal variability of precipitation. Cluster analysis segmented the terrain to 27 regions with distinct seasonal variability of precipitation. The majority of irrigated agricultural land (plains of Macedonia, Thessaly and Thrace) belong to clusters which present the lowest values of annual precipitation (<600 mm year -1 ). The derivation of precipitation signatures for each region of Greece using the proposed terrain segmentation approach can support environmental decision and agricultural planning at a regional (country) scale in relation to water resources management.

show abstract

Section: Methodsmentioning

confidence: 99%

Spatial and seasonal patterns of precipitation in Greece: the terrain segmentation approach

2014

Global NEST: The International Journal

View full text Add to dashboard Cite

show abstract

“…There are several approaches for selection of features from multispectral datasets that could provide the optimum information about the various categories present within a scene. One of them is the selective principal components analysis (Siljestrom et al 2002) that deals with the selection of the most informative principal components based on their eigenvalues. Many a times, classification of the selective principal components yields better classification accuracy of some of the categories as compared to the clas-ORIGINAL PAPER Journal of Forestry Research (2011) 22(1): 99−105 100 sification of all the principal components.…”

Section: Introductionmentioning

confidence: 99%

Classification of features selected through Optimum Index Factor (OIF) for improving classification accuracy

Patel

Kaushal

2011

Journal of Forestry Research

View full text Add to dashboard Cite

The present investigation was performed to determine if the features selected through Optimum Index Factor (OIF) could provide improved classification accuracy of the various categories on the satellite images of the individual years as well as stacked images of two different years as compared to all the features considered together. Further, in order to determine if there occurs increase in the classification accuracy of the different categories with corresponding increase in the OIF values of the features extracted from both the individual years' and stacked images, we performed linear regression between the producer's accuracy (PA) of the various categories with the OIF values of the different combinations of the features. The investigations demonstrated that there occurs significant improvement in the PA of two impervious categories viz. moderate built-up and low density built-up determined from the classification of the bands and principal components associated with the highest OIF value as compared to all the bands and principal components for both the individual years' and stacked images respectively. Regression analyses exhibited positive trends between the regression coefficients and OIF values for the various categories determined for the individual years' and stacked images respectively signifying the prevalence of direct relationship between the increase in the information content with corresponding increase in the OIF values. The research proved that features extracted through OIF from both the individual years' and stacked images are capable of providing significantly improved PA as compared to all the features pooled together.

show abstract

“…For instance, spectral contrast mapping-also known as selective PCA-uses two bands that are only weakly correlated to each other as an input and produces two components from which the second summarizes the changes [23,25,29]. Harsanyi and Chang [30] developed a method called noise-adjusted PCA in which they suppress undesired spectral signatures in order to maximize the signalto-noise ratio of a particular signature of interest.…”

Section: Change Detection Via Pcamentioning

confidence: 99%

“…Multi-spectral bands treated as columns build up a data matrix, which is then centered and whose variance-covariance (or correlation) matrix is the basis for the components extraction algorithm. In addition, past literature on the subject, includes various explorations such as using unitemporal [17,21,22] and multi-temporal [18,21,[23][24][25][26][27][28] multi-spectral data sets. These studies clearly depict multi-temporal approaches as more effective in terms of both qualitative and quantitative accuracy.…”

Section: Change Detection Via Pcamentioning

confidence: 99%

Remote sensing of burned areas via PCA, Part 1; centering, scaling and EVD vs SVD

Alexandris¹,

Gupta

Koutsias

2017

Open geospatial data, softw. stand.

View full text Add to dashboard Cite

Background: Principal components analysis (PCA) is based conventially on the eigenvector decomposition (EVD). Mean-centering the input data prior to the eigenanalysis is treated as an integral part of the algorithm. It ensures that the first principal component is proportional to the maximum variance of the input data. Equivalent to EVD, but numerically more robust, is the singular value decomposition (SVD). Mean-centered data subjected to SVD, yield transformation coefficients identical to EVD. Nevertheless, mean-centering is optional in SVD. Avoiding to center the input data, results in generic first component that mainly reflects their mean. This may, however, detect more accurately distinct clusters in PCA-based change detection applications. Methods: In remote sensing, PCA transforms multi-spectral bands into a new coordinate system. The first, among the transformed components, contain the variance of unchanged landscape features. Succeeding components may contain an enhanced variance of changed features. Such is the case of burned surfaces appearing as distinct clusters in multitemporal composites. Conclusions: Within this framework, a non-centered SVD may increase the spectral separability of burned clusters among other features in some of the higher order principal components.

show abstract

Technical note The application of selective principal components analysis (SPCA) to a Thematic Mapper (TM) image for the recognition of geomorphologic features configuration

Cited by 34 publications

References 1 publication

Spatial and seasonal patterns of precipitation in Greece: the terrain segmentation approach

Spatial and seasonal patterns of precipitation in Greece: the terrain segmentation approach

Classification of features selected through Optimum Index Factor (OIF) for improving classification accuracy

Remote sensing of burned areas via PCA, Part 1; centering, scaling and EVD vs SVD

Contact Info

Product

Resources

About